FileSystem for local reads - #9773
Conversation
Signed-off-by: Mikhail Kot <mikhail@spiraldb.com>
Merging this PR will degrade performance by 9.27%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | WallTime | arrow_checked_add_u32_neon[16384] |
13.3 µs | 20.3 µs | -34.44% |
| ❌ | Simulation | random_i8[0.8] |
69.9 µs | 99.3 µs | -29.58% |
| ❌ | Simulation | decompress[u64, (4000, 1024)] |
70.8 µs | 86.1 µs | -17.78% |
| ❌ | WallTime | mul_u32_nonnull_avx512 |
5.6 µs | 6.2 µs | -10.4% |
| ⚡ | Simulation | random_i8[0.5] |
91.4 µs | 67.5 µs | +35.4% |
| ⚡ | WallTime | arrow_checked_add_u32_avx2[16384] |
21.4 µs | 17.7 µs | +21.11% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing myrrc/local-fs-reads (3fe72c8) with develop (9fdaa38)2
Footnotes
-
218 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩
-
No successful run was found on
develop(272bf1d) during the generation of this report, so 9fdaa38 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩
Polar Signals Profiling ResultsLatest Run
Powered by Polar Signals Cloud |
Polar Signals Profiling ResultsLatest Run
Powered by Polar Signals Cloud |
Benchmarks: PolarSignals Profiling 📖Commits: PR datafusion / vortex-file-compressed / ns (0.998x ➖, 0↑ 0↓)
No file size changes detected. |
Benchmarks: TPC-H SF=1 on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.998x ➖, 0↑ 0↓)
datafusion / parquet / ns (1.000x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (1.002x ➖, 1↑ 2↓)
duckdb / parquet / ns (0.995x ➖, 0↑ 0↓)
No file size changes detected. |
Benchmarks: FineWeb NVMe 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.995x ➖, 0↑ 0↓)
datafusion / parquet / ns (1.001x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (0.977x ➖, 2↑ 3↓)
duckdb / parquet / ns (1.003x ➖, 0↑ 0↓)
No file size changes detected. |
Benchmarks: Clickbench Sorted on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.943x ➖, 2↑ 0↓)
datafusion / parquet / ns (0.998x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (1.001x ➖, 1↑ 1↓)
duckdb / parquet / ns (1.003x ➖, 0↑ 0↓)
File Size Changes (100 files changed, +0.0% overall, 57↑ 43↓)
Totals:
|
Benchmarks: FineWeb S3 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (1.035x ➖, 0↑ 1↓)
datafusion / parquet / ns (0.992x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (1.005x ➖, 0↑ 0↓)
duckdb / parquet / ns (0.936x ➖, 0↑ 0↓)
|
Benchmarks: Clickbench on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.998x ➖, 0↑ 0↓)
datafusion / parquet / ns (1.001x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (1.003x ➖, 1↑ 2↓)
duckdb / parquet / ns (1.013x ➖, 0↑ 3↓)
No file size changes detected. |
Benchmarks: TPC-DS SF=1 on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.999x ➖, 0↑ 0↓)
datafusion / parquet / ns (1.003x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (0.995x ➖, 2↑ 1↓)
duckdb / parquet / ns (0.994x ➖, 2↑ 3↓)
No file size changes detected. |
Benchmarks: TPC-H SF=10 on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (1.003x ➖, 0↑ 0↓)
datafusion / parquet / ns (1.002x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (1.002x ➖, 0↑ 1↓)
duckdb / parquet / ns (0.991x ➖, 0↑ 0↓)
No file size changes detected. |
Benchmarks: Statistical and Population Genetics 📖Commits: PR How to read Verdict and Engines
duckdb / vortex-file-compressed / ns (1.001x ➖, 0↑ 0↓)
duckdb / parquet / ns (1.000x ➖, 0↑ 0↓)
No file size changes detected. |
Benchmarks: TPC-H SF=1 on S3 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.970x ➖, 1↑ 3↓)
datafusion / parquet / ns (0.930x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (0.932x ➖, 1↑ 0↓)
duckdb / parquet / ns (0.941x ➖, 0↑ 0↓)
|
We have vortex-cloud and ObjectStoreFileSystem which is a unified interface for
reads. However, it does many things which are needed for an object store but
excessive for local disk i.e. etag creation.
As another example, get_opts currently calls open() for every segment which we
also don't need because a local read may use the cached file descriptor.
This change implements a local FileSystem and adds it support to vortex-duckdb.