ebf01ec3c1
### 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:  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>
146 lines
13 KiB
Plaintext
146 lines
13 KiB
Plaintext
================================================================================================
|
|
parsing large select expressions
|
|
================================================================================================
|
|
|
|
OpenJDK 64-Bit Server VM 11.0.10+9-LTS on Linux 5.4.0-1043-azure
|
|
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
|
|
parsing large select: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
|
|
------------------------------------------------------------------------------------------------------------------------
|
|
1 select expressions 1 2 1 0.0 1421891.0 1.0X
|
|
100 select expressions 9 11 2 0.0 8906443.0 0.2X
|
|
2500 select expressions 400 444 46 0.0 400078084.0 0.0X
|
|
|
|
|
|
================================================================================================
|
|
many column field read and write
|
|
================================================================================================
|
|
|
|
OpenJDK 64-Bit Server VM 11.0.10+9-LTS on Linux 5.4.0-1043-azure
|
|
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
|
|
many column field r/w: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
|
|
------------------------------------------------------------------------------------------------------------------------
|
|
1 cols x 100000 rows (read in-mem) 19 28 10 5.2 190.6 1.0X
|
|
1 cols x 100000 rows (exec in-mem) 23 40 17 4.4 226.2 0.8X
|
|
1 cols x 100000 rows (read parquet) 37 54 18 2.7 371.8 0.5X
|
|
1 cols x 100000 rows (write parquet) 116 166 62 0.9 1157.6 0.2X
|
|
100 cols x 1000 rows (read in-mem) 23 39 20 4.3 231.5 0.8X
|
|
100 cols x 1000 rows (exec in-mem) 35 56 24 2.8 352.5 0.5X
|
|
100 cols x 1000 rows (read parquet) 36 64 21 2.8 358.0 0.5X
|
|
100 cols x 1000 rows (write parquet) 121 148 36 0.8 1211.1 0.2X
|
|
2500 cols x 40 rows (read in-mem) 147 169 18 0.7 1474.8 0.1X
|
|
2500 cols x 40 rows (exec in-mem) 691 730 41 0.1 6912.5 0.0X
|
|
2500 cols x 40 rows (read parquet) 650 679 44 0.2 6495.8 0.0X
|
|
2500 cols x 40 rows (write parquet) 265 285 13 0.4 2651.2 0.1X
|
|
|
|
|
|
================================================================================================
|
|
wide shallowly nested struct field read and write
|
|
================================================================================================
|
|
|
|
OpenJDK 64-Bit Server VM 11.0.10+9-LTS on Linux 5.4.0-1043-azure
|
|
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
|
|
wide shallowly nested struct field r/w: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
|
|
------------------------------------------------------------------------------------------------------------------------
|
|
1 wide x 100000 rows (read in-mem) 29 47 16 3.4 290.5 1.0X
|
|
1 wide x 100000 rows (exec in-mem) 35 56 19 2.9 350.3 0.8X
|
|
1 wide x 100000 rows (read parquet) 57 82 28 1.8 570.1 0.5X
|
|
1 wide x 100000 rows (write parquet) 124 171 44 0.8 1239.2 0.2X
|
|
100 wide x 1000 rows (read in-mem) 25 37 13 4.1 245.8 1.2X
|
|
100 wide x 1000 rows (exec in-mem) 40 56 16 2.5 404.5 0.7X
|
|
100 wide x 1000 rows (read parquet) 37 62 16 2.7 368.6 0.8X
|
|
100 wide x 1000 rows (write parquet) 134 156 22 0.7 1341.0 0.2X
|
|
2500 wide x 40 rows (read in-mem) 34 49 17 3.0 338.1 0.9X
|
|
2500 wide x 40 rows (exec in-mem) 433 454 16 0.2 4334.6 0.1X
|
|
2500 wide x 40 rows (read parquet) 163 233 53 0.6 1634.9 0.2X
|
|
2500 wide x 40 rows (write parquet) 126 177 45 0.8 1262.7 0.2X
|
|
|
|
|
|
================================================================================================
|
|
deeply nested struct field read and write
|
|
================================================================================================
|
|
|
|
OpenJDK 64-Bit Server VM 11.0.10+9-LTS on Linux 5.4.0-1043-azure
|
|
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
|
|
deeply nested struct field r/w: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
|
|
------------------------------------------------------------------------------------------------------------------------
|
|
1 deep x 100000 rows (read in-mem) 27 40 15 3.7 273.6 1.0X
|
|
1 deep x 100000 rows (exec in-mem) 34 46 15 3.0 336.3 0.8X
|
|
1 deep x 100000 rows (read parquet) 53 74 17 1.9 527.0 0.5X
|
|
1 deep x 100000 rows (write parquet) 134 174 37 0.7 1339.3 0.2X
|
|
100 deep x 1000 rows (read in-mem) 87 99 9 1.1 870.2 0.3X
|
|
100 deep x 1000 rows (exec in-mem) 917 954 34 0.1 9171.4 0.0X
|
|
100 deep x 1000 rows (read parquet) 765 793 28 0.1 7645.5 0.0X
|
|
100 deep x 1000 rows (write parquet) 200 222 22 0.5 2003.2 0.1X
|
|
250 deep x 400 rows (read in-mem) 394 419 18 0.3 3939.3 0.1X
|
|
250 deep x 400 rows (exec in-mem) 6189 6210 29 0.0 61892.7 0.0X
|
|
250 deep x 400 rows (read parquet) 4454 4608 218 0.0 44539.5 0.0X
|
|
250 deep x 400 rows (write parquet) 448 455 4 0.2 4482.4 0.1X
|
|
|
|
|
|
================================================================================================
|
|
bushy struct field read and write
|
|
================================================================================================
|
|
|
|
OpenJDK 64-Bit Server VM 11.0.10+9-LTS on Linux 5.4.0-1043-azure
|
|
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
|
|
bushy struct field r/w: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
|
|
-------------------------------------------------------------------------------------------------------------------------
|
|
1 x 1 deep x 100000 rows (read in-mem) 22 29 9 4.6 217.1 1.0X
|
|
1 x 1 deep x 100000 rows (exec in-mem) 29 36 9 3.4 292.1 0.7X
|
|
1 x 1 deep x 100000 rows (read parquet) 30 39 11 3.3 301.9 0.7X
|
|
1 x 1 deep x 100000 rows (write parquet) 139 181 49 0.7 1388.8 0.2X
|
|
128 x 8 deep x 1000 rows (read in-mem) 27 36 11 3.7 267.1 0.8X
|
|
128 x 8 deep x 1000 rows (exec in-mem) 62 85 22 1.6 618.8 0.4X
|
|
128 x 8 deep x 1000 rows (read parquet) 42 62 19 2.4 422.3 0.5X
|
|
128 x 8 deep x 1000 rows (write parquet) 126 141 11 0.8 1258.9 0.2X
|
|
1024 x 11 deep x 100 rows (read in-mem) 31 44 12 3.2 313.7 0.7X
|
|
1024 x 11 deep x 100 rows (exec in-mem) 363 392 24 0.3 3628.3 0.1X
|
|
1024 x 11 deep x 100 rows (read parquet) 79 107 28 1.3 792.5 0.3X
|
|
1024 x 11 deep x 100 rows (write parquet) 142 160 17 0.7 1424.1 0.2X
|
|
|
|
|
|
================================================================================================
|
|
wide array field read and write
|
|
================================================================================================
|
|
|
|
OpenJDK 64-Bit Server VM 11.0.10+9-LTS on Linux 5.4.0-1043-azure
|
|
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
|
|
wide array field r/w: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
|
|
------------------------------------------------------------------------------------------------------------------------
|
|
1 wide x 100000 rows (read in-mem) 27 44 17 3.6 274.0 1.0X
|
|
1 wide x 100000 rows (exec in-mem) 31 45 17 3.3 305.4 0.9X
|
|
1 wide x 100000 rows (read parquet) 74 94 30 1.3 743.7 0.4X
|
|
1 wide x 100000 rows (write parquet) 127 156 29 0.8 1268.3 0.2X
|
|
100 wide x 1000 rows (read in-mem) 21 32 11 4.8 208.2 1.3X
|
|
100 wide x 1000 rows (exec in-mem) 26 37 12 3.8 260.0 1.1X
|
|
100 wide x 1000 rows (read parquet) 36 49 14 2.8 361.9 0.8X
|
|
100 wide x 1000 rows (write parquet) 128 172 42 0.8 1283.4 0.2X
|
|
2500 wide x 40 rows (read in-mem) 22 32 14 4.5 221.6 1.2X
|
|
2500 wide x 40 rows (exec in-mem) 26 37 13 3.9 257.0 1.1X
|
|
2500 wide x 40 rows (read parquet) 35 48 14 2.8 351.8 0.8X
|
|
2500 wide x 40 rows (write parquet) 124 142 16 0.8 1236.5 0.2X
|
|
|
|
|
|
================================================================================================
|
|
wide map field read and write
|
|
================================================================================================
|
|
|
|
OpenJDK 64-Bit Server VM 11.0.10+9-LTS on Linux 5.4.0-1043-azure
|
|
Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
|
|
wide map field r/w: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
|
|
------------------------------------------------------------------------------------------------------------------------
|
|
1 wide x 100000 rows (read in-mem) 22 35 15 4.5 220.3 1.0X
|
|
1 wide x 100000 rows (exec in-mem) 31 41 14 3.2 312.3 0.7X
|
|
1 wide x 100000 rows (read parquet) 103 116 19 1.0 1032.1 0.2X
|
|
1 wide x 100000 rows (write parquet) 128 156 37 0.8 1277.7 0.2X
|
|
100 wide x 1000 rows (read in-mem) 14 20 9 7.1 140.0 1.6X
|
|
100 wide x 1000 rows (exec in-mem) 20 25 9 5.1 197.8 1.1X
|
|
100 wide x 1000 rows (read parquet) 45 60 14 2.2 445.1 0.5X
|
|
100 wide x 1000 rows (write parquet) 116 138 17 0.9 1162.6 0.2X
|
|
2500 wide x 40 rows (read in-mem) 16 21 9 6.2 160.4 1.4X
|
|
2500 wide x 40 rows (exec in-mem) 21 30 12 4.8 210.0 1.0X
|
|
2500 wide x 40 rows (read parquet) 47 59 19 2.1 467.4 0.5X
|
|
2500 wide x 40 rows (write parquet) 119 125 7 0.8 1186.4 0.2X
|
|
|
|
|