Skip to content

feat: Reuse existing file instead of reopening during shuffle write #2577

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 3 commits into
base: main
Choose a base branch
from
Draft

feat: Reuse existing file instead of reopening during shuffle write #2577

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
393b432
feat: Reuse existing file instead of reopening during shuffle write
zuston Oct 15, 2025
97a398e
allow deadcode
zuston Oct 15, 2025
3ec2047
fix
zuston Oct 16, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 6 additions & 3 deletions native/core/src/execution/shuffle/shuffle_writer.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -779,9 +779,10 @@ impl ShufflePartitioner for MultiPartitionShuffleRepartitioner {

// if we wrote a spill file for this partition then copy the
// contents into the shuffle file
if let Some(spill_data) = self.partition_writers[i].spill_file.as_ref() {
let mut spill_file =
BufReader::new(File::open(spill_data.temp_file.path()).map_err(to_df_err)?);
if let Some(spill_data) = self.partition_writers[i].spill_file.as_mut() {
let file = &mut spill_data.file;
file.seek(SeekFrom::Start(0))?;
let mut spill_file = BufReader::new(file);
let mut write_timer = self.metrics.write_time.timer();
std::io::copy(&mut spill_file, &mut output_data).map_err(to_df_err)?;
write_timer.stop();
Expand Down Expand Up @@ -1113,6 +1114,7 @@ struct PartitionWriter {
}

struct SpillFile {
#[allow(dead_code)]
temp_file: RefCountedTempFile,
file: File,
}
Expand Down Expand Up @@ -1168,6 +1170,7 @@ impl PartitionWriter {
let spill_data = OpenOptions::new()
.write(true)
.create(true)
.read(true)
.truncate(true)
.open(spill_file.path())
.map_err(|e| {
Expand Down
Loading