Closed artsiomkorzun closed 6 months ago
No difference for the standard key set, but 10K key set is actually slower. Standard:
Benchmark 1: timeout -v 300 ./calculate_average_artsiomkorzun.sh 2>&1
Time (mean ± σ): 1.924 s ± 0.015 s [User: 0.001 s, System: 0.004 s]
Range (min … max): 1.898 s … 1.935 s 5 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
artsiomkorzun: trimmed mean 1.9286819972533333, raw times 1.8975089989200002,1.9284043529200001,1.9350505949200003,1.9297026599200002,1.92793897892
Leaderboard
| # | Result (m:s.ms) | Implementation | JDK | Submitter | Notes |
|---|-----------------|--------------------|-----|---------------|-----------|
| | 00:01.928 | [link](https://github.com/gunnarmorling/1brc/blob/main/src/main/java/dev/morling/onebrc/CalculateAverage_artsiomkorzun.java)| 21.0.2-graal | [Artsiom Korzun](https://github.com/artsiomkorzun) | GraalVM native binary, uses Unsafe |
10K:
Benchmark 1: timeout -v 300 ./calculate_average_artsiomkorzun.sh 2>&1
Time (mean ± σ): 3.179 s ± 0.022 s [User: 0.001 s, System: 0.005 s]
Range (min … max): 3.142 s … 3.198 s 5 runs
Summary
artsiomkorzun: trimmed mean 3.1842617462133336, raw times 3.14171556288,3.19848109988,3.18802885388,3.1825212398800002,3.18223514488
Leaderboard
| # | Result (m:s.ms) | Implementation | JDK | Submitter | Notes |
|---|-----------------|--------------------|-----|---------------|-----------|
| | 00:03.184 | [link](https://github.com/gunnarmorling/1brc/blob/main/src/main/java/dev/morling/onebrc/CalculateAverage_artsiomkorzun.java)| 21.0.2-graal | [Artsiom Korzun](https://github.com/artsiomkorzun) | GraalVM native binary, uses Unsafe |
@gunnarmorling what time is for the current entry 29 Jan? I only see the time for 27 Jan.
@gunnarmorling can you please test 413 case one more time? I picked up the compiler's setting from Thomas. Maybe something will change.
Yes, looking much better now:
Benchmark 1: timeout -v 300 ./calculate_average_artsiomkorzun.sh 2>&1
Time (mean ± σ): 1.846 s ± 0.018 s [User: 0.002 s, System: 0.003 s]
Range (min … max): 1.815 s … 1.863 s 5 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
artsiomkorzun: trimmed mean 1.8510420830066667, raw times 1.8148779293400001,1.8510192823400002,1.84887277834,1.86337728234,1.85323418834
Leaderboard
| # | Result (m:s.ms) | Implementation | JDK | Submitter | Notes |
|---|-----------------|--------------------|-----|---------------|-----------|
| | 00:01.851 | [link](https://github.com/gunnarmorling/1brc/blob/main/src/main/java/dev/morling/onebrc/CalculateAverage_artsiomkorzun.java)| 21.0.2-graal | [Artsiom Korzun](https://github.com/artsiomkorzun) | GraalVM native binary, uses Unsafe |
Also back to the previous level for 10K.
Check List:
[x] You have run
./mvnw verify
and the project builds successfully[x] Tests pass (
./test.sh <username>
shows no differences between expected and actual outputs)[x] All formatting changes by the build are committed
[x] Your launch script is named
calculate_average_<username>.sh
(make sure to match casing of your GH user name) and is executable[x] Output matches that of
calculate_average_baseline.sh
[x] For new entries, or after substantial changes: When implementing custom hash structures, please point to where you deal with hash collisions (line number)
Execution time: -
Execution time of reference implementation: -
Reverting map structure to restore 10K performance.