Optimize ByteViewGroupValueBuilder vectorized_append

[Dandandan]

Which issue does this PR close?

N/A - Performance optimization

Rationale for this change

The vectorized_append_inner method in ByteViewGroupValueBuilder was using make_view to reconstruct views from scratch and copying non-inline data byte-by-byte in a per-row loop. This is unnecessary since the input array already has correctly formed views with length and prefix.

What changes are included in this PR?

• Two-phase vectorized append:
  • Phase 1: Fast extend over views, temporarily referencing input array buffers (no per-row memcpy in the hot loop)
  • Phase 2: Smart compaction — count referenced bytes per reused buffer; if >=90% is referenced, keep the buffer as-is (just remap view indices); otherwise compact by copying only referenced data into owned buffers
• View reuse: Copy input views directly instead of reconstructing via make_view; use ByteView builder API for buffer_index/offset updates
• All-inline fast path: Skip buffer work entirely when no non-inline data exists
• Extract shared copy_from_view helper for single-row append_val_inner
• Remove now-unused do_append_val_inner, ensure_in_progress_big_enough, and make_view import

Benchmark results

inline_null_0.0_size_1000/vectorized_append (8B strings, all inline)
                        time:   [458.68 ns 468.83 ns 480.99 ns]
                        change: [−90.156% −89.865% −89.564%] (p = 0.00 < 0.05)
                        Performance has improved.

scenario_null_0.0_size_1000/vectorized_append (64B strings, non-inline)
                        time:   [2.8031 µs 2.8582 µs 2.9244 µs]
                        change: [−74.113% −73.410% −72.688%] (p = 0.00 < 0.05)
                        Performance has improved.

random_null_0.0_size_1000/vectorized_append (up to 400B strings, mixed)
                        time:   [4.2838 µs 4.4440 µs 4.6305 µs]
                        change: [−79.826% −79.116% −78.176%] (p = 0.00 < 0.05)
                        Performance has improved.

Case	Before	After	Improvement
inline (8B)	4.29 µs	0.47 µs	-89%
scenario (64B)	11.43 µs	2.86 µs	-75%
random (≤400B)	20.61 µs	4.44 µs	-78%

Are these changes tested?

Existing tests cover the functionality. Verified with cargo check and cargo clippy.

Are there any user-facing changes?

No - internal performance optimization only.

🤖 Generated with Claude Code

[Dandandan]

run benchmark aggregate_vectorized

[adriangbot]

🤖 Criterion benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231206259-1108-m6xwr 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (8b65fee) to 16e578d (merge-base) diff
BENCH_NAME=aggregate_vectorized
BENCH_COMMAND=cargo bench --features=parquet --bench aggregate_vectorized
BENCH_FILTER=
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

Benchmark for this request failed.

Last 20 lines of output:

Click to expand

  Downloaded enum-ordinalize-derive v4.3.2
  Downloaded dyn-clone v1.0.20
  Downloaded ctor v0.8.0
  Downloaded const_panic v0.2.15
  Downloaded clap_lex v1.1.0
  Downloaded cfg-if v1.0.4
  Downloaded block-buffer v0.12.0
  Downloaded aws-smithy-query v0.60.15
  Downloaded anstyle-query v1.1.5
  Downloaded ciborium-io v0.2.2
  Downloaded atoi v2.0.0
  Downloaded cfg_aliases v0.2.1
  Downloaded base64-simd v0.8.0
  Downloaded cast v0.3.0
  Downloaded block-buffer v0.10.4
  Downloaded arrayref v0.3.9
  Downloaded async-recursion v1.1.1
    Blocking waiting for file lock on package cache
error: target `aggregate_vectorized` in package `datafusion-physical-plan` requires the features: `test_utils`
Consider enabling them by passing, e.g., `--features="test_utils"`

---

File an issue against this benchmark runner

[Dandandan]

run benchmarks

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231211405-1111-jwzj6 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (8b65fee) to 16e578d (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231211405-1110-lcjs9 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (8b65fee) to 16e578d (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231211405-1109-h7rxp 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (8b65fee) to 16e578d (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃      optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │              6.56 / 7.01 ±0.80 / 8.62 ms │              6.60 / 7.03 ±0.80 / 8.64 ms │     no change │
│ QQuery 2  │        145.00 / 145.65 ±0.55 / 146.45 ms │        144.97 / 146.23 ±0.67 / 146.95 ms │     no change │
│ QQuery 3  │        112.34 / 113.11 ±0.47 / 113.66 ms │        113.98 / 114.39 ±0.40 / 115.10 ms │     no change │
│ QQuery 4  │    1272.70 / 1295.46 ±20.12 / 1323.51 ms │    1260.65 / 1295.47 ±22.97 / 1322.60 ms │     no change │
│ QQuery 5  │        171.55 / 172.67 ±0.88 / 173.50 ms │        172.62 / 174.17 ±1.28 / 176.29 ms │     no change │
│ QQuery 6  │       817.31 / 838.82 ±15.76 / 864.58 ms │       845.05 / 857.92 ±12.94 / 881.42 ms │     no change │
│ QQuery 7  │        337.59 / 340.67 ±1.94 / 343.13 ms │        338.05 / 339.68 ±2.10 / 343.77 ms │     no change │
│ QQuery 8  │        115.92 / 116.65 ±0.66 / 117.74 ms │        113.83 / 115.30 ±1.31 / 116.94 ms │     no change │
│ QQuery 9  │        100.94 / 104.18 ±2.48 / 106.74 ms │        100.20 / 105.34 ±2.62 / 107.40 ms │     no change │
│ QQuery 10 │        105.39 / 106.21 ±0.54 / 107.03 ms │        104.54 / 105.67 ±0.94 / 107.31 ms │     no change │
│ QQuery 11 │        864.19 / 879.74 ±9.08 / 892.34 ms │       864.58 / 882.24 ±11.28 / 898.20 ms │     no change │
│ QQuery 12 │           43.90 / 45.43 ±0.80 / 46.15 ms │           43.87 / 44.46 ±0.49 / 45.20 ms │     no change │
│ QQuery 13 │        397.09 / 400.51 ±2.38 / 403.63 ms │        392.36 / 398.31 ±3.21 / 401.56 ms │     no change │
│ QQuery 14 │       989.17 / 997.83 ±6.16 / 1007.43 ms │      998.22 / 1006.39 ±4.68 / 1012.48 ms │     no change │
│ QQuery 15 │           15.16 / 16.64 ±1.30 / 18.31 ms │           15.14 / 16.06 ±0.95 / 17.86 ms │     no change │
│ QQuery 16 │              7.05 / 7.61 ±0.88 / 9.36 ms │             7.39 / 8.72 ±1.42 / 11.34 ms │  1.15x slower │
│ QQuery 17 │        226.09 / 227.42 ±0.86 / 228.68 ms │        226.81 / 228.68 ±1.42 / 230.59 ms │     no change │
│ QQuery 18 │        125.04 / 126.42 ±0.79 / 127.22 ms │        124.98 / 127.36 ±1.56 / 129.52 ms │     no change │
│ QQuery 19 │        153.05 / 155.63 ±2.65 / 160.69 ms │        153.37 / 154.05 ±0.44 / 154.53 ms │     no change │
│ QQuery 20 │           13.53 / 13.92 ±0.29 / 14.32 ms │           13.25 / 13.83 ±0.43 / 14.23 ms │     no change │
│ QQuery 21 │           19.00 / 19.52 ±0.45 / 20.31 ms │           18.98 / 19.66 ±0.53 / 20.18 ms │     no change │
│ QQuery 22 │        479.28 / 482.70 ±2.02 / 485.21 ms │        476.59 / 481.51 ±2.99 / 485.97 ms │     no change │
│ QQuery 23 │        842.33 / 857.93 ±8.23 / 864.67 ms │        845.98 / 857.24 ±7.87 / 870.15 ms │     no change │
│ QQuery 24 │        375.58 / 379.45 ±4.07 / 384.85 ms │        377.51 / 379.59 ±2.09 / 382.99 ms │     no change │
│ QQuery 25 │        337.00 / 339.92 ±1.68 / 341.74 ms │        337.05 / 338.54 ±1.02 / 339.74 ms │     no change │
│ QQuery 26 │           79.00 / 80.90 ±1.70 / 83.29 ms │           79.72 / 81.15 ±1.29 / 83.01 ms │     no change │
│ QQuery 27 │              6.75 / 7.24 ±0.57 / 8.34 ms │              6.71 / 7.59 ±1.02 / 9.23 ms │     no change │
│ QQuery 28 │        146.75 / 148.34 ±1.20 / 149.72 ms │        147.56 / 149.03 ±1.85 / 152.65 ms │     no change │
│ QQuery 29 │        279.36 / 280.41 ±0.97 / 281.77 ms │        279.77 / 281.06 ±2.17 / 285.38 ms │     no change │
│ QQuery 30 │           41.91 / 42.88 ±0.88 / 44.25 ms │           42.92 / 43.62 ±0.48 / 44.13 ms │     no change │
│ QQuery 31 │        167.42 / 169.09 ±1.38 / 170.96 ms │        168.03 / 171.13 ±2.12 / 173.94 ms │     no change │
│ QQuery 32 │           56.18 / 56.47 ±0.25 / 56.84 ms │           55.86 / 57.60 ±1.26 / 59.80 ms │     no change │
│ QQuery 33 │        138.32 / 140.31 ±1.25 / 141.67 ms │        139.01 / 140.66 ±1.41 / 142.85 ms │     no change │
│ QQuery 34 │              6.86 / 7.09 ±0.33 / 7.74 ms │              6.79 / 7.39 ±0.68 / 8.66 ms │     no change │
│ QQuery 35 │        104.80 / 106.62 ±1.15 / 107.86 ms │        103.52 / 106.05 ±1.73 / 107.82 ms │     no change │
│ QQuery 36 │              6.50 / 7.02 ±0.59 / 8.18 ms │              6.37 / 7.02 ±0.82 / 8.61 ms │     no change │
│ QQuery 37 │              8.06 / 8.40 ±0.25 / 8.70 ms │              8.25 / 8.76 ±0.53 / 9.77 ms │     no change │
│ QQuery 38 │           83.98 / 85.43 ±1.94 / 89.15 ms │           85.95 / 88.01 ±1.37 / 89.79 ms │     no change │
│ QQuery 39 │        123.72 / 125.73 ±1.52 / 127.95 ms │        124.16 / 126.72 ±1.39 / 128.08 ms │     no change │
│ QQuery 40 │        108.16 / 113.00 ±6.04 / 124.26 ms │        109.62 / 114.68 ±4.94 / 124.09 ms │     no change │
│ QQuery 41 │           14.18 / 15.72 ±2.14 / 19.93 ms │           14.31 / 15.17 ±0.83 / 16.71 ms │     no change │
│ QQuery 42 │        108.08 / 108.75 ±0.75 / 110.13 ms │        106.21 / 108.61 ±1.31 / 110.06 ms │     no change │
│ QQuery 43 │              6.03 / 6.13 ±0.12 / 6.37 ms │              5.97 / 6.20 ±0.37 / 6.92 ms │     no change │
│ QQuery 44 │           11.65 / 12.04 ±0.31 / 12.39 ms │           11.34 / 11.76 ±0.30 / 12.21 ms │     no change │
│ QQuery 45 │           49.34 / 50.29 ±1.13 / 52.41 ms │           49.60 / 50.64 ±0.65 / 51.36 ms │     no change │
│ QQuery 46 │              8.36 / 8.75 ±0.24 / 9.11 ms │              8.42 / 8.83 ±0.38 / 9.51 ms │     no change │
│ QQuery 47 │        674.84 / 685.01 ±7.57 / 697.94 ms │        684.46 / 689.04 ±3.83 / 693.91 ms │     no change │
│ QQuery 48 │        287.19 / 290.84 ±3.99 / 298.32 ms │        280.78 / 288.25 ±5.03 / 296.58 ms │     no change │
│ QQuery 49 │        249.78 / 251.70 ±1.68 / 253.88 ms │        250.98 / 254.04 ±2.88 / 257.57 ms │     no change │
│ QQuery 50 │        217.41 / 222.32 ±4.36 / 229.06 ms │        219.58 / 223.11 ±2.95 / 226.77 ms │     no change │
│ QQuery 51 │        177.39 / 178.81 ±1.14 / 180.19 ms │        176.89 / 180.54 ±2.21 / 183.86 ms │     no change │
│ QQuery 52 │        106.06 / 108.16 ±1.55 / 110.20 ms │        106.78 / 107.39 ±0.50 / 108.17 ms │     no change │
│ QQuery 53 │        100.87 / 101.69 ±0.50 / 102.33 ms │        102.13 / 103.49 ±1.20 / 105.42 ms │     no change │
│ QQuery 54 │        144.34 / 146.96 ±1.95 / 149.44 ms │        144.60 / 145.77 ±0.80 / 147.10 ms │     no change │
│ QQuery 55 │        106.58 / 107.92 ±0.99 / 109.21 ms │        108.02 / 109.61 ±1.89 / 112.52 ms │     no change │
│ QQuery 56 │        139.54 / 140.65 ±1.22 / 142.85 ms │        140.63 / 142.69 ±1.48 / 145.13 ms │     no change │
│ QQuery 57 │        174.20 / 175.47 ±1.06 / 176.88 ms │        172.93 / 176.50 ±2.09 / 179.36 ms │     no change │
│ QQuery 58 │        283.19 / 295.97 ±7.72 / 304.93 ms │       292.32 / 305.75 ±13.16 / 329.87 ms │     no change │
│ QQuery 59 │        198.09 / 199.10 ±1.29 / 201.58 ms │        200.75 / 201.93 ±1.03 / 203.35 ms │     no change │
│ QQuery 60 │        141.31 / 143.23 ±1.28 / 145.25 ms │        145.44 / 146.58 ±0.88 / 147.92 ms │     no change │
│ QQuery 61 │           13.02 / 13.36 ±0.32 / 13.93 ms │           13.01 / 13.28 ±0.29 / 13.84 ms │     no change │
│ QQuery 62 │       885.72 / 906.83 ±20.43 / 937.25 ms │       883.77 / 930.18 ±37.24 / 989.28 ms │     no change │
│ QQuery 63 │        102.05 / 103.77 ±1.60 / 106.32 ms │        103.84 / 106.24 ±2.36 / 110.75 ms │     no change │
│ QQuery 64 │        670.01 / 676.70 ±4.62 / 683.35 ms │        668.39 / 674.23 ±5.20 / 680.77 ms │     no change │
│ QQuery 65 │        246.70 / 249.12 ±1.85 / 251.21 ms │        247.20 / 249.82 ±2.47 / 253.87 ms │     no change │
│ QQuery 66 │        243.95 / 249.21 ±6.08 / 260.42 ms │        237.96 / 252.22 ±8.14 / 263.08 ms │     no change │
│ QQuery 67 │        308.25 / 314.87 ±4.30 / 320.89 ms │        296.39 / 310.06 ±8.48 / 318.96 ms │     no change │
│ QQuery 68 │            8.62 / 10.08 ±1.76 / 13.49 ms │             8.50 / 9.89 ±0.98 / 10.80 ms │     no change │
│ QQuery 69 │        100.71 / 103.14 ±1.93 / 105.45 ms │        102.15 / 104.53 ±2.14 / 108.43 ms │     no change │
│ QQuery 70 │       335.60 / 346.67 ±10.95 / 367.37 ms │       330.92 / 348.02 ±12.87 / 368.76 ms │     no change │
│ QQuery 71 │        132.78 / 136.13 ±2.62 / 139.71 ms │        133.51 / 135.50 ±1.82 / 138.83 ms │     no change │
│ QQuery 72 │        607.28 / 618.33 ±7.98 / 627.82 ms │        611.17 / 620.16 ±7.98 / 629.99 ms │     no change │
│ QQuery 73 │              7.11 / 8.32 ±0.88 / 9.66 ms │              6.57 / 7.45 ±0.69 / 8.69 ms │ +1.12x faster │
│ QQuery 74 │        548.24 / 551.94 ±4.76 / 561.28 ms │        540.51 / 546.10 ±7.81 / 561.37 ms │     no change │
│ QQuery 75 │        273.60 / 276.68 ±2.04 / 279.78 ms │        276.98 / 278.16 ±0.82 / 279.34 ms │     no change │
│ QQuery 76 │        130.27 / 131.97 ±1.58 / 134.74 ms │        131.09 / 131.89 ±0.53 / 132.50 ms │     no change │
│ QQuery 77 │        187.84 / 189.48 ±1.28 / 191.45 ms │        185.59 / 187.72 ±1.92 / 191.33 ms │     no change │
│ QQuery 78 │        331.75 / 339.13 ±4.19 / 342.65 ms │        334.24 / 337.64 ±3.04 / 341.48 ms │     no change │
│ QQuery 79 │        228.61 / 230.94 ±1.88 / 234.29 ms │        226.90 / 231.36 ±2.94 / 235.38 ms │     no change │
│ QQuery 80 │        318.77 / 321.49 ±2.00 / 324.07 ms │        318.11 / 321.33 ±2.94 / 325.98 ms │     no change │
│ QQuery 81 │           25.39 / 26.39 ±0.86 / 27.35 ms │           25.47 / 26.20 ±0.56 / 26.93 ms │     no change │
│ QQuery 82 │        194.55 / 196.91 ±1.75 / 199.03 ms │        199.64 / 203.19 ±1.90 / 205.04 ms │     no change │
│ QQuery 83 │           38.16 / 38.78 ±0.76 / 40.24 ms │           37.93 / 39.47 ±1.66 / 42.42 ms │     no change │
│ QQuery 84 │           46.82 / 48.06 ±0.69 / 48.87 ms │           47.72 / 48.57 ±0.77 / 49.82 ms │     no change │
│ QQuery 85 │        144.27 / 146.86 ±1.43 / 148.44 ms │        145.23 / 146.95 ±0.95 / 147.85 ms │     no change │
│ QQuery 86 │           39.09 / 40.57 ±1.54 / 43.22 ms │           38.45 / 38.95 ±0.80 / 40.53 ms │     no change │
│ QQuery 87 │           84.87 / 87.39 ±2.36 / 91.64 ms │           85.39 / 87.78 ±2.28 / 91.94 ms │     no change │
│ QQuery 88 │         99.06 / 100.12 ±0.99 / 101.64 ms │         99.80 / 100.11 ±0.31 / 100.56 ms │     no change │
│ QQuery 89 │        115.20 / 117.79 ±1.48 / 119.07 ms │        119.43 / 120.34 ±0.79 / 121.56 ms │     no change │
│ QQuery 90 │           22.94 / 23.60 ±0.38 / 23.91 ms │           23.48 / 24.05 ±0.47 / 24.82 ms │     no change │
│ QQuery 91 │           62.26 / 63.13 ±0.63 / 64.14 ms │           63.79 / 64.43 ±0.66 / 65.61 ms │     no change │
│ QQuery 92 │           56.11 / 57.00 ±0.69 / 58.16 ms │           56.50 / 57.28 ±0.71 / 58.58 ms │     no change │
│ QQuery 93 │        183.49 / 185.32 ±1.51 / 187.25 ms │        184.29 / 186.04 ±1.49 / 188.58 ms │     no change │
│ QQuery 94 │           60.65 / 60.81 ±0.20 / 61.21 ms │           58.95 / 60.76 ±1.47 / 63.32 ms │     no change │
│ QQuery 95 │        125.10 / 126.31 ±0.89 / 127.47 ms │        126.27 / 127.63 ±1.04 / 129.42 ms │     no change │
│ QQuery 96 │           68.90 / 71.83 ±2.34 / 74.06 ms │           71.71 / 73.89 ±1.39 / 75.97 ms │     no change │
│ QQuery 97 │        122.55 / 124.24 ±1.25 / 126.25 ms │        125.77 / 127.37 ±0.98 / 128.79 ms │     no change │
│ QQuery 98 │        152.36 / 153.87 ±1.07 / 155.36 ms │        151.80 / 153.48 ±1.36 / 154.96 ms │     no change │
│ QQuery 99 │ 10715.96 / 10760.43 ±26.36 / 10791.72 ms │ 10703.55 / 10730.20 ±18.80 / 10750.03 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 31027.13ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 31096.66ms │
│ Average Time (HEAD)                                │   313.41ms │
│ Average Time (optimize-byteview-vectorized-append) │   314.11ms │
│ Queries Faster                                     │          1 │
│ Queries Slower                                     │          1 │
│ Queries with No Change                             │         97 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	155.5s
Peak memory	5.5 GiB
Avg memory	4.7 GiB
CPU user	256.4s
CPU sys	16.3s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	155.8s
Peak memory	5.5 GiB
Avg memory	4.6 GiB
CPU user	256.8s
CPU sys	16.1s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃   optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.19 / 4.45 ±6.37 / 17.18 ms │          1.20 / 4.51 ±6.45 / 17.42 ms │     no change │
│ QQuery 1  │        13.97 / 14.36 ±0.20 / 14.54 ms │        14.49 / 14.65 ±0.11 / 14.81 ms │     no change │
│ QQuery 2  │        42.95 / 43.18 ±0.19 / 43.51 ms │        43.83 / 44.17 ±0.39 / 44.92 ms │     no change │
│ QQuery 3  │        42.18 / 44.84 ±2.29 / 48.26 ms │        41.19 / 44.58 ±2.76 / 48.86 ms │     no change │
│ QQuery 4  │    291.52 / 302.54 ±15.70 / 333.52 ms │     287.17 / 295.90 ±8.06 / 309.51 ms │     no change │
│ QQuery 5  │     343.87 / 348.01 ±3.96 / 354.79 ms │     350.32 / 353.73 ±4.83 / 363.20 ms │     no change │
│ QQuery 6  │           5.04 / 5.89 ±0.62 / 6.77 ms │           6.48 / 7.19 ±0.53 / 7.72 ms │  1.22x slower │
│ QQuery 7  │        16.44 / 17.18 ±0.61 / 17.94 ms │        16.86 / 17.09 ±0.20 / 17.38 ms │     no change │
│ QQuery 8  │     426.83 / 435.01 ±9.36 / 452.87 ms │     419.03 / 430.40 ±8.68 / 439.94 ms │     no change │
│ QQuery 9  │     639.66 / 650.22 ±6.64 / 659.73 ms │     667.94 / 681.16 ±7.28 / 689.27 ms │     no change │
│ QQuery 10 │       91.38 / 96.20 ±3.49 / 100.76 ms │       92.80 / 96.35 ±3.92 / 103.66 ms │     no change │
│ QQuery 11 │     103.21 / 104.32 ±0.73 / 105.52 ms │     105.67 / 108.15 ±2.00 / 111.31 ms │     no change │
│ QQuery 12 │     334.27 / 342.17 ±4.35 / 346.47 ms │     346.28 / 352.75 ±4.48 / 359.85 ms │     no change │
│ QQuery 13 │     452.27 / 461.96 ±5.62 / 469.03 ms │     455.06 / 470.61 ±8.59 / 479.49 ms │     no change │
│ QQuery 14 │     347.88 / 350.28 ±1.51 / 352.42 ms │     351.16 / 355.06 ±2.86 / 358.72 ms │     no change │
│ QQuery 15 │     350.83 / 356.28 ±4.29 / 361.63 ms │    366.13 / 383.47 ±19.92 / 416.64 ms │  1.08x slower │
│ QQuery 16 │    712.98 / 727.77 ±14.85 / 753.68 ms │     714.51 / 724.75 ±9.33 / 742.18 ms │     no change │
│ QQuery 17 │    710.16 / 731.50 ±25.60 / 781.71 ms │     708.98 / 715.79 ±5.76 / 725.25 ms │     no change │
│ QQuery 18 │ 1447.68 / 1475.03 ±31.87 / 1533.55 ms │ 1404.69 / 1454.66 ±33.56 / 1489.16 ms │     no change │
│ QQuery 19 │        34.69 / 36.31 ±1.45 / 38.55 ms │      36.47 / 60.81 ±28.21 / 107.25 ms │  1.67x slower │
│ QQuery 20 │    718.30 / 736.27 ±18.57 / 762.10 ms │    716.32 / 742.03 ±26.71 / 793.18 ms │     no change │
│ QQuery 21 │     778.08 / 789.21 ±8.94 / 803.07 ms │     760.24 / 766.40 ±3.36 / 769.49 ms │     no change │
│ QQuery 22 │  1136.94 / 1143.12 ±7.73 / 1158.00 ms │  1133.98 / 1135.82 ±2.35 / 1140.34 ms │     no change │
│ QQuery 23 │ 3067.85 / 3093.70 ±15.80 / 3116.02 ms │ 3068.39 / 3106.08 ±22.19 / 3135.66 ms │     no change │
│ QQuery 24 │      98.94 / 101.42 ±2.20 / 104.02 ms │     100.48 / 106.22 ±4.75 / 114.72 ms │     no change │
│ QQuery 25 │     139.03 / 141.52 ±1.49 / 143.01 ms │     138.15 / 140.76 ±1.92 / 143.42 ms │     no change │
│ QQuery 26 │      98.13 / 100.92 ±1.57 / 102.71 ms │     101.11 / 103.55 ±2.16 / 107.28 ms │     no change │
│ QQuery 27 │     854.39 / 863.51 ±8.28 / 875.39 ms │     850.48 / 858.57 ±8.74 / 874.86 ms │     no change │
│ QQuery 28 │  3319.05 / 3329.83 ±9.77 / 3342.52 ms │ 3244.57 / 3276.34 ±17.82 / 3294.98 ms │     no change │
│ QQuery 29 │        51.14 / 54.20 ±3.64 / 60.75 ms │        50.00 / 53.85 ±3.23 / 58.08 ms │     no change │
│ QQuery 30 │     356.59 / 367.79 ±7.05 / 378.20 ms │     364.23 / 370.48 ±4.69 / 376.09 ms │     no change │
│ QQuery 31 │     378.58 / 383.22 ±5.43 / 392.51 ms │    382.96 / 391.01 ±11.52 / 413.34 ms │     no change │
│ QQuery 32 │ 1199.92 / 1242.18 ±38.79 / 1312.34 ms │ 1224.07 / 1274.37 ±55.38 / 1378.78 ms │     no change │
│ QQuery 33 │ 1528.74 / 1588.14 ±47.80 / 1667.07 ms │ 1532.99 / 1573.26 ±26.21 / 1606.45 ms │     no change │
│ QQuery 34 │ 1518.22 / 1598.06 ±70.04 / 1713.76 ms │ 1479.94 / 1496.52 ±10.09 / 1510.35 ms │ +1.07x faster │
│ QQuery 35 │    389.89 / 405.02 ±15.31 / 434.57 ms │     386.86 / 391.25 ±2.71 / 394.36 ms │     no change │
│ QQuery 36 │     118.25 / 120.11 ±1.50 / 121.98 ms │     119.86 / 121.40 ±1.53 / 124.26 ms │     no change │
│ QQuery 37 │        47.90 / 49.04 ±0.97 / 50.61 ms │        47.85 / 49.44 ±1.62 / 52.17 ms │     no change │
│ QQuery 38 │        77.42 / 79.27 ±1.31 / 80.82 ms │        75.12 / 76.78 ±1.92 / 80.20 ms │     no change │
│ QQuery 39 │     211.98 / 221.07 ±6.47 / 229.63 ms │     201.36 / 213.37 ±9.38 / 228.53 ms │     no change │
│ QQuery 40 │        24.86 / 25.78 ±0.68 / 26.75 ms │        22.69 / 25.14 ±2.77 / 30.25 ms │     no change │
│ QQuery 41 │        20.61 / 22.03 ±0.88 / 23.08 ms │        20.16 / 21.25 ±0.65 / 22.21 ms │     no change │
│ QQuery 42 │        19.13 / 19.77 ±0.39 / 20.30 ms │        19.60 / 20.10 ±0.30 / 20.48 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 23022.68ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 22929.76ms │
│ Average Time (HEAD)                                │   535.41ms │
│ Average Time (optimize-byteview-vectorized-append) │   533.25ms │
│ Queries Faster                                     │          1 │
│ Queries Slower                                     │          3 │
│ Queries with No Change                             │         39 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	116.0s
Peak memory	34.8 GiB
Avg memory	25.9 GiB
CPU user	1078.6s
CPU sys	100.6s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	115.7s
Peak memory	37.9 GiB
Avg memory	27.4 GiB
CPU user	1077.6s
CPU sys	98.8s
Peak spill	0 B

---

File an issue against this benchmark runner

[Dandandan]

run benchmarks

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231241460-1112-67jl5 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (ac29b32) to 16e578d (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231241460-1113-xm55h 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (ac29b32) to 16e578d (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231241460-1114-zldtp 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (ac29b32) to 16e578d (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃   optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.19 / 4.39 ±6.32 / 17.03 ms │          1.17 / 4.34 ±6.25 / 16.83 ms │     no change │
│ QQuery 1  │        14.21 / 14.48 ±0.15 / 14.63 ms │        14.07 / 14.33 ±0.16 / 14.50 ms │     no change │
│ QQuery 2  │        43.13 / 43.50 ±0.31 / 43.92 ms │        43.59 / 44.08 ±0.41 / 44.55 ms │     no change │
│ QQuery 3  │        42.23 / 43.94 ±1.06 / 45.28 ms │        43.92 / 44.30 ±0.26 / 44.64 ms │     no change │
│ QQuery 4  │     278.30 / 290.21 ±8.62 / 301.00 ms │     273.05 / 283.16 ±5.24 / 286.92 ms │     no change │
│ QQuery 5  │     333.74 / 339.87 ±5.25 / 346.67 ms │     333.89 / 338.50 ±3.72 / 344.62 ms │     no change │
│ QQuery 6  │           5.27 / 6.13 ±0.63 / 7.05 ms │           5.63 / 6.70 ±0.72 / 7.80 ms │  1.09x slower │
│ QQuery 7  │        16.72 / 17.39 ±0.49 / 17.93 ms │        16.25 / 17.09 ±0.87 / 18.43 ms │     no change │
│ QQuery 8  │     407.28 / 413.37 ±6.34 / 424.37 ms │     410.32 / 413.00 ±2.99 / 418.66 ms │     no change │
│ QQuery 9  │     615.05 / 624.84 ±6.33 / 631.95 ms │     627.20 / 634.26 ±4.06 / 638.27 ms │     no change │
│ QQuery 10 │        90.54 / 91.51 ±1.17 / 93.71 ms │        91.44 / 92.49 ±1.04 / 93.79 ms │     no change │
│ QQuery 11 │     102.59 / 104.09 ±1.45 / 106.62 ms │     103.26 / 103.98 ±0.51 / 104.56 ms │     no change │
│ QQuery 12 │     328.25 / 333.08 ±3.78 / 337.44 ms │     334.76 / 336.64 ±1.48 / 338.57 ms │     no change │
│ QQuery 13 │    448.73 / 460.63 ±10.72 / 476.82 ms │    453.79 / 467.47 ±12.34 / 489.81 ms │     no change │
│ QQuery 14 │     341.63 / 342.52 ±0.97 / 343.89 ms │     336.33 / 339.82 ±3.74 / 345.38 ms │     no change │
│ QQuery 15 │     332.96 / 348.74 ±9.65 / 363.08 ms │    342.59 / 362.67 ±19.61 / 396.84 ms │     no change │
│ QQuery 16 │     693.89 / 702.50 ±5.36 / 707.04 ms │    827.54 / 856.23 ±23.83 / 890.98 ms │  1.22x slower │
│ QQuery 17 │     693.99 / 697.84 ±2.77 / 701.41 ms │    826.22 / 835.31 ±11.04 / 856.97 ms │  1.20x slower │
│ QQuery 18 │ 1330.55 / 1337.86 ±11.80 / 1361.40 ms │ 1634.12 / 1689.75 ±31.17 / 1721.24 ms │  1.26x slower │
│ QQuery 19 │        34.79 / 36.74 ±1.70 / 39.63 ms │        34.56 / 35.12 ±0.42 / 35.73 ms │     no change │
│ QQuery 20 │    712.74 / 723.52 ±12.55 / 745.93 ms │     711.32 / 719.11 ±8.47 / 731.52 ms │     no change │
│ QQuery 21 │     753.63 / 756.45 ±2.69 / 759.75 ms │     754.50 / 759.91 ±4.80 / 767.71 ms │     no change │
│ QQuery 22 │  1121.34 / 1129.24 ±7.04 / 1141.46 ms │  1121.08 / 1123.58 ±3.14 / 1129.74 ms │     no change │
│ QQuery 23 │ 3027.63 / 3044.83 ±19.22 / 3081.71 ms │ 3031.08 / 3056.76 ±19.14 / 3088.65 ms │     no change │
│ QQuery 24 │      98.67 / 102.21 ±3.52 / 108.59 ms │      96.23 / 100.97 ±3.71 / 107.13 ms │     no change │
│ QQuery 25 │     138.14 / 139.84 ±0.94 / 140.75 ms │     139.95 / 141.25 ±1.23 / 143.06 ms │     no change │
│ QQuery 26 │      98.88 / 101.19 ±1.38 / 103.10 ms │      97.87 / 101.11 ±1.69 / 102.40 ms │     no change │
│ QQuery 27 │     845.92 / 850.63 ±3.60 / 855.88 ms │     841.87 / 846.48 ±6.12 / 857.46 ms │     no change │
│ QQuery 28 │ 3255.10 / 3284.83 ±25.28 / 3325.90 ms │ 3249.87 / 3276.75 ±17.16 / 3294.22 ms │     no change │
│ QQuery 29 │        49.60 / 54.61 ±5.89 / 64.47 ms │       50.15 / 56.75 ±11.44 / 79.53 ms │     no change │
│ QQuery 30 │     352.83 / 358.95 ±5.08 / 366.70 ms │     356.31 / 360.84 ±4.18 / 368.14 ms │     no change │
│ QQuery 31 │    356.65 / 378.13 ±16.73 / 399.47 ms │    349.61 / 370.13 ±10.66 / 380.76 ms │     no change │
│ QQuery 32 │ 1010.83 / 1025.88 ±14.40 / 1053.26 ms │ 1233.04 / 1269.32 ±36.71 / 1338.42 ms │  1.24x slower │
│ QQuery 33 │  1428.76 / 1442.02 ±9.72 / 1452.74 ms │ 1433.17 / 1474.96 ±42.41 / 1556.44 ms │     no change │
│ QQuery 34 │  1422.90 / 1425.98 ±2.35 / 1429.67 ms │  1421.21 / 1433.34 ±6.86 / 1439.39 ms │     no change │
│ QQuery 35 │     368.20 / 373.87 ±3.26 / 378.32 ms │     370.12 / 376.11 ±4.28 / 380.84 ms │     no change │
│ QQuery 36 │     111.71 / 118.71 ±4.63 / 124.65 ms │     109.46 / 117.43 ±4.36 / 121.69 ms │     no change │
│ QQuery 37 │        47.99 / 49.81 ±1.37 / 52.00 ms │        48.31 / 49.50 ±0.99 / 50.98 ms │     no change │
│ QQuery 38 │        74.04 / 76.54 ±1.40 / 78.19 ms │        73.84 / 75.48 ±0.94 / 76.74 ms │     no change │
│ QQuery 39 │     200.36 / 215.67 ±8.30 / 225.15 ms │     212.24 / 219.44 ±3.82 / 223.69 ms │     no change │
│ QQuery 40 │        21.28 / 24.57 ±1.98 / 27.45 ms │        21.15 / 23.31 ±1.70 / 26.36 ms │ +1.05x faster │
│ QQuery 41 │        19.23 / 21.18 ±1.33 / 22.40 ms │        19.73 / 21.45 ±1.28 / 23.18 ms │     no change │
│ QQuery 42 │        19.11 / 19.46 ±0.30 / 19.97 ms │        18.01 / 18.98 ±0.57 / 19.49 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 21971.74ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 22912.22ms │
│ Average Time (HEAD)                                │   510.97ms │
│ Average Time (optimize-byteview-vectorized-append) │   532.84ms │
│ Queries Faster                                     │          1 │
│ Queries Slower                                     │          5 │
│ Queries with No Change                             │         37 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	110.9s
Peak memory	42.2 GiB
Avg memory	33.2 GiB
CPU user	1059.9s
CPU sys	73.8s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	115.6s
Peak memory	42.8 GiB
Avg memory	29.6 GiB
CPU user	1084.1s
CPU sys	93.9s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃      optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │              6.57 / 7.08 ±0.84 / 8.75 ms │              6.62 / 7.10 ±0.80 / 8.70 ms │     no change │
│ QQuery 2  │        146.79 / 147.65 ±0.56 / 148.42 ms │        144.66 / 145.52 ±0.76 / 146.81 ms │     no change │
│ QQuery 3  │        114.15 / 114.78 ±0.34 / 115.13 ms │        114.02 / 114.68 ±0.97 / 116.61 ms │     no change │
│ QQuery 4  │    1357.16 / 1390.43 ±22.24 / 1423.60 ms │    1346.32 / 1382.33 ±18.48 / 1398.82 ms │     no change │
│ QQuery 5  │        173.82 / 176.29 ±1.92 / 179.13 ms │        173.64 / 175.13 ±1.23 / 177.05 ms │     no change │
│ QQuery 6  │        854.27 / 865.13 ±8.27 / 878.57 ms │       848.14 / 870.39 ±22.26 / 910.47 ms │     no change │
│ QQuery 7  │        343.08 / 347.65 ±3.43 / 353.21 ms │        348.16 / 352.31 ±2.88 / 355.67 ms │     no change │
│ QQuery 8  │        116.52 / 117.65 ±0.69 / 118.40 ms │        118.14 / 119.48 ±1.00 / 121.23 ms │     no change │
│ QQuery 9  │        101.28 / 103.12 ±2.66 / 108.42 ms │        102.59 / 106.09 ±2.70 / 108.99 ms │     no change │
│ QQuery 10 │        106.03 / 108.37 ±1.38 / 109.51 ms │        108.81 / 109.82 ±0.78 / 110.91 ms │     no change │
│ QQuery 11 │        978.91 / 985.39 ±5.84 / 995.87 ms │        969.86 / 978.07 ±7.74 / 990.85 ms │     no change │
│ QQuery 12 │           46.76 / 48.32 ±1.14 / 50.00 ms │           45.11 / 47.37 ±1.20 / 48.66 ms │     no change │
│ QQuery 13 │        404.56 / 408.87 ±3.39 / 414.00 ms │        403.26 / 405.51 ±1.49 / 407.87 ms │     no change │
│ QQuery 14 │     1017.34 / 1021.28 ±6.12 / 1033.40 ms │      995.63 / 1007.82 ±8.43 / 1017.38 ms │     no change │
│ QQuery 15 │           16.46 / 17.75 ±0.85 / 18.88 ms │           15.31 / 16.60 ±0.85 / 17.59 ms │ +1.07x faster │
│ QQuery 16 │              7.30 / 7.56 ±0.17 / 7.76 ms │              7.05 / 7.52 ±0.48 / 8.35 ms │     no change │
│ QQuery 17 │        229.65 / 230.37 ±0.52 / 230.95 ms │        226.64 / 228.34 ±0.98 / 229.47 ms │     no change │
│ QQuery 18 │        127.72 / 128.64 ±0.87 / 130.09 ms │        129.78 / 130.27 ±0.45 / 131.09 ms │     no change │
│ QQuery 19 │        155.91 / 157.92 ±1.43 / 159.66 ms │        154.25 / 156.14 ±1.26 / 157.89 ms │     no change │
│ QQuery 20 │           14.21 / 14.79 ±0.60 / 15.86 ms │           13.54 / 14.01 ±0.44 / 14.69 ms │ +1.06x faster │
│ QQuery 21 │           20.78 / 21.45 ±0.75 / 22.48 ms │           19.22 / 19.60 ±0.57 / 20.73 ms │ +1.09x faster │
│ QQuery 22 │        484.69 / 491.27 ±5.60 / 500.39 ms │        483.84 / 490.39 ±5.47 / 497.57 ms │     no change │
│ QQuery 23 │        879.48 / 892.40 ±8.17 / 904.96 ms │        865.08 / 877.15 ±6.36 / 882.86 ms │     no change │
│ QQuery 24 │        382.77 / 386.85 ±2.28 / 389.12 ms │        378.79 / 385.53 ±4.72 / 391.07 ms │     no change │
│ QQuery 25 │        341.31 / 342.83 ±1.33 / 345.02 ms │        339.33 / 342.50 ±1.96 / 345.09 ms │     no change │
│ QQuery 26 │           79.96 / 82.11 ±1.42 / 84.35 ms │           82.35 / 83.72 ±1.31 / 85.97 ms │     no change │
│ QQuery 27 │              6.83 / 7.17 ±0.26 / 7.57 ms │              6.93 / 7.78 ±0.78 / 9.09 ms │  1.09x slower │
│ QQuery 28 │        148.23 / 150.22 ±1.37 / 152.35 ms │        148.66 / 150.10 ±1.37 / 152.30 ms │     no change │
│ QQuery 29 │        282.19 / 284.37 ±1.81 / 287.57 ms │        282.46 / 283.82 ±1.15 / 285.36 ms │     no change │
│ QQuery 30 │           43.39 / 44.79 ±1.47 / 46.83 ms │           43.74 / 45.03 ±1.04 / 46.88 ms │     no change │
│ QQuery 31 │        170.53 / 173.05 ±1.36 / 174.28 ms │        169.74 / 171.92 ±1.63 / 174.72 ms │     no change │
│ QQuery 32 │           57.18 / 58.81 ±0.95 / 60.15 ms │           56.96 / 57.97 ±1.03 / 59.60 ms │     no change │
│ QQuery 33 │        139.93 / 141.45 ±1.11 / 143.06 ms │        140.86 / 142.14 ±0.94 / 143.46 ms │     no change │
│ QQuery 34 │              6.90 / 7.23 ±0.29 / 7.56 ms │              6.99 / 7.17 ±0.16 / 7.46 ms │     no change │
│ QQuery 35 │        109.17 / 110.76 ±1.12 / 112.63 ms │        107.90 / 110.39 ±1.60 / 112.83 ms │     no change │
│ QQuery 36 │              6.59 / 7.14 ±0.60 / 8.29 ms │              6.50 / 6.92 ±0.35 / 7.54 ms │     no change │
│ QQuery 37 │              8.32 / 8.67 ±0.36 / 9.17 ms │              8.59 / 8.81 ±0.21 / 9.09 ms │     no change │
│ QQuery 38 │           82.48 / 89.38 ±4.88 / 97.42 ms │           85.15 / 88.46 ±3.32 / 93.27 ms │     no change │
│ QQuery 39 │        124.89 / 126.75 ±1.46 / 129.29 ms │        122.90 / 126.87 ±2.89 / 129.91 ms │     no change │
│ QQuery 40 │        108.56 / 115.81 ±7.68 / 130.44 ms │        108.19 / 114.50 ±6.29 / 125.92 ms │     no change │
│ QQuery 41 │           14.34 / 15.54 ±0.98 / 16.51 ms │           14.45 / 15.47 ±0.58 / 16.11 ms │     no change │
│ QQuery 42 │        109.76 / 110.30 ±0.48 / 111.08 ms │        108.10 / 109.72 ±1.23 / 111.60 ms │     no change │
│ QQuery 43 │              6.08 / 6.23 ±0.15 / 6.52 ms │              5.90 / 6.11 ±0.22 / 6.54 ms │     no change │
│ QQuery 44 │           11.60 / 12.09 ±0.38 / 12.54 ms │           11.47 / 11.93 ±0.51 / 12.92 ms │     no change │
│ QQuery 45 │           50.64 / 51.23 ±0.48 / 51.80 ms │           50.33 / 52.02 ±1.40 / 54.04 ms │     no change │
│ QQuery 46 │              8.73 / 8.94 ±0.19 / 9.25 ms │              8.53 / 8.95 ±0.49 / 9.88 ms │     no change │
│ QQuery 47 │        731.84 / 737.75 ±3.63 / 742.36 ms │        715.80 / 720.70 ±3.19 / 725.64 ms │     no change │
│ QQuery 48 │        287.33 / 291.60 ±3.74 / 296.31 ms │        284.40 / 293.34 ±6.52 / 302.69 ms │     no change │
│ QQuery 49 │        253.70 / 257.59 ±2.46 / 260.17 ms │        251.30 / 252.48 ±0.87 / 253.62 ms │     no change │
│ QQuery 50 │        225.18 / 229.44 ±2.93 / 234.11 ms │        223.48 / 228.99 ±4.12 / 234.51 ms │     no change │
│ QQuery 51 │        178.77 / 182.62 ±2.11 / 185.03 ms │        179.34 / 182.36 ±2.34 / 186.06 ms │     no change │
│ QQuery 52 │        107.83 / 109.74 ±1.53 / 111.27 ms │        108.37 / 109.53 ±0.81 / 110.83 ms │     no change │
│ QQuery 53 │        104.05 / 104.76 ±0.93 / 106.52 ms │        102.62 / 103.68 ±0.63 / 104.40 ms │     no change │
│ QQuery 54 │        145.68 / 147.23 ±1.25 / 148.91 ms │        147.05 / 147.95 ±1.00 / 149.57 ms │     no change │
│ QQuery 55 │        107.69 / 110.52 ±2.00 / 113.36 ms │        106.16 / 108.20 ±1.36 / 110.19 ms │     no change │
│ QQuery 56 │        141.47 / 142.90 ±1.16 / 144.72 ms │        140.50 / 142.72 ±1.30 / 143.91 ms │     no change │
│ QQuery 57 │        175.20 / 177.78 ±2.08 / 180.90 ms │        172.81 / 174.55 ±1.26 / 176.56 ms │     no change │
│ QQuery 58 │       295.06 / 312.54 ±15.80 / 339.04 ms │        285.07 / 292.28 ±4.54 / 296.50 ms │ +1.07x faster │
│ QQuery 59 │        198.44 / 201.49 ±2.29 / 204.67 ms │        198.69 / 199.68 ±0.89 / 201.04 ms │     no change │
│ QQuery 60 │        145.33 / 147.25 ±1.41 / 148.84 ms │        142.64 / 144.42 ±1.47 / 146.76 ms │     no change │
│ QQuery 61 │           13.23 / 13.37 ±0.14 / 13.56 ms │           13.48 / 13.71 ±0.22 / 14.12 ms │     no change │
│ QQuery 62 │      911.56 / 953.30 ±36.38 / 1008.32 ms │       909.54 / 937.47 ±24.77 / 980.41 ms │     no change │
│ QQuery 63 │        102.71 / 107.18 ±4.30 / 114.67 ms │        102.35 / 105.26 ±1.92 / 107.95 ms │     no change │
│ QQuery 64 │        688.91 / 691.61 ±3.30 / 698.02 ms │        685.22 / 690.74 ±3.51 / 696.07 ms │     no change │
│ QQuery 65 │        258.17 / 261.41 ±2.59 / 264.43 ms │        253.33 / 257.37 ±2.80 / 261.51 ms │     no change │
│ QQuery 66 │       240.31 / 262.23 ±14.36 / 278.87 ms │        244.52 / 251.88 ±5.29 / 258.74 ms │     no change │
│ QQuery 67 │        310.70 / 321.03 ±7.11 / 328.98 ms │        308.65 / 312.81 ±3.40 / 317.46 ms │     no change │
│ QQuery 68 │            9.01 / 10.13 ±0.67 / 10.69 ms │            9.17 / 10.27 ±0.90 / 11.54 ms │     no change │
│ QQuery 69 │        102.67 / 104.50 ±1.16 / 106.14 ms │        104.50 / 105.04 ±0.42 / 105.69 ms │     no change │
│ QQuery 70 │       343.14 / 370.10 ±17.55 / 386.88 ms │        335.18 / 348.47 ±9.72 / 364.07 ms │ +1.06x faster │
│ QQuery 71 │        135.10 / 137.66 ±3.55 / 144.48 ms │        133.63 / 135.86 ±1.39 / 137.91 ms │     no change │
│ QQuery 72 │        617.76 / 625.83 ±5.11 / 631.84 ms │        614.32 / 623.27 ±6.92 / 633.98 ms │     no change │
│ QQuery 73 │              6.59 / 7.43 ±0.87 / 9.10 ms │              6.73 / 7.89 ±0.80 / 9.10 ms │  1.06x slower │
│ QQuery 74 │        606.37 / 615.68 ±9.52 / 630.58 ms │        596.91 / 604.48 ±6.96 / 615.87 ms │     no change │
│ QQuery 75 │        275.93 / 279.17 ±1.85 / 281.44 ms │        276.76 / 280.31 ±2.41 / 283.73 ms │     no change │
│ QQuery 76 │        133.08 / 134.37 ±1.26 / 136.21 ms │        131.73 / 134.30 ±1.76 / 136.59 ms │     no change │
│ QQuery 77 │        186.45 / 188.77 ±1.70 / 191.32 ms │        187.29 / 189.97 ±2.47 / 193.58 ms │     no change │
│ QQuery 78 │        340.02 / 347.33 ±4.77 / 354.75 ms │        341.14 / 343.25 ±1.34 / 345.28 ms │     no change │
│ QQuery 79 │        230.21 / 235.16 ±2.86 / 238.34 ms │        233.93 / 234.93 ±0.61 / 235.63 ms │     no change │
│ QQuery 80 │        322.58 / 324.60 ±2.25 / 328.26 ms │        320.60 / 324.26 ±2.98 / 329.44 ms │     no change │
│ QQuery 81 │           26.16 / 27.30 ±0.74 / 28.21 ms │           26.93 / 27.55 ±0.60 / 28.53 ms │     no change │
│ QQuery 82 │        199.70 / 200.34 ±0.50 / 200.81 ms │        197.71 / 200.57 ±2.97 / 206.08 ms │     no change │
│ QQuery 83 │           38.25 / 39.98 ±1.16 / 41.83 ms │           38.84 / 40.35 ±1.91 / 43.95 ms │     no change │
│ QQuery 84 │           48.05 / 49.20 ±0.87 / 50.21 ms │           48.44 / 49.40 ±0.65 / 50.23 ms │     no change │
│ QQuery 85 │        148.34 / 150.02 ±1.80 / 153.49 ms │        147.92 / 149.53 ±0.89 / 150.39 ms │     no change │
│ QQuery 86 │           39.54 / 40.19 ±0.44 / 40.73 ms │           39.03 / 40.34 ±1.49 / 43.24 ms │     no change │
│ QQuery 87 │           86.12 / 90.41 ±3.04 / 94.81 ms │           83.60 / 87.88 ±3.45 / 92.16 ms │     no change │
│ QQuery 88 │        101.81 / 102.64 ±0.62 / 103.59 ms │        100.79 / 101.55 ±0.65 / 102.38 ms │     no change │
│ QQuery 89 │        118.61 / 121.47 ±1.57 / 122.92 ms │        117.82 / 119.69 ±1.63 / 121.87 ms │     no change │
│ QQuery 90 │           23.25 / 24.50 ±0.70 / 25.41 ms │           23.41 / 24.20 ±0.73 / 25.33 ms │     no change │
│ QQuery 91 │           63.76 / 64.55 ±0.70 / 65.84 ms │           64.81 / 65.83 ±0.90 / 67.39 ms │     no change │
│ QQuery 92 │           57.45 / 58.71 ±0.76 / 59.61 ms │           57.71 / 58.99 ±0.80 / 60.02 ms │     no change │
│ QQuery 93 │        187.95 / 189.96 ±1.51 / 192.30 ms │        188.22 / 189.49 ±1.19 / 191.35 ms │     no change │
│ QQuery 94 │           63.17 / 63.68 ±0.64 / 64.90 ms │           61.30 / 61.78 ±0.56 / 62.87 ms │     no change │
│ QQuery 95 │        127.86 / 130.03 ±1.70 / 132.79 ms │        127.19 / 129.02 ±1.93 / 132.56 ms │     no change │
│ QQuery 96 │           71.47 / 73.93 ±1.28 / 75.12 ms │           72.97 / 74.31 ±1.13 / 76.29 ms │     no change │
│ QQuery 97 │        124.30 / 127.39 ±1.77 / 129.82 ms │        125.48 / 126.60 ±0.88 / 128.05 ms │     no change │
│ QQuery 98 │        152.99 / 156.59 ±2.33 / 160.09 ms │        153.60 / 158.22 ±2.70 / 161.01 ms │     no change │
│ QQuery 99 │ 10852.34 / 10876.26 ±16.24 / 10899.14 ms │ 10791.65 / 10832.18 ±30.58 / 10883.24 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 31877.01ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 31649.37ms │
│ Average Time (HEAD)                                │   321.99ms │
│ Average Time (optimize-byteview-vectorized-append) │   319.69ms │
│ Queries Faster                                     │          5 │
│ Queries Slower                                     │          2 │
│ Queries with No Change                             │         92 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	159.7s
Peak memory	5.3 GiB
Avg memory	4.4 GiB
CPU user	263.0s
CPU sys	17.9s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	158.5s
Peak memory	5.2 GiB
Avg memory	4.4 GiB
CPU user	261.3s
CPU sys	16.8s
Peak spill	0 B

---

File an issue against this benchmark runner

[Dandandan]

run benchmark clickbench_partitioned

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231314993-1115-twclp 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (ac29b32) to 16e578d (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃   optimize-byteview-vectorized-append ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.20 / 4.45 ±6.36 / 17.17 ms │          1.23 / 4.47 ±6.37 / 17.21 ms │    no change │
│ QQuery 1  │        14.06 / 14.40 ±0.18 / 14.57 ms │        14.16 / 14.52 ±0.20 / 14.70 ms │    no change │
│ QQuery 2  │        43.32 / 44.65 ±1.54 / 47.38 ms │        43.49 / 43.82 ±0.24 / 44.23 ms │    no change │
│ QQuery 3  │        43.22 / 45.32 ±1.09 / 46.39 ms │        41.42 / 44.13 ±3.00 / 49.48 ms │    no change │
│ QQuery 4  │     282.91 / 291.90 ±6.19 / 300.76 ms │    290.04 / 306.17 ±17.84 / 336.46 ms │    no change │
│ QQuery 5  │     341.30 / 346.41 ±4.36 / 351.14 ms │     337.60 / 348.18 ±6.28 / 355.81 ms │    no change │
│ QQuery 6  │           5.72 / 6.82 ±0.87 / 8.12 ms │           5.78 / 6.67 ±0.62 / 7.47 ms │    no change │
│ QQuery 7  │        16.63 / 17.85 ±1.61 / 20.97 ms │        16.61 / 17.38 ±0.70 / 18.52 ms │    no change │
│ QQuery 8  │     408.63 / 420.09 ±9.74 / 431.25 ms │    415.66 / 435.19 ±15.85 / 461.22 ms │    no change │
│ QQuery 9  │     643.60 / 657.06 ±7.43 / 665.77 ms │    649.01 / 663.83 ±10.30 / 678.18 ms │    no change │
│ QQuery 10 │       92.44 / 95.49 ±3.17 / 101.27 ms │        94.05 / 95.74 ±1.17 / 96.91 ms │    no change │
│ QQuery 11 │     105.41 / 106.65 ±1.06 / 108.35 ms │     105.41 / 108.40 ±4.97 / 118.33 ms │    no change │
│ QQuery 12 │     345.81 / 351.58 ±4.21 / 358.45 ms │     340.14 / 351.43 ±9.23 / 363.95 ms │    no change │
│ QQuery 13 │     458.22 / 469.35 ±9.26 / 484.07 ms │     453.49 / 459.24 ±6.18 / 470.62 ms │    no change │
│ QQuery 14 │     348.41 / 353.77 ±4.44 / 359.59 ms │     349.40 / 353.80 ±2.67 / 356.96 ms │    no change │
│ QQuery 15 │     358.16 / 369.98 ±7.26 / 380.67 ms │    359.78 / 376.06 ±17.03 / 399.47 ms │    no change │
│ QQuery 16 │     720.05 / 723.64 ±2.40 / 727.24 ms │    833.51 / 857.94 ±12.96 / 869.97 ms │ 1.19x slower │
│ QQuery 17 │    711.90 / 744.53 ±26.74 / 774.67 ms │     838.35 / 845.77 ±7.03 / 856.05 ms │ 1.14x slower │
│ QQuery 18 │ 1467.63 / 1493.46 ±19.70 / 1519.93 ms │ 1629.93 / 1706.80 ±43.63 / 1751.82 ms │ 1.14x slower │
│ QQuery 19 │        34.96 / 36.43 ±1.80 / 39.96 ms │        36.77 / 43.23 ±9.57 / 62.27 ms │ 1.19x slower │
│ QQuery 20 │    724.38 / 743.32 ±20.57 / 772.39 ms │    717.48 / 726.74 ±13.91 / 754.38 ms │    no change │
│ QQuery 21 │     767.24 / 774.29 ±4.25 / 779.79 ms │     760.31 / 766.57 ±3.83 / 770.73 ms │    no change │
│ QQuery 22 │  1137.95 / 1144.68 ±5.43 / 1154.27 ms │  1139.96 / 1143.52 ±2.56 / 1147.45 ms │    no change │
│ QQuery 23 │ 3078.09 / 3090.18 ±11.52 / 3110.74 ms │ 3109.17 / 3132.56 ±14.47 / 3148.40 ms │    no change │
│ QQuery 24 │     101.18 / 103.28 ±2.62 / 108.30 ms │     103.69 / 106.27 ±2.59 / 110.59 ms │    no change │
│ QQuery 25 │     138.62 / 140.18 ±1.54 / 142.90 ms │     140.01 / 141.80 ±1.70 / 144.13 ms │    no change │
│ QQuery 26 │      98.25 / 100.96 ±1.36 / 101.77 ms │      99.96 / 102.55 ±1.79 / 105.26 ms │    no change │
│ QQuery 27 │     853.21 / 858.69 ±3.76 / 864.80 ms │     855.38 / 860.12 ±5.55 / 867.82 ms │    no change │
│ QQuery 28 │ 3272.08 / 3293.80 ±12.21 / 3309.71 ms │ 3281.76 / 3306.20 ±19.54 / 3329.90 ms │    no change │
│ QQuery 29 │        51.44 / 55.45 ±4.22 / 62.90 ms │        50.34 / 54.24 ±3.60 / 60.33 ms │    no change │
│ QQuery 30 │     361.78 / 367.73 ±5.21 / 376.74 ms │     360.51 / 368.73 ±4.27 / 372.19 ms │    no change │
│ QQuery 31 │    366.09 / 378.97 ±12.13 / 400.75 ms │     366.84 / 377.50 ±9.53 / 393.44 ms │    no change │
│ QQuery 32 │ 1043.30 / 1065.99 ±26.08 / 1114.60 ms │ 1161.01 / 1273.95 ±70.11 / 1374.22 ms │ 1.20x slower │
│ QQuery 33 │ 1439.47 / 1457.25 ±11.82 / 1472.44 ms │ 1511.32 / 1558.19 ±29.20 / 1593.54 ms │ 1.07x slower │
│ QQuery 34 │ 1470.05 / 1477.82 ±10.04 / 1496.79 ms │ 1523.93 / 1589.31 ±37.82 / 1629.77 ms │ 1.08x slower │
│ QQuery 35 │     389.55 / 396.52 ±5.57 / 404.77 ms │     391.45 / 396.90 ±6.14 / 407.14 ms │    no change │
│ QQuery 36 │     114.33 / 121.85 ±4.50 / 127.11 ms │     114.07 / 120.63 ±3.64 / 124.15 ms │    no change │
│ QQuery 37 │        46.74 / 49.54 ±1.50 / 51.11 ms │        46.89 / 48.26 ±0.72 / 48.94 ms │    no change │
│ QQuery 38 │        76.77 / 78.26 ±0.91 / 79.29 ms │        74.44 / 76.26 ±1.74 / 79.15 ms │    no change │
│ QQuery 39 │     209.28 / 217.76 ±5.83 / 226.04 ms │     216.08 / 220.08 ±3.89 / 225.70 ms │    no change │
│ QQuery 40 │        24.51 / 26.81 ±2.98 / 32.49 ms │        23.74 / 27.49 ±2.53 / 30.80 ms │    no change │
│ QQuery 41 │        20.28 / 21.62 ±1.18 / 23.12 ms │        21.02 / 21.62 ±0.81 / 23.23 ms │    no change │
│ QQuery 42 │        19.51 / 20.06 ±0.34 / 20.53 ms │        19.09 / 20.34 ±1.26 / 22.45 ms │    no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 22578.87ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 23522.61ms │
│ Average Time (HEAD)                                │   525.09ms │
│ Average Time (optimize-byteview-vectorized-append) │   547.04ms │
│ Queries Faster                                     │          0 │
│ Queries Slower                                     │          7 │
│ Queries with No Change                             │         36 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	114.0s
Peak memory	40.2 GiB
Avg memory	31.1 GiB
CPU user	1075.5s
CPU sys	86.7s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	118.6s
Peak memory	36.0 GiB
Avg memory	26.2 GiB
CPU user	1101.1s
CPU sys	104.9s
Peak spill	0 B

---

File an issue against this benchmark runner

[Dandandan]

run benchmarks

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231749567-1121-6qj5c 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (8867a6e) to 16e578d (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231749567-1122-dl7cn 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (8867a6e) to 16e578d (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231749567-1120-lfmbd 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (8867a6e) to 16e578d (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃   optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.18 / 4.39 ±6.31 / 17.01 ms │          1.20 / 4.52 ±6.46 / 17.45 ms │     no change │
│ QQuery 1  │        14.08 / 14.48 ±0.24 / 14.82 ms │        14.41 / 14.67 ±0.15 / 14.84 ms │     no change │
│ QQuery 2  │        43.61 / 44.03 ±0.44 / 44.84 ms │        44.42 / 44.84 ±0.49 / 45.79 ms │     no change │
│ QQuery 3  │        40.78 / 44.86 ±3.18 / 48.66 ms │        43.13 / 45.42 ±1.47 / 47.09 ms │     no change │
│ QQuery 4  │     291.78 / 297.48 ±5.51 / 306.55 ms │    287.97 / 302.47 ±12.47 / 319.22 ms │     no change │
│ QQuery 5  │     344.25 / 348.77 ±5.07 / 356.03 ms │     339.77 / 349.61 ±8.84 / 364.11 ms │     no change │
│ QQuery 6  │           5.14 / 5.82 ±0.55 / 6.76 ms │         4.87 / 10.39 ±5.95 / 19.27 ms │  1.79x slower │
│ QQuery 7  │        16.62 / 17.28 ±0.83 / 18.88 ms │        16.65 / 17.12 ±0.50 / 18.04 ms │     no change │
│ QQuery 8  │     418.37 / 429.91 ±5.93 / 434.77 ms │    417.56 / 431.74 ±11.43 / 443.33 ms │     no change │
│ QQuery 9  │     655.53 / 661.25 ±6.85 / 674.11 ms │     658.39 / 667.76 ±6.63 / 675.25 ms │     no change │
│ QQuery 10 │       92.03 / 97.83 ±6.27 / 109.98 ms │       91.85 / 96.18 ±3.49 / 101.40 ms │     no change │
│ QQuery 11 │     104.11 / 105.71 ±1.35 / 107.35 ms │     105.67 / 106.71 ±0.65 / 107.66 ms │     no change │
│ QQuery 12 │     345.67 / 349.34 ±3.39 / 355.68 ms │     344.65 / 355.06 ±8.40 / 369.80 ms │     no change │
│ QQuery 13 │    461.03 / 477.13 ±13.67 / 497.35 ms │    448.81 / 474.41 ±13.97 / 488.54 ms │     no change │
│ QQuery 14 │     340.64 / 349.50 ±7.01 / 358.92 ms │     346.12 / 348.84 ±2.72 / 352.65 ms │     no change │
│ QQuery 15 │    354.33 / 372.89 ±17.33 / 400.78 ms │    360.86 / 378.19 ±30.75 / 439.62 ms │     no change │
│ QQuery 16 │    711.48 / 726.47 ±18.79 / 763.45 ms │    713.66 / 732.14 ±14.82 / 756.78 ms │     no change │
│ QQuery 17 │     708.75 / 717.34 ±6.47 / 725.49 ms │     695.79 / 704.99 ±7.05 / 717.41 ms │     no change │
│ QQuery 18 │  1367.93 / 1381.06 ±9.66 / 1397.18 ms │ 1402.50 / 1448.12 ±39.52 / 1498.26 ms │     no change │
│ QQuery 19 │        37.14 / 38.46 ±0.83 / 39.30 ms │      36.36 / 66.25 ±31.63 / 117.01 ms │  1.72x slower │
│ QQuery 20 │    711.87 / 725.05 ±13.21 / 743.85 ms │    718.35 / 741.84 ±31.65 / 804.08 ms │     no change │
│ QQuery 21 │     758.55 / 764.62 ±4.77 / 772.73 ms │     758.13 / 762.06 ±2.51 / 765.35 ms │     no change │
│ QQuery 22 │  1135.61 / 1141.76 ±7.72 / 1156.97 ms │  1127.76 / 1136.77 ±7.21 / 1148.91 ms │     no change │
│ QQuery 23 │ 3090.75 / 3102.32 ±12.80 / 3126.60 ms │ 3099.01 / 3125.71 ±20.46 / 3149.39 ms │     no change │
│ QQuery 24 │      99.42 / 101.93 ±1.84 / 105.13 ms │     100.56 / 104.66 ±3.26 / 108.33 ms │     no change │
│ QQuery 25 │     138.44 / 141.65 ±2.37 / 144.28 ms │     138.82 / 141.20 ±1.86 / 143.84 ms │     no change │
│ QQuery 26 │      99.67 / 102.36 ±1.81 / 104.15 ms │     102.90 / 105.35 ±1.52 / 107.55 ms │     no change │
│ QQuery 27 │     852.42 / 856.94 ±2.51 / 859.37 ms │     847.02 / 852.66 ±4.68 / 861.09 ms │     no change │
│ QQuery 28 │ 3281.97 / 3296.88 ±15.33 / 3321.57 ms │ 3258.52 / 3288.83 ±18.81 / 3315.28 ms │     no change │
│ QQuery 29 │        50.20 / 53.37 ±3.75 / 59.81 ms │        51.84 / 56.07 ±3.99 / 62.64 ms │  1.05x slower │
│ QQuery 30 │     362.82 / 368.36 ±6.37 / 380.67 ms │     362.97 / 369.92 ±4.94 / 376.80 ms │     no change │
│ QQuery 31 │     363.78 / 373.30 ±5.23 / 378.33 ms │    356.88 / 383.82 ±14.92 / 399.14 ms │     no change │
│ QQuery 32 │ 1031.10 / 1044.29 ±15.18 / 1072.82 ms │ 1044.21 / 1066.77 ±27.76 / 1120.80 ms │     no change │
│ QQuery 33 │  1462.92 / 1466.29 ±2.47 / 1470.32 ms │ 1484.60 / 1506.90 ±11.97 / 1517.23 ms │     no change │
│ QQuery 34 │ 1484.18 / 1503.08 ±11.74 / 1517.08 ms │ 1495.26 / 1515.74 ±15.26 / 1540.09 ms │     no change │
│ QQuery 35 │    393.06 / 404.13 ±15.77 / 434.77 ms │     387.92 / 396.44 ±6.54 / 404.58 ms │     no change │
│ QQuery 36 │     114.67 / 120.24 ±4.31 / 124.12 ms │     115.46 / 124.35 ±5.86 / 133.01 ms │     no change │
│ QQuery 37 │        49.61 / 50.73 ±0.92 / 52.23 ms │        47.97 / 49.98 ±1.34 / 51.90 ms │     no change │
│ QQuery 38 │        75.57 / 77.43 ±1.44 / 79.58 ms │        74.48 / 76.84 ±1.45 / 78.90 ms │     no change │
│ QQuery 39 │     215.12 / 223.63 ±7.73 / 235.76 ms │     210.01 / 219.33 ±6.13 / 226.31 ms │     no change │
│ QQuery 40 │        22.77 / 26.45 ±2.41 / 29.89 ms │        25.32 / 26.87 ±1.42 / 29.13 ms │     no change │
│ QQuery 41 │        20.42 / 22.85 ±2.50 / 27.51 ms │        19.54 / 21.52 ±1.92 / 24.90 ms │ +1.06x faster │
│ QQuery 42 │        19.87 / 20.86 ±1.34 / 23.49 ms │        19.87 / 20.36 ±0.26 / 20.58 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 22472.49ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 22693.39ms │
│ Average Time (HEAD)                                │   522.62ms │
│ Average Time (optimize-byteview-vectorized-append) │   527.75ms │
│ Queries Faster                                     │          1 │
│ Queries Slower                                     │          3 │
│ Queries with No Change                             │         39 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	113.4s
Peak memory	38.3 GiB
Avg memory	32.6 GiB
CPU user	1080.0s
CPU sys	78.6s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	114.7s
Peak memory	41.2 GiB
Avg memory	30.1 GiB
CPU user	1079.8s
CPU sys	86.4s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃      optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │              7.03 / 7.47 ±0.76 / 8.98 ms │              6.95 / 7.41 ±0.75 / 8.90 ms │     no change │
│ QQuery 2  │        150.48 / 151.14 ±0.61 / 152.26 ms │        147.17 / 147.53 ±0.47 / 148.44 ms │     no change │
│ QQuery 3  │        116.82 / 117.89 ±0.67 / 118.72 ms │        114.34 / 116.03 ±1.11 / 117.21 ms │     no change │
│ QQuery 4  │    1350.06 / 1428.41 ±62.36 / 1502.40 ms │    1365.52 / 1396.82 ±32.60 / 1449.46 ms │     no change │
│ QQuery 5  │        176.38 / 176.75 ±0.40 / 177.36 ms │        174.32 / 175.61 ±0.88 / 176.60 ms │     no change │
│ QQuery 6  │       826.51 / 850.85 ±18.20 / 873.73 ms │       827.23 / 845.66 ±18.53 / 874.95 ms │     no change │
│ QQuery 7  │        340.81 / 345.52 ±5.43 / 355.71 ms │        341.27 / 346.91 ±3.65 / 351.71 ms │     no change │
│ QQuery 8  │        115.88 / 117.31 ±1.17 / 119.23 ms │        116.34 / 117.88 ±1.12 / 119.42 ms │     no change │
│ QQuery 9  │        101.52 / 106.55 ±2.58 / 108.57 ms │        103.06 / 110.76 ±9.50 / 129.50 ms │     no change │
│ QQuery 10 │        105.44 / 107.66 ±1.15 / 108.56 ms │        106.78 / 108.13 ±0.69 / 108.70 ms │     no change │
│ QQuery 11 │        908.33 / 910.56 ±1.44 / 912.35 ms │     966.29 / 1048.62 ±41.45 / 1078.01 ms │  1.15x slower │
│ QQuery 12 │           44.81 / 46.33 ±1.01 / 47.91 ms │           44.73 / 46.60 ±1.41 / 48.56 ms │     no change │
│ QQuery 13 │        396.61 / 401.52 ±3.24 / 405.61 ms │        403.50 / 411.77 ±6.24 / 418.61 ms │     no change │
│ QQuery 14 │     1017.16 / 1029.94 ±8.75 / 1044.62 ms │    1010.93 / 1027.21 ±10.67 / 1037.45 ms │     no change │
│ QQuery 15 │           16.60 / 17.85 ±1.59 / 20.90 ms │           15.11 / 16.10 ±0.93 / 17.83 ms │ +1.11x faster │
│ QQuery 16 │              7.63 / 8.28 ±0.59 / 9.22 ms │              7.00 / 7.71 ±0.89 / 9.44 ms │ +1.07x faster │
│ QQuery 17 │        238.47 / 240.28 ±1.94 / 243.26 ms │        229.62 / 230.95 ±1.37 / 233.16 ms │     no change │
│ QQuery 18 │        127.80 / 130.84 ±1.55 / 132.05 ms │        125.59 / 127.85 ±1.29 / 129.32 ms │     no change │
│ QQuery 19 │        160.15 / 162.27 ±1.41 / 164.43 ms │        156.88 / 158.60 ±1.32 / 160.46 ms │     no change │
│ QQuery 20 │           14.71 / 15.16 ±0.53 / 16.18 ms │           13.32 / 13.92 ±0.47 / 14.76 ms │ +1.09x faster │
│ QQuery 21 │           20.83 / 21.76 ±0.77 / 23.09 ms │           19.13 / 19.94 ±0.53 / 20.57 ms │ +1.09x faster │
│ QQuery 22 │       485.13 / 502.06 ±14.06 / 518.14 ms │        485.77 / 493.73 ±4.47 / 498.60 ms │     no change │
│ QQuery 23 │       869.64 / 903.86 ±28.01 / 944.89 ms │        860.14 / 868.61 ±5.44 / 873.83 ms │     no change │
│ QQuery 24 │        397.44 / 400.88 ±3.82 / 408.11 ms │        380.57 / 384.80 ±2.76 / 388.18 ms │     no change │
│ QQuery 25 │        341.34 / 347.62 ±5.37 / 354.51 ms │        336.10 / 340.49 ±2.34 / 342.38 ms │     no change │
│ QQuery 26 │           82.78 / 83.61 ±0.58 / 84.40 ms │           79.60 / 83.16 ±2.24 / 86.50 ms │     no change │
│ QQuery 27 │              7.17 / 7.38 ±0.15 / 7.64 ms │              6.81 / 6.95 ±0.16 / 7.22 ms │ +1.06x faster │
│ QQuery 28 │        149.22 / 152.40 ±2.28 / 155.68 ms │        150.46 / 151.28 ±0.83 / 152.81 ms │     no change │
│ QQuery 29 │        281.52 / 282.89 ±1.13 / 284.63 ms │        280.94 / 283.95 ±1.69 / 285.60 ms │     no change │
│ QQuery 30 │           42.27 / 44.37 ±1.48 / 46.16 ms │           42.75 / 44.06 ±1.33 / 46.52 ms │     no change │
│ QQuery 31 │        168.67 / 172.62 ±2.16 / 174.86 ms │        169.66 / 173.60 ±2.99 / 177.93 ms │     no change │
│ QQuery 32 │           56.38 / 58.42 ±1.21 / 59.93 ms │           58.38 / 59.84 ±1.18 / 61.57 ms │     no change │
│ QQuery 33 │        141.28 / 142.00 ±0.84 / 143.59 ms │        144.03 / 146.06 ±2.06 / 149.97 ms │     no change │
│ QQuery 34 │              6.94 / 7.23 ±0.35 / 7.91 ms │              7.49 / 7.67 ±0.18 / 8.01 ms │  1.06x slower │
│ QQuery 35 │        106.75 / 109.84 ±1.77 / 111.83 ms │        111.51 / 113.70 ±2.26 / 116.78 ms │     no change │
│ QQuery 36 │              6.52 / 6.89 ±0.30 / 7.37 ms │              6.91 / 7.10 ±0.15 / 7.27 ms │     no change │
│ QQuery 37 │              8.39 / 8.78 ±0.33 / 9.22 ms │              8.76 / 9.22 ±0.26 / 9.50 ms │     no change │
│ QQuery 38 │           86.17 / 89.53 ±4.24 / 97.77 ms │          85.23 / 92.46 ±5.65 / 102.62 ms │     no change │
│ QQuery 39 │        124.98 / 129.38 ±3.07 / 133.81 ms │        134.01 / 135.32 ±1.67 / 138.55 ms │     no change │
│ QQuery 40 │        109.83 / 117.33 ±8.95 / 134.78 ms │        113.38 / 120.23 ±6.54 / 132.65 ms │     no change │
│ QQuery 41 │           14.13 / 14.67 ±0.44 / 15.29 ms │           15.25 / 15.99 ±0.57 / 16.71 ms │  1.09x slower │
│ QQuery 42 │        107.17 / 108.21 ±0.88 / 109.76 ms │        107.41 / 110.72 ±2.45 / 113.56 ms │     no change │
│ QQuery 43 │              5.94 / 6.40 ±0.56 / 7.44 ms │              5.85 / 6.04 ±0.22 / 6.48 ms │ +1.06x faster │
│ QQuery 44 │           11.74 / 11.92 ±0.12 / 12.08 ms │           11.31 / 11.51 ±0.17 / 11.71 ms │     no change │
│ QQuery 45 │           51.25 / 52.21 ±1.15 / 54.27 ms │           50.60 / 51.49 ±0.90 / 53.20 ms │     no change │
│ QQuery 46 │              8.25 / 8.57 ±0.23 / 8.90 ms │              8.90 / 9.15 ±0.13 / 9.25 ms │  1.07x slower │
│ QQuery 47 │        708.51 / 717.03 ±7.53 / 731.01 ms │       714.73 / 763.24 ±49.18 / 834.32 ms │  1.06x slower │
│ QQuery 48 │        292.11 / 296.35 ±3.29 / 300.55 ms │        285.84 / 292.41 ±4.08 / 297.06 ms │     no change │
│ QQuery 49 │        255.68 / 256.84 ±1.47 / 259.49 ms │        250.82 / 254.47 ±2.01 / 256.79 ms │     no change │
│ QQuery 50 │        233.64 / 238.68 ±3.12 / 242.85 ms │        234.52 / 238.03 ±3.04 / 243.40 ms │     no change │
│ QQuery 51 │        183.92 / 184.76 ±0.65 / 185.57 ms │        187.21 / 189.24 ±2.99 / 195.08 ms │     no change │
│ QQuery 52 │        109.35 / 110.36 ±0.82 / 111.80 ms │        110.39 / 112.70 ±1.28 / 114.08 ms │     no change │
│ QQuery 53 │        103.34 / 104.82 ±0.90 / 106.04 ms │        104.64 / 105.96 ±0.93 / 107.23 ms │     no change │
│ QQuery 54 │        145.94 / 147.59 ±1.69 / 150.28 ms │        152.18 / 153.15 ±0.72 / 154.30 ms │     no change │
│ QQuery 55 │        106.92 / 107.39 ±0.49 / 108.27 ms │        109.58 / 111.45 ±1.13 / 112.98 ms │     no change │
│ QQuery 56 │        140.63 / 142.92 ±1.66 / 145.43 ms │        141.58 / 143.43 ±1.44 / 145.72 ms │     no change │
│ QQuery 57 │        176.56 / 178.80 ±3.32 / 185.34 ms │        173.47 / 175.52 ±1.43 / 177.85 ms │     no change │
│ QQuery 58 │       293.17 / 304.50 ±11.52 / 325.44 ms │        291.75 / 296.23 ±3.87 / 302.01 ms │     no change │
│ QQuery 59 │        204.17 / 207.34 ±1.80 / 209.21 ms │        198.59 / 201.90 ±3.48 / 207.69 ms │     no change │
│ QQuery 60 │        148.25 / 149.18 ±0.68 / 150.03 ms │        144.04 / 145.12 ±1.02 / 146.79 ms │     no change │
│ QQuery 61 │           13.61 / 13.70 ±0.06 / 13.78 ms │           12.94 / 13.21 ±0.28 / 13.73 ms │     no change │
│ QQuery 62 │      873.28 / 943.50 ±56.02 / 1020.90 ms │      872.29 / 954.99 ±60.94 / 1047.97 ms │     no change │
│ QQuery 63 │        103.73 / 105.48 ±1.30 / 107.71 ms │        105.13 / 107.55 ±2.12 / 110.95 ms │     no change │
│ QQuery 64 │        685.90 / 688.80 ±1.93 / 691.58 ms │       686.72 / 702.05 ±13.31 / 719.36 ms │     no change │
│ QQuery 65 │        247.33 / 253.68 ±5.93 / 262.82 ms │        271.84 / 276.75 ±3.67 / 282.30 ms │  1.09x slower │
│ QQuery 66 │       239.85 / 256.01 ±13.20 / 272.40 ms │        246.83 / 256.17 ±9.15 / 273.16 ms │     no change │
│ QQuery 67 │        308.27 / 314.30 ±8.75 / 331.59 ms │       302.70 / 315.29 ±14.08 / 341.42 ms │     no change │
│ QQuery 68 │             8.37 / 9.77 ±1.44 / 12.30 ms │            8.89 / 10.57 ±1.49 / 13.11 ms │  1.08x slower │
│ QQuery 69 │        101.68 / 104.00 ±1.94 / 106.57 ms │        101.90 / 103.20 ±1.10 / 104.76 ms │     no change │
│ QQuery 70 │       345.55 / 360.42 ±15.47 / 390.25 ms │        344.64 / 355.82 ±7.05 / 363.49 ms │     no change │
│ QQuery 71 │        133.13 / 137.66 ±2.43 / 140.31 ms │        134.57 / 137.71 ±2.79 / 141.59 ms │     no change │
│ QQuery 72 │       618.32 / 630.87 ±11.87 / 651.02 ms │       615.95 / 647.11 ±16.26 / 660.06 ms │     no change │
│ QQuery 73 │              7.25 / 7.94 ±0.74 / 9.25 ms │              7.76 / 8.88 ±0.91 / 9.89 ms │  1.12x slower │
│ QQuery 74 │        666.42 / 677.67 ±6.71 / 686.84 ms │       599.12 / 639.24 ±32.84 / 674.87 ms │ +1.06x faster │
│ QQuery 75 │        276.07 / 278.26 ±2.16 / 281.98 ms │        276.90 / 279.16 ±1.70 / 281.61 ms │     no change │
│ QQuery 76 │        133.71 / 134.77 ±0.86 / 136.33 ms │        133.40 / 134.50 ±0.91 / 135.76 ms │     no change │
│ QQuery 77 │        187.52 / 192.11 ±2.59 / 194.39 ms │        187.35 / 189.34 ±1.63 / 191.50 ms │     no change │
│ QQuery 78 │        336.93 / 342.62 ±5.50 / 353.01 ms │        338.17 / 341.97 ±2.55 / 345.11 ms │     no change │
│ QQuery 79 │        229.41 / 235.73 ±3.80 / 240.73 ms │        230.34 / 233.01 ±2.62 / 237.62 ms │     no change │
│ QQuery 80 │        321.11 / 324.35 ±1.85 / 326.82 ms │        320.79 / 323.58 ±2.21 / 326.33 ms │     no change │
│ QQuery 81 │           26.10 / 26.99 ±0.72 / 28.04 ms │           25.49 / 26.70 ±1.05 / 28.37 ms │     no change │
│ QQuery 82 │        202.51 / 205.69 ±2.02 / 207.78 ms │        200.03 / 202.69 ±2.15 / 206.40 ms │     no change │
│ QQuery 83 │           40.56 / 41.75 ±0.83 / 42.68 ms │           38.05 / 39.35 ±0.99 / 40.66 ms │ +1.06x faster │
│ QQuery 84 │           49.55 / 49.91 ±0.31 / 50.47 ms │           48.59 / 49.47 ±0.75 / 50.29 ms │     no change │
│ QQuery 85 │        151.59 / 153.62 ±1.35 / 155.03 ms │        148.73 / 150.53 ±2.00 / 153.88 ms │     no change │
│ QQuery 86 │           39.86 / 41.87 ±2.02 / 45.58 ms │           38.94 / 40.73 ±1.26 / 42.40 ms │     no change │
│ QQuery 87 │          91.34 / 95.14 ±3.80 / 102.04 ms │           86.39 / 89.37 ±2.61 / 94.25 ms │ +1.06x faster │
│ QQuery 88 │        104.75 / 105.21 ±0.55 / 106.04 ms │        100.50 / 102.17 ±1.12 / 103.78 ms │     no change │
│ QQuery 89 │        122.91 / 124.48 ±0.99 / 125.67 ms │        117.82 / 119.22 ±0.98 / 120.49 ms │     no change │
│ QQuery 90 │           24.47 / 25.09 ±0.64 / 26.28 ms │           22.32 / 23.70 ±0.74 / 24.58 ms │ +1.06x faster │
│ QQuery 91 │           64.13 / 65.34 ±0.96 / 66.81 ms │           62.80 / 63.76 ±0.90 / 65.27 ms │     no change │
│ QQuery 92 │           58.39 / 59.96 ±1.08 / 61.26 ms │           58.09 / 58.54 ±0.38 / 59.25 ms │     no change │
│ QQuery 93 │        188.06 / 191.07 ±3.44 / 197.28 ms │        187.96 / 190.00 ±1.54 / 192.44 ms │     no change │
│ QQuery 94 │           62.90 / 64.56 ±0.96 / 65.38 ms │           62.41 / 64.12 ±1.07 / 65.29 ms │     no change │
│ QQuery 95 │        131.64 / 133.36 ±0.92 / 134.17 ms │        128.80 / 132.01 ±2.57 / 136.26 ms │     no change │
│ QQuery 96 │           76.48 / 77.37 ±0.91 / 78.76 ms │           74.32 / 75.85 ±1.02 / 77.16 ms │     no change │
│ QQuery 97 │        126.34 / 129.40 ±1.99 / 131.61 ms │        127.27 / 129.85 ±1.38 / 131.08 ms │     no change │
│ QQuery 98 │        153.46 / 155.87 ±1.51 / 157.60 ms │        155.30 / 158.28 ±2.99 / 163.20 ms │     no change │
│ QQuery 99 │ 10808.14 / 10867.98 ±70.48 / 11005.97 ms │ 10813.39 / 10876.96 ±46.26 / 10936.52 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 31936.31ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 32015.34ms │
│ Average Time (HEAD)                                │   322.59ms │
│ Average Time (optimize-byteview-vectorized-append) │   323.39ms │
│ Queries Faster                                     │         10 │
│ Queries Slower                                     │          8 │
│ Queries with No Change                             │         81 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	160.1s
Peak memory	5.1 GiB
Avg memory	4.1 GiB
CPU user	264.1s
CPU sys	17.7s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	160.4s
Peak memory	4.9 GiB
Avg memory	4.0 GiB
CPU user	263.5s
CPU sys	18.0s
Peak spill	0 B

---

File an issue against this benchmark runner

[Dandandan]

run benchmarks

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231782112-1124-p7j68 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (faeabd2) to 16e578d (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231782112-1123-7fxgs 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (faeabd2) to 16e578d (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231782112-1125-qxnsj 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (faeabd2) to 16e578d (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃      optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │              6.76 / 7.32 ±0.74 / 8.77 ms │              6.59 / 7.04 ±0.82 / 8.68 ms │     no change │
│ QQuery 2  │        147.29 / 148.48 ±1.23 / 150.34 ms │        144.98 / 145.99 ±0.96 / 147.73 ms │     no change │
│ QQuery 3  │        115.24 / 115.73 ±0.60 / 116.90 ms │        113.72 / 114.33 ±0.54 / 115.31 ms │     no change │
│ QQuery 4  │    1395.47 / 1442.10 ±26.04 / 1472.31 ms │    1284.76 / 1333.38 ±59.41 / 1447.11 ms │ +1.08x faster │
│ QQuery 5  │        172.00 / 174.08 ±1.57 / 176.83 ms │        175.12 / 176.54 ±1.09 / 178.05 ms │     no change │
│ QQuery 6  │       844.46 / 861.52 ±10.94 / 878.58 ms │       824.16 / 866.24 ±26.70 / 898.72 ms │     no change │
│ QQuery 7  │        341.41 / 345.23 ±4.09 / 352.69 ms │        343.80 / 347.11 ±2.46 / 350.08 ms │     no change │
│ QQuery 8  │        115.46 / 117.74 ±1.24 / 118.83 ms │        115.55 / 117.25 ±1.15 / 118.51 ms │     no change │
│ QQuery 9  │        102.02 / 110.52 ±9.24 / 128.51 ms │        99.96 / 110.63 ±11.17 / 132.29 ms │     no change │
│ QQuery 10 │        107.14 / 108.25 ±0.61 / 108.91 ms │        105.79 / 107.09 ±0.84 / 108.14 ms │     no change │
│ QQuery 11 │     938.84 / 1009.73 ±52.09 / 1067.84 ms │      912.22 / 977.63 ±70.86 / 1067.21 ms │     no change │
│ QQuery 12 │           48.09 / 49.30 ±0.79 / 50.23 ms │           48.04 / 49.15 ±0.76 / 50.39 ms │     no change │
│ QQuery 13 │        402.93 / 412.32 ±4.88 / 417.06 ms │        420.74 / 424.10 ±3.41 / 430.16 ms │     no change │
│ QQuery 14 │    1012.84 / 1028.90 ±13.16 / 1045.47 ms │    1015.95 / 1032.15 ±11.64 / 1045.56 ms │     no change │
│ QQuery 15 │           17.24 / 18.14 ±0.49 / 18.68 ms │           15.82 / 17.26 ±0.85 / 18.42 ms │     no change │
│ QQuery 16 │             7.64 / 8.59 ±1.05 / 10.53 ms │              7.07 / 7.53 ±0.61 / 8.68 ms │ +1.14x faster │
│ QQuery 17 │        236.09 / 238.75 ±1.45 / 240.20 ms │        229.00 / 230.86 ±1.73 / 233.48 ms │     no change │
│ QQuery 18 │        130.19 / 131.48 ±0.82 / 132.49 ms │        128.45 / 129.45 ±0.80 / 130.55 ms │     no change │
│ QQuery 19 │        156.98 / 159.17 ±2.12 / 162.47 ms │        155.85 / 157.66 ±1.83 / 160.51 ms │     no change │
│ QQuery 20 │           13.28 / 14.34 ±0.58 / 14.85 ms │           13.23 / 13.89 ±0.44 / 14.41 ms │     no change │
│ QQuery 21 │           19.05 / 19.86 ±0.47 / 20.52 ms │           19.15 / 19.60 ±0.36 / 20.07 ms │     no change │
│ QQuery 22 │        483.65 / 488.31 ±4.43 / 496.65 ms │        483.60 / 490.68 ±5.15 / 498.95 ms │     no change │
│ QQuery 23 │        859.51 / 868.28 ±6.93 / 875.25 ms │       861.55 / 893.24 ±30.62 / 945.74 ms │     no change │
│ QQuery 24 │        375.93 / 380.17 ±2.50 / 383.60 ms │        398.09 / 402.79 ±4.04 / 410.01 ms │  1.06x slower │
│ QQuery 25 │        340.69 / 342.35 ±1.56 / 344.79 ms │        341.98 / 348.64 ±6.19 / 356.86 ms │     no change │
│ QQuery 26 │           80.72 / 81.80 ±0.70 / 82.67 ms │           81.61 / 84.35 ±1.47 / 85.91 ms │     no change │
│ QQuery 27 │              6.80 / 7.12 ±0.25 / 7.45 ms │              6.86 / 7.41 ±0.38 / 8.01 ms │     no change │
│ QQuery 28 │        148.81 / 149.35 ±0.67 / 150.62 ms │        147.56 / 150.31 ±1.65 / 152.41 ms │     no change │
│ QQuery 29 │        280.28 / 281.45 ±1.20 / 283.54 ms │        281.96 / 283.04 ±1.12 / 284.53 ms │     no change │
│ QQuery 30 │           44.86 / 45.97 ±1.15 / 48.16 ms │           45.20 / 46.72 ±1.47 / 49.26 ms │     no change │
│ QQuery 31 │        170.89 / 173.85 ±1.83 / 176.23 ms │        170.20 / 174.62 ±2.65 / 178.23 ms │     no change │
│ QQuery 32 │           57.57 / 59.05 ±1.74 / 62.38 ms │           58.27 / 59.87 ±0.90 / 60.63 ms │     no change │
│ QQuery 33 │        141.17 / 142.70 ±0.84 / 143.45 ms │        143.38 / 144.18 ±0.57 / 145.12 ms │     no change │
│ QQuery 34 │              7.20 / 7.42 ±0.19 / 7.77 ms │              7.55 / 7.70 ±0.10 / 7.84 ms │     no change │
│ QQuery 35 │        107.39 / 109.19 ±1.21 / 110.65 ms │        110.09 / 111.29 ±0.91 / 112.39 ms │     no change │
│ QQuery 36 │              6.81 / 7.01 ±0.18 / 7.32 ms │              6.82 / 6.90 ±0.05 / 6.95 ms │     no change │
│ QQuery 37 │             8.45 / 9.14 ±0.81 / 10.68 ms │             9.06 / 9.81 ±0.77 / 11.21 ms │  1.07x slower │
│ QQuery 38 │           84.69 / 88.85 ±4.56 / 97.61 ms │          88.23 / 92.41 ±5.08 / 102.07 ms │     no change │
│ QQuery 39 │        127.32 / 138.16 ±6.07 / 143.90 ms │        129.86 / 133.54 ±3.41 / 139.23 ms │     no change │
│ QQuery 40 │        118.08 / 122.85 ±5.75 / 133.31 ms │        116.73 / 120.67 ±6.50 / 133.58 ms │     no change │
│ QQuery 41 │           15.30 / 16.44 ±0.76 / 17.47 ms │           14.59 / 15.97 ±1.10 / 17.61 ms │     no change │
│ QQuery 42 │        110.65 / 112.32 ±1.15 / 114.08 ms │        110.35 / 111.92 ±1.23 / 113.63 ms │     no change │
│ QQuery 43 │              6.50 / 6.60 ±0.10 / 6.72 ms │              6.30 / 6.41 ±0.10 / 6.57 ms │     no change │
│ QQuery 44 │           12.78 / 13.52 ±0.71 / 14.49 ms │           11.97 / 12.38 ±0.21 / 12.55 ms │ +1.09x faster │
│ QQuery 45 │           52.11 / 52.91 ±0.89 / 54.62 ms │           50.61 / 51.92 ±2.15 / 56.20 ms │     no change │
│ QQuery 46 │              8.84 / 9.18 ±0.28 / 9.69 ms │              8.38 / 8.57 ±0.15 / 8.79 ms │ +1.07x faster │
│ QQuery 47 │       713.16 / 786.11 ±49.39 / 850.42 ms │        692.27 / 697.79 ±3.31 / 701.04 ms │ +1.13x faster │
│ QQuery 48 │        285.47 / 290.88 ±4.41 / 298.18 ms │        287.07 / 293.46 ±3.26 / 296.26 ms │     no change │
│ QQuery 49 │        249.48 / 251.72 ±1.90 / 254.12 ms │        250.95 / 253.58 ±2.27 / 256.41 ms │     no change │
│ QQuery 50 │        218.15 / 223.07 ±2.68 / 225.77 ms │        221.27 / 223.73 ±2.03 / 227.44 ms │     no change │
│ QQuery 51 │        181.47 / 183.61 ±2.30 / 187.96 ms │        177.19 / 180.38 ±2.11 / 182.72 ms │     no change │
│ QQuery 52 │        106.31 / 109.83 ±2.09 / 112.30 ms │        106.77 / 107.81 ±0.99 / 109.12 ms │     no change │
│ QQuery 53 │        103.81 / 105.49 ±1.31 / 107.15 ms │        102.97 / 103.78 ±0.80 / 105.25 ms │     no change │
│ QQuery 54 │        150.68 / 153.47 ±2.02 / 156.54 ms │        145.23 / 146.97 ±1.36 / 148.83 ms │     no change │
│ QQuery 55 │        107.95 / 108.90 ±0.51 / 109.45 ms │        106.54 / 108.02 ±1.18 / 109.43 ms │     no change │
│ QQuery 56 │        145.17 / 146.78 ±1.17 / 148.15 ms │        139.98 / 140.95 ±0.89 / 142.55 ms │     no change │
│ QQuery 57 │        180.90 / 182.69 ±1.01 / 183.93 ms │        170.79 / 172.35 ±1.85 / 175.94 ms │ +1.06x faster │
│ QQuery 58 │        292.88 / 304.12 ±8.91 / 317.91 ms │        288.38 / 295.38 ±6.49 / 305.41 ms │     no change │
│ QQuery 59 │        203.82 / 207.79 ±2.05 / 209.63 ms │        197.41 / 200.24 ±2.40 / 203.79 ms │     no change │
│ QQuery 60 │        141.44 / 144.89 ±2.28 / 147.85 ms │        143.46 / 145.03 ±1.55 / 147.73 ms │     no change │
│ QQuery 61 │           12.88 / 13.36 ±0.39 / 13.91 ms │           12.87 / 13.31 ±0.36 / 13.82 ms │     no change │
│ QQuery 62 │       902.08 / 927.25 ±22.98 / 970.00 ms │      923.46 / 987.22 ±65.99 / 1086.52 ms │  1.06x slower │
│ QQuery 63 │        104.89 / 106.46 ±1.37 / 108.53 ms │        103.33 / 107.81 ±3.57 / 114.27 ms │     no change │
│ QQuery 64 │        685.37 / 688.19 ±2.41 / 692.33 ms │       682.27 / 695.03 ±13.78 / 719.28 ms │     no change │
│ QQuery 65 │        251.06 / 254.19 ±2.19 / 256.99 ms │        249.74 / 253.19 ±2.45 / 256.41 ms │     no change │
│ QQuery 66 │       245.53 / 258.19 ±12.51 / 278.83 ms │        244.28 / 257.01 ±8.67 / 271.10 ms │     no change │
│ QQuery 67 │        323.42 / 326.10 ±3.09 / 331.81 ms │        347.16 / 351.61 ±4.57 / 359.11 ms │  1.08x slower │
│ QQuery 68 │            9.51 / 12.98 ±2.85 / 16.73 ms │            9.42 / 11.19 ±1.18 / 12.90 ms │ +1.16x faster │
│ QQuery 69 │        104.25 / 105.88 ±1.28 / 107.91 ms │        105.28 / 107.74 ±1.70 / 109.62 ms │     no change │
│ QQuery 70 │       346.25 / 357.70 ±11.20 / 379.15 ms │       353.44 / 373.14 ±14.80 / 395.24 ms │     no change │
│ QQuery 71 │        137.18 / 139.71 ±2.05 / 142.36 ms │        137.88 / 139.20 ±1.21 / 141.28 ms │     no change │
│ QQuery 72 │       602.47 / 622.37 ±14.34 / 645.70 ms │       617.63 / 637.33 ±15.90 / 662.41 ms │     no change │
│ QQuery 73 │              6.50 / 7.80 ±1.00 / 8.96 ms │             7.12 / 8.28 ±1.13 / 10.17 ms │  1.06x slower │
│ QQuery 74 │       584.22 / 629.52 ±39.30 / 675.56 ms │        586.00 / 596.38 ±7.65 / 609.82 ms │ +1.06x faster │
│ QQuery 75 │        283.68 / 284.55 ±1.00 / 286.05 ms │        276.81 / 278.36 ±1.15 / 279.75 ms │     no change │
│ QQuery 76 │        137.69 / 138.88 ±0.97 / 140.45 ms │        133.86 / 134.37 ±0.63 / 135.59 ms │     no change │
│ QQuery 77 │        190.83 / 193.69 ±1.70 / 195.60 ms │        188.25 / 191.31 ±2.22 / 194.76 ms │     no change │
│ QQuery 78 │        333.78 / 342.97 ±7.91 / 357.52 ms │        340.31 / 342.97 ±1.87 / 346.11 ms │     no change │
│ QQuery 79 │        232.24 / 235.69 ±2.33 / 238.35 ms │        230.83 / 234.80 ±3.04 / 239.05 ms │     no change │
│ QQuery 80 │        323.34 / 326.96 ±3.06 / 332.62 ms │        323.16 / 324.71 ±1.56 / 326.98 ms │     no change │
│ QQuery 81 │           26.14 / 27.08 ±1.40 / 29.84 ms │           26.45 / 27.32 ±0.77 / 28.58 ms │     no change │
│ QQuery 82 │        195.81 / 197.97 ±1.94 / 201.25 ms │        199.82 / 204.53 ±2.53 / 207.24 ms │     no change │
│ QQuery 83 │           38.33 / 40.38 ±1.96 / 43.71 ms │           39.08 / 40.47 ±1.50 / 42.66 ms │     no change │
│ QQuery 84 │           48.85 / 50.06 ±1.13 / 51.90 ms │           47.86 / 49.76 ±1.53 / 52.26 ms │     no change │
│ QQuery 85 │        147.57 / 149.19 ±1.08 / 150.35 ms │        149.18 / 150.98 ±1.11 / 152.35 ms │     no change │
│ QQuery 86 │           38.87 / 40.15 ±0.75 / 41.10 ms │           40.06 / 41.00 ±0.77 / 41.80 ms │     no change │
│ QQuery 87 │           84.43 / 87.54 ±3.27 / 93.75 ms │           90.47 / 92.95 ±3.33 / 99.52 ms │  1.06x slower │
│ QQuery 88 │         99.92 / 101.22 ±0.98 / 102.86 ms │        102.43 / 103.77 ±0.94 / 104.86 ms │     no change │
│ QQuery 89 │        119.60 / 120.77 ±0.80 / 121.98 ms │        120.76 / 122.60 ±1.04 / 123.85 ms │     no change │
│ QQuery 90 │           23.68 / 24.68 ±1.53 / 27.67 ms │           23.74 / 24.08 ±0.31 / 24.56 ms │     no change │
│ QQuery 91 │           63.36 / 64.48 ±0.65 / 65.23 ms │           65.11 / 66.80 ±1.53 / 68.65 ms │     no change │
│ QQuery 92 │           57.01 / 58.88 ±1.46 / 61.25 ms │           58.59 / 59.54 ±0.57 / 60.37 ms │     no change │
│ QQuery 93 │        185.57 / 187.21 ±1.43 / 188.95 ms │        190.02 / 193.25 ±1.77 / 194.88 ms │     no change │
│ QQuery 94 │           61.34 / 62.60 ±1.00 / 63.77 ms │           62.57 / 62.99 ±0.49 / 63.84 ms │     no change │
│ QQuery 95 │        127.16 / 128.24 ±0.98 / 129.95 ms │        128.41 / 130.67 ±1.66 / 133.28 ms │     no change │
│ QQuery 96 │           70.58 / 74.21 ±2.21 / 76.96 ms │           74.81 / 75.51 ±0.57 / 76.56 ms │     no change │
│ QQuery 97 │        125.94 / 127.37 ±1.27 / 129.75 ms │        126.66 / 127.58 ±1.02 / 129.34 ms │     no change │
│ QQuery 98 │        153.18 / 154.98 ±1.63 / 157.47 ms │        150.31 / 154.71 ±3.03 / 157.80 ms │     no change │
│ QQuery 99 │ 10807.81 / 10864.66 ±47.88 / 10925.39 ms │ 10844.88 / 10890.14 ±32.08 / 10924.97 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 31969.04ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 31893.33ms │
│ Average Time (HEAD)                                │   322.92ms │
│ Average Time (optimize-byteview-vectorized-append) │   322.15ms │
│ Queries Faster                                     │          8 │
│ Queries Slower                                     │          6 │
│ Queries with No Change                             │         85 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	160.2s
Peak memory	5.3 GiB
Avg memory	4.4 GiB
CPU user	262.9s
CPU sys	17.6s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	159.8s
Peak memory	4.7 GiB
Avg memory	4.1 GiB
CPU user	263.8s
CPU sys	18.0s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃   optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.21 / 4.53 ±6.46 / 17.46 ms │          1.22 / 4.60 ±6.53 / 17.66 ms │     no change │
│ QQuery 1  │        14.56 / 14.90 ±0.32 / 15.47 ms │        14.40 / 14.84 ±0.25 / 15.14 ms │     no change │
│ QQuery 2  │        43.54 / 43.81 ±0.29 / 44.32 ms │        44.16 / 44.71 ±0.33 / 45.12 ms │     no change │
│ QQuery 3  │        41.68 / 45.13 ±2.98 / 49.12 ms │        44.97 / 46.48 ±0.92 / 47.85 ms │     no change │
│ QQuery 4  │     295.44 / 299.59 ±3.11 / 303.29 ms │     298.75 / 305.50 ±6.38 / 315.39 ms │     no change │
│ QQuery 5  │     349.09 / 350.24 ±1.66 / 353.49 ms │     351.11 / 357.39 ±6.69 / 369.88 ms │     no change │
│ QQuery 6  │           4.89 / 6.34 ±0.84 / 7.36 ms │           5.42 / 6.11 ±0.73 / 7.31 ms │     no change │
│ QQuery 7  │        16.82 / 16.96 ±0.15 / 17.19 ms │        16.96 / 17.39 ±0.38 / 17.99 ms │     no change │
│ QQuery 8  │     424.09 / 432.48 ±5.82 / 440.52 ms │     428.63 / 437.70 ±6.08 / 446.77 ms │     no change │
│ QQuery 9  │     660.88 / 667.10 ±4.42 / 673.23 ms │    650.95 / 675.60 ±14.31 / 691.84 ms │     no change │
│ QQuery 10 │        92.20 / 94.68 ±2.62 / 99.75 ms │       95.54 / 97.80 ±3.15 / 103.65 ms │     no change │
│ QQuery 11 │     105.94 / 108.13 ±2.00 / 111.55 ms │     110.00 / 110.79 ±0.53 / 111.37 ms │     no change │
│ QQuery 12 │     343.81 / 349.20 ±3.49 / 353.34 ms │    341.76 / 355.93 ±12.78 / 377.58 ms │     no change │
│ QQuery 13 │    458.85 / 472.84 ±14.24 / 498.15 ms │    521.57 / 544.59 ±19.02 / 573.37 ms │  1.15x slower │
│ QQuery 14 │     343.68 / 347.35 ±2.09 / 349.43 ms │     764.46 / 770.46 ±4.28 / 775.40 ms │  2.22x slower │
│ QQuery 15 │    356.27 / 368.77 ±13.34 / 392.50 ms │    360.60 / 380.64 ±23.48 / 422.71 ms │     no change │
│ QQuery 16 │     710.85 / 725.69 ±9.85 / 740.54 ms │ 2530.75 / 2594.58 ±38.10 / 2644.69 ms │  3.58x slower │
│ QQuery 17 │     715.85 / 725.40 ±6.61 / 733.74 ms │ 2536.43 / 2552.08 ±13.46 / 2575.14 ms │  3.52x slower │
│ QQuery 18 │ 1440.55 / 1499.90 ±75.78 / 1645.31 ms │ 3224.35 / 3266.43 ±31.55 / 3305.57 ms │  2.18x slower │
│ QQuery 19 │      37.28 / 57.70 ±38.31 / 134.25 ms │        34.90 / 36.57 ±1.62 / 39.53 ms │ +1.58x faster │
│ QQuery 20 │    727.83 / 755.12 ±33.80 / 811.28 ms │    712.78 / 727.80 ±16.41 / 759.77 ms │     no change │
│ QQuery 21 │     769.60 / 781.41 ±9.12 / 797.43 ms │     765.22 / 771.56 ±4.06 / 775.82 ms │     no change │
│ QQuery 22 │  1160.26 / 1168.30 ±4.30 / 1172.54 ms │  1142.30 / 1151.76 ±6.26 / 1158.32 ms │     no change │
│ QQuery 23 │ 3124.29 / 3153.44 ±22.59 / 3191.50 ms │ 3108.80 / 3143.50 ±23.70 / 3177.55 ms │     no change │
│ QQuery 24 │     101.09 / 103.32 ±1.75 / 106.24 ms │     100.03 / 102.91 ±1.85 / 105.42 ms │     no change │
│ QQuery 25 │     142.40 / 143.86 ±1.17 / 145.92 ms │     142.39 / 143.42 ±0.91 / 144.75 ms │     no change │
│ QQuery 26 │      99.94 / 102.36 ±1.81 / 104.71 ms │     100.38 / 103.26 ±2.23 / 105.38 ms │     no change │
│ QQuery 27 │     861.19 / 869.00 ±6.35 / 880.38 ms │     853.57 / 864.81 ±7.69 / 876.96 ms │     no change │
│ QQuery 28 │ 3287.78 / 3323.93 ±25.13 / 3351.25 ms │ 3288.03 / 3311.93 ±12.91 / 3327.07 ms │     no change │
│ QQuery 29 │        50.59 / 54.47 ±5.10 / 64.16 ms │        51.67 / 54.48 ±2.27 / 57.33 ms │     no change │
│ QQuery 30 │     365.74 / 372.09 ±7.58 / 386.82 ms │     367.82 / 373.55 ±5.66 / 383.37 ms │     no change │
│ QQuery 31 │    373.67 / 387.48 ±10.52 / 403.67 ms │    364.25 / 379.85 ±13.00 / 403.39 ms │     no change │
│ QQuery 32 │ 1205.63 / 1254.30 ±34.13 / 1296.04 ms │ 1038.58 / 1059.52 ±18.14 / 1091.32 ms │ +1.18x faster │
│ QQuery 33 │ 1526.82 / 1565.19 ±29.76 / 1612.13 ms │ 1474.07 / 1500.88 ±21.76 / 1534.60 ms │     no change │
│ QQuery 34 │ 1518.36 / 1549.81 ±42.29 / 1632.11 ms │ 1473.21 / 1516.68 ±32.13 / 1555.34 ms │     no change │
│ QQuery 35 │    395.92 / 418.31 ±27.35 / 470.07 ms │     403.07 / 408.71 ±5.11 / 416.05 ms │     no change │
│ QQuery 36 │     118.69 / 124.38 ±3.26 / 127.57 ms │     117.69 / 122.81 ±3.20 / 127.75 ms │     no change │
│ QQuery 37 │        49.79 / 51.97 ±2.15 / 55.56 ms │        48.00 / 50.80 ±1.67 / 52.94 ms │     no change │
│ QQuery 38 │        76.08 / 77.53 ±1.29 / 79.09 ms │        77.02 / 77.76 ±0.70 / 79.09 ms │     no change │
│ QQuery 39 │     204.72 / 215.80 ±5.81 / 220.58 ms │     361.43 / 373.29 ±6.31 / 379.61 ms │  1.73x slower │
│ QQuery 40 │        23.81 / 25.95 ±1.84 / 28.13 ms │        22.86 / 25.65 ±2.49 / 29.82 ms │     no change │
│ QQuery 41 │        20.52 / 22.04 ±1.25 / 24.13 ms │        19.44 / 20.95 ±0.95 / 21.96 ms │     no change │
│ QQuery 42 │        20.07 / 20.54 ±0.28 / 20.84 ms │        19.93 / 21.30 ±1.00 / 22.76 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 23171.33ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 28927.37ms │
│ Average Time (HEAD)                                │   538.87ms │
│ Average Time (optimize-byteview-vectorized-append) │   672.73ms │
│ Queries Faster                                     │          2 │
│ Queries Slower                                     │          6 │
│ Queries with No Change                             │         35 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	116.8s
Peak memory	37.8 GiB
Avg memory	27.9 GiB
CPU user	1091.3s
CPU sys	95.9s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	145.9s
Peak memory	40.5 GiB
Avg memory	30.8 GiB
CPU user	1319.3s
CPU sys	111.7s
Peak spill	0 B

---

File an issue against this benchmark runner

[Dandandan]

run benchmarks

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231862016-1128-cvq9f 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (9fc0180) to 16e578d (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231862016-1127-ql96w 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (9fc0180) to 16e578d (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231862016-1126-k4ncj 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (9fc0180) to 16e578d (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[Dandandan]

run benchmark clickbench_partitioned clickbench_extended

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231879548-1129-4qnxm 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (c6044ad) to 16e578d (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4231879548-1130-xcbz2 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing optimize-byteview-vectorized-append (c6044ad) to 16e578d (merge-base) diff using: clickbench_extended
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃      optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │              6.54 / 7.08 ±0.79 / 8.64 ms │              6.76 / 7.21 ±0.81 / 8.83 ms │     no change │
│ QQuery 2  │        145.59 / 146.18 ±0.35 / 146.64 ms │        144.69 / 146.26 ±1.03 / 147.43 ms │     no change │
│ QQuery 3  │        113.75 / 114.41 ±0.46 / 115.01 ms │        112.94 / 114.25 ±1.05 / 115.44 ms │     no change │
│ QQuery 4  │    1316.43 / 1343.09 ±16.99 / 1358.85 ms │    1301.86 / 1368.96 ±47.33 / 1429.33 ms │     no change │
│ QQuery 5  │        172.74 / 174.45 ±1.40 / 176.75 ms │        176.44 / 178.33 ±1.56 / 181.09 ms │     no change │
│ QQuery 6  │       807.78 / 841.40 ±17.73 / 859.83 ms │       892.69 / 917.62 ±17.57 / 942.60 ms │  1.09x slower │
│ QQuery 7  │        338.39 / 343.11 ±2.40 / 345.01 ms │        344.12 / 346.14 ±1.50 / 348.23 ms │     no change │
│ QQuery 8  │        115.57 / 116.76 ±0.74 / 117.57 ms │        116.14 / 117.57 ±1.41 / 119.36 ms │     no change │
│ QQuery 9  │        101.16 / 104.69 ±2.95 / 108.37 ms │        100.49 / 102.35 ±2.13 / 106.38 ms │     no change │
│ QQuery 10 │        105.55 / 106.74 ±0.77 / 107.57 ms │        106.30 / 107.62 ±0.98 / 109.31 ms │     no change │
│ QQuery 11 │        926.17 / 934.86 ±7.66 / 944.75 ms │       940.71 / 957.90 ±10.84 / 966.86 ms │     no change │
│ QQuery 12 │           45.28 / 46.42 ±1.07 / 48.37 ms │           45.12 / 47.31 ±1.29 / 49.16 ms │     no change │
│ QQuery 13 │        397.60 / 403.86 ±3.29 / 406.88 ms │        400.74 / 403.36 ±2.09 / 406.50 ms │     no change │
│ QQuery 14 │     990.56 / 1009.77 ±10.78 / 1021.06 ms │      998.74 / 1006.10 ±5.48 / 1014.79 ms │     no change │
│ QQuery 15 │           15.43 / 17.31 ±1.49 / 19.97 ms │           15.23 / 16.45 ±0.72 / 17.34 ms │     no change │
│ QQuery 16 │              7.34 / 8.17 ±0.76 / 9.44 ms │             7.10 / 8.35 ±1.13 / 10.25 ms │     no change │
│ QQuery 17 │        228.32 / 229.32 ±0.60 / 230.21 ms │        228.82 / 231.54 ±2.42 / 235.04 ms │     no change │
│ QQuery 18 │        126.94 / 127.96 ±0.89 / 129.23 ms │        127.19 / 129.17 ±1.25 / 130.40 ms │     no change │
│ QQuery 19 │        155.06 / 155.79 ±0.73 / 157.18 ms │        155.01 / 156.08 ±0.81 / 157.10 ms │     no change │
│ QQuery 20 │           13.82 / 14.73 ±0.91 / 16.43 ms │           13.78 / 14.21 ±0.45 / 14.96 ms │     no change │
│ QQuery 21 │           19.63 / 19.92 ±0.23 / 20.23 ms │           19.19 / 19.59 ±0.28 / 20.04 ms │     no change │
│ QQuery 22 │        486.06 / 488.46 ±2.12 / 492.11 ms │        479.93 / 490.50 ±5.99 / 495.77 ms │     no change │
│ QQuery 23 │        870.06 / 884.07 ±8.38 / 892.26 ms │        861.91 / 867.36 ±4.57 / 875.53 ms │     no change │
│ QQuery 24 │        380.29 / 380.69 ±0.31 / 381.21 ms │        379.45 / 381.55 ±3.15 / 387.77 ms │     no change │
│ QQuery 25 │        338.96 / 340.40 ±0.86 / 341.43 ms │        335.95 / 340.15 ±2.61 / 344.21 ms │     no change │
│ QQuery 26 │           80.34 / 82.27 ±2.10 / 86.08 ms │           80.96 / 83.38 ±1.83 / 85.87 ms │     no change │
│ QQuery 27 │              6.95 / 7.35 ±0.56 / 8.46 ms │              6.85 / 7.08 ±0.19 / 7.27 ms │     no change │
│ QQuery 28 │       148.58 / 159.35 ±18.03 / 195.33 ms │        148.97 / 149.38 ±0.28 / 149.79 ms │ +1.07x faster │
│ QQuery 29 │        278.93 / 281.01 ±1.58 / 283.15 ms │        279.13 / 281.08 ±1.69 / 284.17 ms │     no change │
│ QQuery 30 │           43.16 / 44.49 ±0.89 / 45.35 ms │           42.94 / 44.02 ±0.94 / 45.73 ms │     no change │
│ QQuery 31 │        171.36 / 172.54 ±1.41 / 175.16 ms │        171.33 / 173.70 ±2.65 / 178.70 ms │     no change │
│ QQuery 32 │           56.41 / 57.42 ±1.15 / 59.62 ms │           57.17 / 57.93 ±0.60 / 58.89 ms │     no change │
│ QQuery 33 │        139.77 / 143.27 ±2.40 / 147.15 ms │        138.99 / 140.85 ±1.49 / 142.96 ms │     no change │
│ QQuery 34 │              6.89 / 7.19 ±0.28 / 7.62 ms │              6.88 / 7.05 ±0.17 / 7.37 ms │     no change │
│ QQuery 35 │        105.35 / 108.47 ±2.33 / 111.32 ms │        107.56 / 109.29 ±1.04 / 110.35 ms │     no change │
│ QQuery 36 │              6.47 / 6.72 ±0.20 / 6.99 ms │              6.47 / 6.67 ±0.16 / 6.96 ms │     no change │
│ QQuery 37 │             8.35 / 9.10 ±0.96 / 10.96 ms │              8.53 / 8.82 ±0.24 / 9.19 ms │     no change │
│ QQuery 38 │           86.06 / 87.65 ±1.33 / 89.93 ms │           84.04 / 88.05 ±3.20 / 92.70 ms │     no change │
│ QQuery 39 │        127.27 / 129.28 ±1.63 / 131.19 ms │        123.41 / 126.96 ±2.36 / 129.90 ms │     no change │
│ QQuery 40 │        106.36 / 114.28 ±7.93 / 128.75 ms │        106.95 / 114.64 ±6.30 / 125.93 ms │     no change │
│ QQuery 41 │           14.42 / 15.73 ±1.13 / 17.13 ms │           14.76 / 15.18 ±0.34 / 15.76 ms │     no change │
│ QQuery 42 │        106.26 / 108.54 ±1.64 / 110.45 ms │        107.14 / 109.33 ±1.81 / 111.64 ms │     no change │
│ QQuery 43 │              6.01 / 6.23 ±0.27 / 6.65 ms │              5.97 / 6.12 ±0.23 / 6.58 ms │     no change │
│ QQuery 44 │           11.51 / 11.88 ±0.22 / 12.13 ms │           11.75 / 12.39 ±0.75 / 13.85 ms │     no change │
│ QQuery 45 │           49.57 / 51.53 ±1.26 / 52.89 ms │           49.91 / 51.40 ±1.00 / 52.52 ms │     no change │
│ QQuery 46 │             8.39 / 9.08 ±0.67 / 10.16 ms │             8.56 / 9.17 ±0.62 / 10.29 ms │     no change │
│ QQuery 47 │        705.31 / 708.56 ±3.08 / 714.40 ms │       732.56 / 747.70 ±12.37 / 759.72 ms │  1.06x slower │
│ QQuery 48 │        279.98 / 288.44 ±4.60 / 293.74 ms │        289.51 / 294.29 ±5.39 / 304.42 ms │     no change │
│ QQuery 49 │        249.28 / 253.01 ±2.21 / 255.68 ms │        249.30 / 252.24 ±2.14 / 254.28 ms │     no change │
│ QQuery 50 │        219.33 / 222.82 ±3.70 / 229.93 ms │        225.45 / 227.81 ±2.00 / 231.41 ms │     no change │
│ QQuery 51 │        176.90 / 182.89 ±4.32 / 190.16 ms │        180.76 / 182.96 ±1.99 / 186.28 ms │     no change │
│ QQuery 52 │        106.76 / 108.98 ±2.08 / 112.45 ms │        106.00 / 108.65 ±1.99 / 111.69 ms │     no change │
│ QQuery 53 │        100.74 / 102.78 ±1.42 / 105.02 ms │        102.67 / 104.41 ±1.63 / 107.33 ms │     no change │
│ QQuery 54 │        146.40 / 147.58 ±0.95 / 149.17 ms │        145.86 / 148.03 ±3.05 / 154.08 ms │     no change │
│ QQuery 55 │        106.41 / 107.77 ±1.11 / 109.47 ms │        106.30 / 109.01 ±1.86 / 111.18 ms │     no change │
│ QQuery 56 │        139.74 / 141.91 ±1.98 / 145.03 ms │        140.14 / 141.32 ±1.21 / 143.51 ms │     no change │
│ QQuery 57 │        173.90 / 177.60 ±1.98 / 179.38 ms │        178.49 / 180.19 ±0.98 / 181.55 ms │     no change │
│ QQuery 58 │        287.77 / 298.01 ±9.69 / 311.79 ms │        302.91 / 310.99 ±6.74 / 319.34 ms │     no change │
│ QQuery 59 │        198.98 / 202.61 ±3.77 / 209.10 ms │        198.05 / 201.42 ±2.58 / 204.87 ms │     no change │
│ QQuery 60 │        143.79 / 144.37 ±0.35 / 144.81 ms │        141.87 / 143.85 ±1.23 / 145.68 ms │     no change │
│ QQuery 61 │           13.02 / 13.67 ±0.75 / 15.04 ms │           12.84 / 13.19 ±0.36 / 13.73 ms │     no change │
│ QQuery 62 │       893.68 / 923.03 ±20.29 / 947.50 ms │       906.86 / 945.07 ±35.53 / 989.32 ms │     no change │
│ QQuery 63 │        103.57 / 106.24 ±2.40 / 110.24 ms │        103.59 / 105.47 ±1.36 / 107.24 ms │     no change │
│ QQuery 64 │        671.71 / 677.52 ±6.03 / 689.11 ms │        678.29 / 687.44 ±7.96 / 699.28 ms │     no change │
│ QQuery 65 │        248.02 / 252.97 ±3.42 / 257.55 ms │        249.20 / 251.37 ±2.07 / 254.77 ms │     no change │
│ QQuery 66 │        241.76 / 255.01 ±7.04 / 261.29 ms │       242.70 / 266.01 ±18.42 / 298.70 ms │     no change │
│ QQuery 67 │        309.02 / 312.34 ±3.01 / 316.76 ms │        346.84 / 350.01 ±1.80 / 352.08 ms │  1.12x slower │
│ QQuery 68 │            8.94 / 11.29 ±2.58 / 16.15 ms │            9.11 / 10.49 ±1.22 / 12.66 ms │ +1.08x faster │
│ QQuery 69 │         99.92 / 103.09 ±2.65 / 107.73 ms │        102.60 / 103.67 ±1.06 / 105.31 ms │     no change │
│ QQuery 70 │       339.98 / 351.21 ±11.28 / 371.57 ms │       329.84 / 347.77 ±16.26 / 378.42 ms │     no change │
│ QQuery 71 │        133.27 / 135.23 ±1.36 / 136.87 ms │        135.26 / 140.23 ±5.42 / 149.77 ms │     no change │
│ QQuery 72 │        612.82 / 620.56 ±7.72 / 635.08 ms │        620.53 / 629.00 ±7.83 / 641.47 ms │     no change │
│ QQuery 73 │             6.53 / 8.17 ±2.34 / 12.79 ms │              6.68 / 8.32 ±0.95 / 9.57 ms │     no change │
│ QQuery 74 │        569.77 / 581.96 ±7.74 / 591.15 ms │        578.15 / 585.57 ±4.21 / 590.37 ms │     no change │
│ QQuery 75 │        274.99 / 278.58 ±3.21 / 283.73 ms │        279.19 / 281.01 ±1.48 / 283.25 ms │     no change │
│ QQuery 76 │        130.58 / 132.80 ±2.58 / 137.17 ms │        134.44 / 135.87 ±1.43 / 138.37 ms │     no change │
│ QQuery 77 │        188.64 / 191.29 ±2.30 / 194.63 ms │        190.22 / 191.94 ±1.93 / 195.35 ms │     no change │
│ QQuery 78 │        337.59 / 340.77 ±2.00 / 343.46 ms │        339.65 / 342.28 ±3.62 / 349.30 ms │     no change │
│ QQuery 79 │        230.94 / 234.72 ±3.04 / 238.40 ms │        231.89 / 234.83 ±1.58 / 236.31 ms │     no change │
│ QQuery 80 │        320.54 / 322.96 ±1.91 / 325.85 ms │        322.63 / 323.27 ±0.51 / 324.15 ms │     no change │
│ QQuery 81 │           26.29 / 26.90 ±0.85 / 28.57 ms │           26.06 / 26.49 ±0.30 / 26.89 ms │     no change │
│ QQuery 82 │        194.88 / 200.07 ±4.52 / 206.90 ms │        197.73 / 201.84 ±2.84 / 204.61 ms │     no change │
│ QQuery 83 │           38.54 / 39.15 ±0.47 / 39.93 ms │           38.05 / 39.54 ±1.00 / 41.09 ms │     no change │
│ QQuery 84 │           48.38 / 49.85 ±1.14 / 51.20 ms │           48.27 / 49.32 ±0.86 / 50.75 ms │     no change │
│ QQuery 85 │        147.14 / 148.14 ±0.64 / 148.80 ms │        147.86 / 148.76 ±0.79 / 150.00 ms │     no change │
│ QQuery 86 │           39.07 / 40.98 ±1.26 / 42.64 ms │           39.08 / 40.05 ±0.75 / 41.04 ms │     no change │
│ QQuery 87 │           86.01 / 90.69 ±4.13 / 96.26 ms │           84.37 / 89.25 ±3.78 / 95.72 ms │     no change │
│ QQuery 88 │        100.73 / 101.21 ±0.35 / 101.68 ms │         99.64 / 100.29 ±0.68 / 101.61 ms │     no change │
│ QQuery 89 │        117.22 / 118.83 ±1.15 / 120.14 ms │        118.03 / 119.72 ±1.07 / 121.00 ms │     no change │
│ QQuery 90 │           23.35 / 24.77 ±0.99 / 26.13 ms │           23.00 / 25.15 ±1.36 / 27.05 ms │     no change │
│ QQuery 91 │           65.25 / 66.12 ±0.70 / 66.98 ms │           63.15 / 64.60 ±1.14 / 66.08 ms │     no change │
│ QQuery 92 │           57.41 / 58.04 ±0.38 / 58.53 ms │           57.60 / 58.38 ±0.59 / 59.24 ms │     no change │
│ QQuery 93 │        185.01 / 187.21 ±2.03 / 189.98 ms │        184.87 / 186.98 ±1.68 / 189.49 ms │     no change │
│ QQuery 94 │           60.81 / 61.71 ±0.62 / 62.37 ms │           61.61 / 62.33 ±0.76 / 63.70 ms │     no change │
│ QQuery 95 │        125.25 / 127.04 ±1.44 / 128.49 ms │        127.00 / 128.16 ±1.06 / 129.57 ms │     no change │
│ QQuery 96 │           72.94 / 74.57 ±0.89 / 75.45 ms │           70.38 / 73.92 ±2.01 / 76.43 ms │     no change │
│ QQuery 97 │        126.55 / 128.68 ±1.76 / 131.20 ms │        125.80 / 128.77 ±2.38 / 132.39 ms │     no change │
│ QQuery 98 │        153.24 / 155.03 ±1.04 / 156.47 ms │        154.12 / 155.59 ±1.07 / 157.19 ms │     no change │
│ QQuery 99 │ 10758.45 / 10813.96 ±34.55 / 10861.29 ms │ 10846.28 / 10891.64 ±29.53 / 10932.77 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 31426.72ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 31774.83ms │
│ Average Time (HEAD)                                │   317.44ms │
│ Average Time (optimize-byteview-vectorized-append) │   320.96ms │
│ Queries Faster                                     │          2 │
│ Queries Slower                                     │          3 │
│ Queries with No Change                             │         94 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	157.5s
Peak memory	5.9 GiB
Avg memory	4.8 GiB
CPU user	259.6s
CPU sys	16.5s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	159.2s
Peak memory	5.2 GiB
Avg memory	4.5 GiB
CPU user	262.0s
CPU sys	17.3s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃   optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.23 / 4.63 ±6.64 / 17.90 ms │          1.22 / 4.50 ±6.45 / 17.39 ms │     no change │
│ QQuery 1  │        14.81 / 15.16 ±0.26 / 15.58 ms │        14.33 / 14.62 ±0.23 / 15.03 ms │     no change │
│ QQuery 2  │        43.45 / 43.87 ±0.28 / 44.28 ms │        43.97 / 44.16 ±0.15 / 44.33 ms │     no change │
│ QQuery 3  │        41.62 / 45.11 ±2.88 / 48.31 ms │        41.69 / 45.22 ±2.07 / 47.41 ms │     no change │
│ QQuery 4  │    290.74 / 299.19 ±11.38 / 321.71 ms │     284.38 / 298.34 ±7.19 / 303.94 ms │     no change │
│ QQuery 5  │     343.39 / 348.65 ±3.11 / 352.89 ms │     344.11 / 347.80 ±2.89 / 350.92 ms │     no change │
│ QQuery 6  │           5.34 / 6.53 ±1.30 / 8.64 ms │           5.59 / 6.13 ±0.33 / 6.49 ms │ +1.07x faster │
│ QQuery 7  │        17.13 / 17.79 ±0.81 / 19.38 ms │        16.80 / 17.91 ±0.87 / 18.88 ms │     no change │
│ QQuery 8  │     413.96 / 421.25 ±4.20 / 425.98 ms │     434.38 / 439.59 ±3.85 / 444.11 ms │     no change │
│ QQuery 9  │     643.58 / 655.48 ±8.00 / 668.43 ms │     679.52 / 687.39 ±4.74 / 693.98 ms │     no change │
│ QQuery 10 │        91.63 / 93.41 ±1.27 / 95.23 ms │        94.53 / 96.33 ±1.33 / 98.29 ms │     no change │
│ QQuery 11 │     105.05 / 107.82 ±4.01 / 115.61 ms │     107.29 / 109.56 ±1.96 / 112.39 ms │     no change │
│ QQuery 12 │     339.96 / 344.39 ±2.82 / 347.45 ms │     351.28 / 356.45 ±5.71 / 366.61 ms │     no change │
│ QQuery 13 │    455.00 / 472.66 ±15.40 / 496.01 ms │     492.65 / 505.31 ±8.44 / 515.82 ms │  1.07x slower │
│ QQuery 14 │     348.20 / 349.82 ±0.99 / 351.21 ms │     412.45 / 417.68 ±3.28 / 421.69 ms │  1.19x slower │
│ QQuery 15 │    358.94 / 382.91 ±19.65 / 408.64 ms │     362.31 / 370.54 ±6.16 / 378.60 ms │     no change │
│ QQuery 16 │     719.45 / 725.51 ±5.17 / 732.41 ms │ 1422.05 / 1443.55 ±16.58 / 1471.81 ms │  1.99x slower │
│ QQuery 17 │     713.43 / 715.79 ±2.32 / 719.47 ms │  1409.68 / 1422.96 ±6.87 / 1428.50 ms │  1.99x slower │
│ QQuery 18 │ 1441.12 / 1465.71 ±21.73 / 1490.33 ms │ 2567.11 / 2637.39 ±45.19 / 2702.41 ms │  1.80x slower │
│ QQuery 19 │        36.23 / 37.70 ±1.25 / 39.59 ms │        37.32 / 38.36 ±0.94 / 39.93 ms │     no change │
│ QQuery 20 │    719.04 / 732.92 ±14.00 / 756.36 ms │     726.18 / 740.29 ±8.12 / 750.37 ms │     no change │
│ QQuery 21 │     764.36 / 766.18 ±1.49 / 768.36 ms │     761.56 / 772.70 ±6.30 / 780.96 ms │     no change │
│ QQuery 22 │  1133.04 / 1140.02 ±6.30 / 1150.64 ms │  1136.19 / 1144.12 ±4.81 / 1150.12 ms │     no change │
│ QQuery 23 │ 3087.94 / 3097.85 ±11.90 / 3118.30 ms │ 3096.58 / 3120.79 ±14.46 / 3138.46 ms │     no change │
│ QQuery 24 │     101.32 / 103.18 ±1.51 / 105.80 ms │      99.51 / 103.59 ±3.50 / 109.88 ms │     no change │
│ QQuery 25 │     138.49 / 140.30 ±1.42 / 142.34 ms │     136.42 / 140.39 ±2.69 / 143.46 ms │     no change │
│ QQuery 26 │     101.41 / 102.30 ±1.06 / 104.30 ms │      99.92 / 103.46 ±1.93 / 105.79 ms │     no change │
│ QQuery 27 │     849.76 / 858.90 ±6.81 / 868.81 ms │     853.93 / 858.40 ±5.20 / 867.87 ms │     no change │
│ QQuery 28 │ 3256.14 / 3297.56 ±21.60 / 3316.89 ms │ 3280.29 / 3301.02 ±13.86 / 3320.78 ms │     no change │
│ QQuery 29 │        51.37 / 55.00 ±4.94 / 64.68 ms │        51.41 / 54.73 ±2.96 / 58.59 ms │     no change │
│ QQuery 30 │     361.07 / 365.61 ±2.88 / 369.08 ms │     366.50 / 371.80 ±4.38 / 379.69 ms │     no change │
│ QQuery 31 │    369.11 / 381.04 ±10.63 / 399.76 ms │    364.54 / 378.46 ±13.38 / 400.36 ms │     no change │
│ QQuery 32 │  1051.40 / 1056.43 ±8.68 / 1073.74 ms │ 1054.95 / 1091.16 ±47.14 / 1183.23 ms │     no change │
│ QQuery 33 │ 1455.78 / 1474.78 ±13.90 / 1492.50 ms │  1483.34 / 1500.59 ±9.51 / 1508.70 ms │     no change │
│ QQuery 34 │ 1483.21 / 1498.29 ±10.60 / 1514.44 ms │ 1511.68 / 1526.84 ±12.93 / 1544.24 ms │     no change │
│ QQuery 35 │     389.18 / 391.50 ±2.43 / 395.65 ms │     393.45 / 399.89 ±5.67 / 409.27 ms │     no change │
│ QQuery 36 │     112.19 / 122.49 ±6.18 / 130.19 ms │     123.40 / 125.92 ±2.30 / 129.99 ms │     no change │
│ QQuery 37 │        46.76 / 48.40 ±1.18 / 49.89 ms │        49.00 / 50.08 ±0.84 / 51.38 ms │     no change │
│ QQuery 38 │        75.10 / 77.69 ±1.51 / 79.71 ms │        77.76 / 78.61 ±1.08 / 80.71 ms │     no change │
│ QQuery 39 │     207.91 / 217.73 ±7.70 / 229.12 ms │     224.46 / 233.10 ±5.80 / 242.20 ms │  1.07x slower │
│ QQuery 40 │        22.78 / 25.53 ±3.09 / 31.26 ms │        25.05 / 25.96 ±0.75 / 27.26 ms │     no change │
│ QQuery 41 │        20.43 / 21.41 ±0.79 / 22.33 ms │        20.91 / 22.29 ±1.24 / 24.53 ms │     no change │
│ QQuery 42 │        20.03 / 20.48 ±0.24 / 20.70 ms │        20.82 / 21.23 ±0.40 / 21.98 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 22548.99ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 25469.21ms │
│ Average Time (HEAD)                                │   524.40ms │
│ Average Time (optimize-byteview-vectorized-append) │   592.31ms │
│ Queries Faster                                     │          1 │
│ Queries Slower                                     │          6 │
│ Queries with No Change                             │         36 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	113.9s
Peak memory	38.7 GiB
Avg memory	31.1 GiB
CPU user	1079.8s
CPU sys	80.9s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	128.9s
Peak memory	58.2 GiB
Avg memory	45.2 GiB
CPU user	1220.6s
CPU sys	89.3s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃   optimize-byteview-vectorized-append ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.20 / 4.43 ±6.32 / 17.06 ms │          1.21 / 4.51 ±6.44 / 17.39 ms │     no change │
│ QQuery 1  │        14.06 / 14.41 ±0.19 / 14.63 ms │        14.01 / 14.63 ±0.36 / 15.00 ms │     no change │
│ QQuery 2  │        43.25 / 43.65 ±0.32 / 44.06 ms │        44.14 / 44.45 ±0.23 / 44.74 ms │     no change │
│ QQuery 3  │        41.49 / 43.62 ±2.39 / 47.83 ms │        43.21 / 45.76 ±1.50 / 47.68 ms │     no change │
│ QQuery 4  │     291.46 / 295.26 ±2.78 / 299.38 ms │     294.74 / 303.15 ±7.61 / 315.93 ms │     no change │
│ QQuery 5  │     344.29 / 347.34 ±2.07 / 350.79 ms │     347.73 / 353.98 ±4.83 / 361.43 ms │     no change │
│ QQuery 6  │          5.74 / 7.47 ±2.11 / 10.92 ms │          5.11 / 8.11 ±2.89 / 13.50 ms │  1.08x slower │
│ QQuery 7  │        16.69 / 17.33 ±1.11 / 19.53 ms │        16.93 / 17.19 ±0.16 / 17.32 ms │     no change │
│ QQuery 8  │     419.13 / 426.24 ±6.42 / 435.16 ms │     419.00 / 428.85 ±6.62 / 438.34 ms │     no change │
│ QQuery 9  │     649.64 / 655.07 ±4.64 / 660.94 ms │     656.15 / 665.34 ±7.39 / 676.66 ms │     no change │
│ QQuery 10 │        94.46 / 95.49 ±0.94 / 96.68 ms │       94.57 / 98.11 ±2.38 / 101.37 ms │     no change │
│ QQuery 11 │     105.93 / 106.82 ±0.99 / 108.28 ms │     106.24 / 108.26 ±1.45 / 110.31 ms │     no change │
│ QQuery 12 │     342.54 / 349.23 ±4.57 / 355.02 ms │     350.00 / 357.16 ±5.80 / 365.67 ms │     no change │
│ QQuery 13 │     475.33 / 483.95 ±8.70 / 499.62 ms │     471.55 / 481.24 ±9.95 / 499.75 ms │     no change │
│ QQuery 14 │     352.34 / 359.62 ±5.24 / 368.61 ms │     347.44 / 352.37 ±2.62 / 354.63 ms │     no change │
│ QQuery 15 │    356.75 / 367.56 ±11.70 / 388.86 ms │    370.11 / 388.00 ±17.83 / 422.19 ms │  1.06x slower │
│ QQuery 16 │    730.88 / 749.84 ±19.26 / 781.70 ms │    745.62 / 764.19 ±14.37 / 782.06 ms │     no change │
│ QQuery 17 │     717.57 / 723.28 ±3.34 / 727.60 ms │    707.09 / 729.46 ±26.32 / 778.67 ms │     no change │
│ QQuery 18 │ 1465.73 / 1495.27 ±24.90 / 1527.79 ms │ 1395.96 / 1450.38 ±47.36 / 1512.96 ms │     no change │
│ QQuery 19 │       36.93 / 45.30 ±15.77 / 76.81 ms │        35.64 / 37.77 ±1.62 / 39.79 ms │ +1.20x faster │
│ QQuery 20 │    721.26 / 742.35 ±22.93 / 779.37 ms │     713.54 / 724.52 ±8.45 / 738.60 ms │     no change │
│ QQuery 21 │     768.05 / 772.61 ±3.39 / 778.57 ms │     759.21 / 763.41 ±2.51 / 767.12 ms │     no change │
│ QQuery 22 │  1150.09 / 1158.27 ±7.69 / 1172.77 ms │  1136.29 / 1138.37 ±2.47 / 1143.16 ms │     no change │
│ QQuery 23 │ 3124.91 / 3148.64 ±20.92 / 3182.45 ms │ 3107.52 / 3136.21 ±17.35 / 3155.67 ms │     no change │
│ QQuery 24 │     100.14 / 105.01 ±3.46 / 108.07 ms │     101.77 / 103.86 ±1.97 / 106.71 ms │     no change │
│ QQuery 25 │     140.50 / 142.55 ±2.28 / 146.99 ms │     142.57 / 145.03 ±1.55 / 146.80 ms │     no change │
│ QQuery 26 │     100.77 / 102.54 ±1.77 / 105.78 ms │     101.52 / 104.58 ±1.71 / 106.71 ms │     no change │
│ QQuery 27 │    853.55 / 861.81 ±11.24 / 883.74 ms │     854.23 / 860.76 ±6.92 / 872.34 ms │     no change │
│ QQuery 28 │ 3294.97 / 3323.20 ±20.20 / 3346.05 ms │ 3307.54 / 3335.91 ±17.94 / 3352.51 ms │     no change │
│ QQuery 29 │        52.05 / 55.74 ±3.37 / 59.90 ms │        51.07 / 54.74 ±4.12 / 62.25 ms │     no change │
│ QQuery 30 │     368.33 / 369.65 ±0.92 / 370.91 ms │     362.70 / 365.97 ±2.60 / 369.81 ms │     no change │
│ QQuery 31 │    371.49 / 386.28 ±12.07 / 407.56 ms │    369.49 / 382.89 ±11.38 / 397.61 ms │     no change │
│ QQuery 32 │ 1051.23 / 1075.38 ±27.04 / 1126.91 ms │ 1053.21 / 1071.70 ±25.11 / 1121.36 ms │     no change │
│ QQuery 33 │  1478.14 / 1490.69 ±8.73 / 1504.14 ms │ 1473.44 / 1496.66 ±13.38 / 1512.75 ms │     no change │
│ QQuery 34 │  1500.52 / 1512.13 ±8.00 / 1525.06 ms │ 1490.17 / 1520.53 ±25.09 / 1559.33 ms │     no change │
│ QQuery 35 │     394.48 / 409.06 ±8.79 / 420.41 ms │     385.80 / 391.63 ±3.56 / 396.14 ms │     no change │
│ QQuery 36 │     122.39 / 124.21 ±1.33 / 126.47 ms │     112.71 / 119.64 ±3.73 / 122.76 ms │     no change │
│ QQuery 37 │        48.37 / 50.62 ±2.80 / 55.96 ms │        48.79 / 49.60 ±0.90 / 50.88 ms │     no change │
│ QQuery 38 │        76.33 / 77.34 ±0.93 / 79.05 ms │        75.57 / 78.29 ±2.35 / 81.19 ms │     no change │
│ QQuery 39 │     213.20 / 216.91 ±2.95 / 220.37 ms │     215.24 / 221.45 ±3.99 / 225.38 ms │     no change │
│ QQuery 40 │        25.43 / 26.98 ±1.39 / 29.07 ms │        21.97 / 24.33 ±2.33 / 28.23 ms │ +1.11x faster │
│ QQuery 41 │        21.22 / 22.05 ±0.51 / 22.80 ms │        20.63 / 22.10 ±1.69 / 25.21 ms │     no change │
│ QQuery 42 │        19.93 / 20.07 ±0.11 / 20.27 ms │        19.70 / 20.49 ±0.51 / 21.15 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 22825.31ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 22785.53ms │
│ Average Time (HEAD)                                │   530.82ms │
│ Average Time (optimize-byteview-vectorized-append) │   529.90ms │
│ Queries Faster                                     │          2 │
│ Queries Slower                                     │          2 │
│ Queries with No Change                             │         39 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	115.2s
Peak memory	37.3 GiB
Avg memory	27.8 GiB
CPU user	1088.6s
CPU sys	85.8s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	115.2s
Peak memory	42.1 GiB
Avg memory	35.5 GiB
CPU user	1087.4s
CPU sys	86.0s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and optimize-byteview-vectorized-append
--------------------
Benchmark clickbench_extended.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                                    HEAD ┃     optimize-byteview-vectorized-append ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 0  │       836.51 / 847.47 ±9.66 / 864.87 ms │       868.01 / 880.33 ±7.91 / 888.78 ms │ no change │
│ QQuery 1  │       209.81 / 210.89 ±1.35 / 213.51 ms │       213.60 / 214.83 ±0.70 / 215.54 ms │ no change │
│ QQuery 2  │       503.24 / 504.79 ±1.50 / 507.51 ms │       507.58 / 512.59 ±2.72 / 515.40 ms │ no change │
│ QQuery 3  │       316.99 / 319.39 ±1.82 / 321.66 ms │       317.93 / 320.67 ±1.86 / 322.84 ms │ no change │
│ QQuery 4  │      664.33 / 676.03 ±12.88 / 700.16 ms │       674.65 / 683.91 ±6.45 / 692.96 ms │ no change │
│ QQuery 5  │ 9793.25 / 9960.13 ±138.31 / 10143.01 ms │ 9520.36 / 9911.18 ±356.48 / 10583.73 ms │ no change │
│ QQuery 6  │   1000.92 / 1012.03 ±14.91 / 1041.11 ms │    1003.55 / 1010.83 ±6.84 / 1021.00 ms │ no change │
│ QQuery 7  │       777.11 / 783.45 ±5.79 / 793.48 ms │       787.93 / 793.13 ±5.09 / 801.06 ms │ no change │
│ QQuery 8  │       397.82 / 404.16 ±4.86 / 412.71 ms │       405.79 / 411.20 ±3.96 / 417.27 ms │ no change │
│ QQuery 9  │   2842.60 / 2860.70 ±17.73 / 2891.25 ms │   2823.85 / 2836.25 ±13.53 / 2859.19 ms │ no change │
│ QQuery 10 │      634.51 / 646.02 ±13.76 / 672.03 ms │       633.10 / 639.29 ±6.03 / 649.54 ms │ no change │
│ QQuery 11 │   2049.98 / 2065.82 ±10.38 / 2078.93 ms │   2055.38 / 2075.65 ±15.54 / 2094.62 ms │ no change │
│ QQuery 12 │       198.62 / 205.14 ±7.88 / 220.46 ms │       198.88 / 203.04 ±3.97 / 210.18 ms │ no change │
└───────────┴─────────────────────────────────────────┴─────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                  │ 20496.03ms │
│ Total Time (optimize-byteview-vectorized-append)   │ 20492.91ms │
│ Average Time (HEAD)                                │  1576.62ms │
│ Average Time (optimize-byteview-vectorized-append) │  1576.38ms │
│ Queries Faster                                     │          0 │
│ Queries Slower                                     │          0 │
│ Queries with No Change                             │         13 │
│ Queries with Failure                               │          0 │
└────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_extended — base (merge-base)

Metric	Value
Wall time	103.3s
Peak memory	33.7 GiB
Avg memory	29.4 GiB
CPU user	995.4s
CPU sys	46.6s
Peak spill	0 B

clickbench_extended — branch

Metric	Value
Wall time	103.3s
Peak memory	33.6 GiB
Avg memory	28.6 GiB
CPU user	992.6s
CPU sys	47.6s
Peak spill	0 B

---

File an issue against this benchmark runner