ODIn/spark-instrumented-optimizer
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
spark-instrumented-optimizer/sql/core/benchmarks/DateTimeRebaseBenchmark-results.txt
HyukjinKwon ebf01ec3c1 [SPARK-34950][TESTS] Update benchmark results to the ones created by GitHub Actions machines 
### What changes were proposed in this pull request?

https://github.com/apache/spark/pull/32015 added a way to run benchmarks much more easily in the same GitHub Actions build. This PR updates the benchmark results by using the way.

**NOTE** that looks like GitHub Actions use four types of CPU given my observations:

- Intel(R) Xeon(R) Platinum 8171M CPU  2.60GHz
- Intel(R) Xeon(R) CPU E5-2673 v4  2.30GHz
- Intel(R) Xeon(R) CPU E5-2673 v3  2.40GHz
- Intel(R) Xeon(R) Platinum 8272CL CPU  2.60GHz

Given my quick research, seems like they perform roughly similarly:

![Screen Shot 2021-04-03 at 9 31 23 PM](https://user-images.githubusercontent.com/6477701/113478478-f4b57b80-94c3-11eb-9047-f81ca8c59672.png)

I couldn't find enough information about Intel(R) Xeon(R) Platinum 8272CL CPU  2.60GHz but the performance seems roughly similar given the numbers.

So shouldn't be a big deal especially given that this way is much easier, encourages contributors to run more and guarantee the same number of cores and same memory with the same softwares.

### Why are the changes needed?

To have a base line of the benchmarks accordingly.

### Does this PR introduce _any_ user-facing change?

No, dev-only.

### How was this patch tested?

It was generated from:

- [Run benchmarks: * (JDK 11)](https://github.com/HyukjinKwon/spark/actions/runs/713575465)
- [Run benchmarks: * (JDK 8)](https://github.com/HyukjinKwon/spark/actions/runs/713154337)

Closes #32044 from HyukjinKwon/SPARK-34950.

Authored-by: HyukjinKwon <gurwls223@apache.org>
Signed-off-by: Max Gekk <max.gekk@gmail.com>
2021-04-03 23:02:56 +03:00

155 lines
14 KiB
Plaintext
Raw Blame History

================================================================================================
Rebasing dates/timestamps in Parquet datasource
================================================================================================
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Save DATE to parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1582, noop 24145 24145 0 4.1 241.4 1.0X
before 1582, noop 9863 9863 0 10.1 98.6 2.4X
after 1582, rebase EXCEPTION 36123 36123 0 2.8 361.2 0.7X
after 1582, rebase LEGACY 36800 36800 0 2.7 368.0 0.7X
after 1582, rebase CORRECTED 37083 37083 0 2.7 370.8 0.7X
before 1582, rebase LEGACY 27894 27894 0 3.6 278.9 0.9X
before 1582, rebase CORRECTED 22720 22720 0 4.4 227.2 1.1X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Load DATE from parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1582, vec off, rebase EXCEPTION 13975 14142 207 7.2 139.7 1.0X
after 1582, vec off, rebase LEGACY 14288 14464 165 7.0 142.9 1.0X
after 1582, vec off, rebase CORRECTED 14192 14398 332 7.0 141.9 1.0X
after 1582, vec on, rebase EXCEPTION 3418 3473 56 29.3 34.2 4.1X
after 1582, vec on, rebase LEGACY 3708 3794 91 27.0 37.1 3.8X
after 1582, vec on, rebase CORRECTED 3474 3542 67 28.8 34.7 4.0X
before 1582, vec off, rebase LEGACY 14162 14333 198 7.1 141.6 1.0X
before 1582, vec off, rebase CORRECTED 14201 14254 48 7.0 142.0 1.0X
before 1582, vec on, rebase LEGACY 4497 4538 70 22.2 45.0 3.1X
before 1582, vec on, rebase CORRECTED 3371 3420 46 29.7 33.7 4.1X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Save TIMESTAMP_INT96 to parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, noop 2359 2359 0 42.4 23.6 1.0X
before 1900, noop 2500 2500 0 40.0 25.0 0.9X
after 1900, rebase EXCEPTION 21442 21442 0 4.7 214.4 0.1X
after 1900, rebase LEGACY 21791 21791 0 4.6 217.9 0.1X
after 1900, rebase CORRECTED 22458 22458 0 4.5 224.6 0.1X
before 1900, rebase LEGACY 26749 26749 0 3.7 267.5 0.1X
before 1900, rebase CORRECTED 22908 22908 0 4.4 229.1 0.1X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Load TIMESTAMP_INT96 from parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, vec off, rebase EXCEPTION 18065 18166 91 5.5 180.7 1.0X
after 1900, vec off, rebase LEGACY 17358 17602 217 5.8 173.6 1.0X
after 1900, vec off, rebase CORRECTED 17900 18615 1194 5.6 179.0 1.0X
after 1900, vec on, rebase EXCEPTION 8235 8301 100 12.1 82.3 2.2X
after 1900, vec on, rebase LEGACY 8295 8476 225 12.1 83.0 2.2X
after 1900, vec on, rebase CORRECTED 8128 8224 98 12.3 81.3 2.2X
before 1900, vec off, rebase LEGACY 21413 21906 507 4.7 214.1 0.8X
before 1900, vec off, rebase CORRECTED 18217 18288 103 5.5 182.2 1.0X
before 1900, vec on, rebase LEGACY 12115 12260 138 8.3 121.2 1.5X
before 1900, vec on, rebase CORRECTED 8742 8772 42 11.4 87.4 2.1X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Save TIMESTAMP_MICROS to parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, noop 2556 2556 0 39.1 25.6 1.0X
before 1900, noop 2619 2619 0 38.2 26.2 1.0X
after 1900, rebase EXCEPTION 18377 18377 0 5.4 183.8 0.1X
after 1900, rebase LEGACY 18986 18986 0 5.3 189.9 0.1X
after 1900, rebase CORRECTED 20329 20329 0 4.9 203.3 0.1X
before 1900, rebase LEGACY 24590 24590 0 4.1 245.9 0.1X
before 1900, rebase CORRECTED 19737 19737 0 5.1 197.4 0.1X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Load TIMESTAMP_MICROS from parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, vec off, rebase EXCEPTION 16350 16622 272 6.1 163.5 1.0X
after 1900, vec off, rebase LEGACY 15887 16060 209 6.3 158.9 1.0X
after 1900, vec off, rebase CORRECTED 16156 16349 175 6.2 161.6 1.0X
after 1900, vec on, rebase EXCEPTION 4902 4922 32 20.4 49.0 3.3X
after 1900, vec on, rebase LEGACY 5245 5354 135 19.1 52.5 3.1X
after 1900, vec on, rebase CORRECTED 4905 5019 99 20.4 49.1 3.3X
before 1900, vec off, rebase LEGACY 19924 20162 240 5.0 199.2 0.8X
before 1900, vec off, rebase CORRECTED 16070 16180 162 6.2 160.7 1.0X
before 1900, vec on, rebase LEGACY 9381 9522 127 10.7 93.8 1.7X
before 1900, vec on, rebase CORRECTED 4942 4990 49 20.2 49.4 3.3X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Save TIMESTAMP_MILLIS to parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, noop 2583 2583 0 38.7 25.8 1.0X
before 1900, noop 2514 2514 0 39.8 25.1 1.0X
after 1900, rebase EXCEPTION 16857 16857 0 5.9 168.6 0.2X
after 1900, rebase LEGACY 17211 17211 0 5.8 172.1 0.2X
after 1900, rebase CORRECTED 18908 18908 0 5.3 189.1 0.1X
before 1900, rebase LEGACY 22901 22901 0 4.4 229.0 0.1X
before 1900, rebase CORRECTED 19047 19047 0 5.3 190.5 0.1X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Load TIMESTAMP_MILLIS from parquet: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, vec off, rebase EXCEPTION 16820 16922 116 5.9 168.2 1.0X
after 1900, vec off, rebase LEGACY 17201 17547 406 5.8 172.0 1.0X
after 1900, vec off, rebase CORRECTED 17087 17321 220 5.9 170.9 1.0X
after 1900, vec on, rebase EXCEPTION 6204 6237 55 16.1 62.0 2.7X
after 1900, vec on, rebase LEGACY 6853 6887 51 14.6 68.5 2.5X
after 1900, vec on, rebase CORRECTED 6388 6509 193 15.7 63.9 2.6X
before 1900, vec off, rebase LEGACY 20223 20522 454 4.9 202.2 0.8X
before 1900, vec off, rebase CORRECTED 16680 17002 330 6.0 166.8 1.0X
before 1900, vec on, rebase LEGACY 9893 10044 162 10.1 98.9 1.7X
before 1900, vec on, rebase CORRECTED 6381 6486 140 15.7 63.8 2.6X
================================================================================================
Rebasing dates/timestamps in ORC datasource
================================================================================================
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Save DATE to ORC: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1582, noop 24733 24733 0 4.0 247.3 1.0X
before 1582, noop 10658 10658 0 9.4 106.6 2.3X
after 1582 34442 34442 0 2.9 344.4 0.7X
before 1582 20847 20847 0 4.8 208.5 1.2X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Load DATE from ORC: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1582, vec off 10339 10498 180 9.7 103.4 1.0X
after 1582, vec on 3812 3825 18 26.2 38.1 2.7X
before 1582, vec off 11374 11531 150 8.8 113.7 0.9X
before 1582, vec on 4387 4641 424 22.8 43.9 2.4X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Save TIMESTAMP to ORC: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, noop 2709 2709 0 36.9 27.1 1.0X
before 1900, noop 2563 2563 0 39.0 25.6 1.1X
after 1900 18967 18967 0 5.3 189.7 0.1X
before 1900 21109 21109 0 4.7 211.1 0.1X
OpenJDK 64-Bit Server VM 1.8.0_282-b08 on Linux 5.4.0-1043-azure
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
Load TIMESTAMP from ORC: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
after 1900, vec off 11339 11393 86 8.8 113.4 1.0X
after 1900, vec on 5269 5298 47 19.0 52.7 2.2X
before 1900, vec off 14399 14551 241 6.9 144.0 0.8X
before 1900, vec on 8171 8266 97 12.2 81.7 1.4X
Reference in a new issue View git blame Copy permalink