Repository navigation
perf: wip sub seg p3 - #7392
Closed
joseph-isaacs wants to merge 22 commits into
Closed
perf: wip sub seg p3#7392joseph-isaacs wants to merge 22 commits into
joseph-isaacs wants to merge 22 commits into
Conversation
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk> # Conflicts: # vortex-array/public-api.lock # vortex-array/src/array/vtable/dyn_.rs # vortex-array/src/array/vtable/mod.rs # vortex-array/src/arrays/bool/array.rs # vortex-array/src/arrays/fixed_size_list/compute/slice.rs # vortex-layout/src/layouts/flat/reader.rs # vortex-layout/src/layouts/flat/writer.rs # vortex-layout/src/segments/cache.rs
The lazy buffer path was issuing separate request_ranges() calls per buffer within a segment, bypassing SharedSegmentSource deduplication and SegmentCache. This caused N I/O requests per segment instead of 1, with catastrophic impact on S3 (2x slower on some queries). Key changes: - materialize_selection: Selection::Range now uses request() + local slice instead of request_ranges(), benefiting from caching and deduplication when multiple buffers share the same segment - Replace resolve_filter() with effective_selection() returning Cow<Selection>, avoiding a full LazyBufferHandle clone on every slice()/select_ranges() call when no filter is pending - Add MAX_SPARSE_RANGES (32) limit in materialize() to prevent hundreds of tiny I/O requests from scattered masks - Downgrade hot-path tracing from debug! to trace! Signed-off-by: Claude <noreply@anthropic.com> https://claude.ai/code/session_01LM4hVS2zQBoFA8X1YZmQ5Z
Contributor
Polar Signals Profiling ResultsLatest Run
Powered by Polar Signals Cloud |
Contributor
BENCHMARK FAILEDBenchmark |
Contributor
Benchmarks: PolarSignals ProfilingVortex (geomean): 1.087x ➖ datafusion / vortex-file-compressed (1.087x ➖, 0↑ 4↓)
|
Contributor
File Sizes: PolarSignals ProfilingFile Size Changes (1 files changed, +0.0% overall, 1↑ 0↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=1 on NVMEVerdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.966x ➖, 1↑ 1↓)
datafusion / vortex-compact (0.953x ➖, 4↑ 1↓)
datafusion / parquet (0.929x ➖, 6↑ 1↓)
datafusion / arrow (0.917x ➖, 8↑ 0↓)
duckdb / vortex-file-compressed (0.998x ➖, 1↑ 1↓)
duckdb / vortex-compact (0.977x ➖, 2↑ 1↓)
duckdb / parquet (0.978x ➖, 1↑ 3↓)
duckdb / duckdb (0.923x ➖, 5↑ 0↓)
Full attributed analysis
|
Contributor
File Sizes: TPC-H SF=1 on NVMENo file size changes detected. |
Contributor
Benchmarks: FineWeb NVMeVerdict: No clear signal (medium confidence) datafusion / vortex-file-compressed (0.943x ➖, 3↑ 3↓)
datafusion / vortex-compact (0.905x ➖, 1↑ 0↓)
datafusion / parquet (1.006x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.042x ➖, 1↑ 6↓)
duckdb / vortex-compact (0.933x ➖, 1↑ 0↓)
duckdb / parquet (0.998x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
File Sizes: FineWeb NVMeNo file size changes detected. |
Contributor
Benchmarks: TPC-DS SF=1 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.986x ➖, 9↑ 3↓)
datafusion / vortex-compact (0.997x ➖, 0↑ 2↓)
datafusion / parquet (0.933x ➖, 21↑ 0↓)
duckdb / vortex-file-compressed (1.031x ➖, 2↑ 14↓)
duckdb / vortex-compact (1.018x ➖, 2↑ 5↓)
duckdb / parquet (0.999x ➖, 1↑ 0↓)
duckdb / duckdb (0.973x ➖, 7↑ 1↓)
Full attributed analysis
|
Contributor
File Sizes: TPC-DS SF=1 on NVMENo file size changes detected. |
Contributor
Benchmarks: TPC-H SF=10 on NVMEVerdict: No clear signal (medium confidence) datafusion / vortex-file-compressed (0.971x ➖, 2↑ 1↓)
datafusion / vortex-compact (1.027x ➖, 0↑ 1↓)
datafusion / parquet (0.902x ➖, 11↑ 0↓)
datafusion / arrow (1.026x ➖, 2↑ 3↓)
duckdb / vortex-file-compressed (1.123x ❌, 0↑ 11↓)
duckdb / vortex-compact (1.062x ➖, 0↑ 4↓)
duckdb / parquet (0.977x ➖, 1↑ 1↓)
duckdb / duckdb (1.005x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
File Sizes: TPC-H SF=10 on NVMENo file size changes detected. |
Contributor
Benchmarks: TPC-H SF=1 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.745x ➖, 8↑ 0↓)
datafusion / vortex-compact (1.068x ➖, 0↑ 3↓)
datafusion / parquet (0.746x ➖, 6↑ 0↓)
duckdb / vortex-file-compressed (1.133x ➖, 0↑ 7↓)
duckdb / vortex-compact (1.098x ➖, 0↑ 6↓)
duckdb / parquet (0.885x ➖, 2↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: FineWeb S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.756x ➖, 3↑ 3↓)
datafusion / vortex-compact (0.702x ➖, 3↑ 0↓)
datafusion / parquet (0.828x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.969x ➖, 2↑ 3↓)
duckdb / vortex-compact (0.838x ➖, 3↑ 3↓)
duckdb / parquet (0.961x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: Statistical and Population GeneticsVerdict: No clear signal (low confidence) duckdb / vortex-file-compressed (0.984x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.993x ➖, 0↑ 0↓)
duckdb / parquet (0.943x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
File Sizes: Statistical and Population GeneticsNo file size changes detected. |
Contributor
Benchmarks: Clickbench on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (1.080x ➖, 1↑ 16↓)
datafusion / parquet (1.042x ➖, 0↑ 3↓)
duckdb / vortex-file-compressed (1.058x ➖, 6↑ 11↓)
duckdb / parquet (1.028x ➖, 0↑ 1↓)
duckdb / duckdb (1.002x ➖, 2↑ 2↓)
Full attributed analysis
|
Contributor
File Sizes: Clickbench on NVMEFile Size Changes (1 files changed, -0.0% overall, 0↑ 1↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=10 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (1.043x ➖, 1↑ 3↓)
datafusion / vortex-compact (1.044x ➖, 1↑ 4↓)
datafusion / parquet (0.988x ➖, 3↑ 5↓)
duckdb / vortex-file-compressed (1.175x ➖, 0↑ 5↓)
duckdb / vortex-compact (1.153x ➖, 0↑ 6↓)
duckdb / parquet (0.862x ➖, 1↑ 0↓)
Full attributed analysis
|
Contributor
|
This PR has been marked as stale because it has been open for 14 days with no activity. Please comment or remove the stale label if you wish to keep it active, otherwise it will be closed in 7 days |
Contributor
|
This PR was closed because it has been inactive for 7 days since being marked as stale. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Closes: #000
Testing