Skip to content

Commit

Permalink
[SPARK-25492][TEST] Refactor WideSchemaBenchmark to use main method
Browse files Browse the repository at this point in the history 
## What changes were proposed in this pull request?

Refactor `WideSchemaBenchmark` to use main method.
1. use `spark-submit`:
```console
bin/spark-submit --class  org.apache.spark.sql.execution.benchmark.WideSchemaBenchmark --jars ./core/target/spark-core_2.11-3.0.0-SNAPSHOT-tests.jar ./sql/core/target/spark-sql_2.11-3.0.0-SNAPSHOT-tests.jar
```

2. Generate benchmark result:
```console
SPARK_GENERATE_BENCHMARK_FILES=1 build/sbt "sql/test:runMain org.apache.spark.sql.execution.benchmark.WideSchemaBenchmark"
```

## How was this patch tested?

manual tests

Closes apache#22501 from wangyum/SPARK-25492.

Lead-authored-by: Yuming Wang <[email protected]>
Co-authored-by: Dongjoon Hyun <[email protected]>
Signed-off-by: Dongjoon Hyun <[email protected]>
  • Loading branch information
@wangyum @dongjoon-hyun @jackylee-ch
2 people authored and jackylee-ch committed Feb 18, 2019
1 parent 2ae3301 commit 47d15e0
Show file tree
Hide file tree
Showing 2 changed files with 197 additions and 153 deletions.
216 changes: 122 additions & 94 deletions sql/core/benchmarks/WideSchemaBenchmark-results.txt
View file
Original file line number Diff line number Diff line change
@@ -1,117 +1,145 @@
Java HotSpot(TM) 64-Bit Server VM 1.8.0_92-b14 on Mac OS X 10.11.6
Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz
================================================================================================
parsing large select expressions
================================================================================================

OpenJDK 64-Bit Server VM 1.8.0_181-b13 on Linux 3.10.0-862.3.2.el7.x86_64
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
parsing large select: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
1 select expressions 2 / 4 0.0 2050147.0 1.0X
100 select expressions 6 / 7 0.0 6123412.0 0.3X
2500 select expressions 135 / 141 0.0 134623148.0 0.0X
1 select expressions 6 / 13 0.0 5997373.0 1.0X
100 select expressions 7 / 10 0.0 7204596.0 0.8X
2500 select expressions 103 / 107 0.0 102962705.0 0.1X

Java HotSpot(TM) 64-Bit Server VM 1.8.0_92-b14 on Mac OS X 10.11.6
Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz

================================================================================================
many column field read and write
================================================================================================

OpenJDK 64-Bit Server VM 1.8.0_181-b13 on Linux 3.10.0-862.3.2.el7.x86_64
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
many column field r/w: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
1 cols x 100000 rows (read in-mem) 16 / 18 6.3 158.6 1.0X
1 cols x 100000 rows (exec in-mem) 17 / 19 6.0 166.7 1.0X
1 cols x 100000 rows (read parquet) 24 / 26 4.3 235.1 0.7X
1 cols x 100000 rows (write parquet) 81 / 85 1.2 811.3 0.2X
100 cols x 1000 rows (read in-mem) 17 / 19 6.0 166.2 1.0X
100 cols x 1000 rows (exec in-mem) 25 / 27 4.0 249.2 0.6X
100 cols x 1000 rows (read parquet) 23 / 25 4.4 226.0 0.7X
100 cols x 1000 rows (write parquet) 83 / 87 1.2 831.0 0.2X
2500 cols x 40 rows (read in-mem) 132 / 137 0.8 1322.9 0.1X
2500 cols x 40 rows (exec in-mem) 326 / 330 0.3 3260.6 0.0X
2500 cols x 40 rows (read parquet) 831 / 839 0.1 8305.8 0.0X
2500 cols x 40 rows (write parquet) 237 / 245 0.4 2372.6 0.1X

Java HotSpot(TM) 64-Bit Server VM 1.8.0_92-b14 on Mac OS X 10.11.6
Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz
1 cols x 100000 rows (read in-mem) 40 / 51 2.5 396.5 1.0X
1 cols x 100000 rows (exec in-mem) 41 / 48 2.4 414.4 1.0X
1 cols x 100000 rows (read parquet) 61 / 70 1.6 610.2 0.6X
1 cols x 100000 rows (write parquet) 209 / 233 0.5 2086.1 0.2X
100 cols x 1000 rows (read in-mem) 43 / 49 2.3 433.8 0.9X
100 cols x 1000 rows (exec in-mem) 57 / 66 1.8 568.4 0.7X
100 cols x 1000 rows (read parquet) 60 / 66 1.7 599.0 0.7X
100 cols x 1000 rows (write parquet) 212 / 224 0.5 2120.6 0.2X
2500 cols x 40 rows (read in-mem) 268 / 275 0.4 2676.5 0.1X
2500 cols x 40 rows (exec in-mem) 494 / 504 0.2 4936.9 0.1X
2500 cols x 40 rows (read parquet) 132 / 139 0.8 1319.7 0.3X
2500 cols x 40 rows (write parquet) 371 / 381 0.3 3710.1 0.1X


================================================================================================
wide shallowly nested struct field read and write
================================================================================================

OpenJDK 64-Bit Server VM 1.8.0_181-b13 on Linux 3.10.0-862.3.2.el7.x86_64
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
wide shallowly nested struct field r/w: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
1 wide x 100000 rows (read in-mem) 15 / 17 6.6 151.0 1.0X
1 wide x 100000 rows (exec in-mem) 20 / 22 5.1 196.6 0.8X
1 wide x 100000 rows (read parquet) 59 / 63 1.7 592.8 0.3X
1 wide x 100000 rows (write parquet) 81 / 87 1.2 814.6 0.2X
100 wide x 1000 rows (read in-mem) 21 / 25 4.8 208.7 0.7X
100 wide x 1000 rows (exec in-mem) 72 / 81 1.4 718.5 0.2X
100 wide x 1000 rows (read parquet) 75 / 85 1.3 752.6 0.2X
100 wide x 1000 rows (write parquet) 88 / 95 1.1 876.7 0.2X
2500 wide x 40 rows (read in-mem) 28 / 34 3.5 282.2 0.5X
2500 wide x 40 rows (exec in-mem) 1269 / 1284 0.1 12688.1 0.0X
2500 wide x 40 rows (read parquet) 549 / 578 0.2 5493.4 0.0X
2500 wide x 40 rows (write parquet) 96 / 104 1.0 959.1 0.2X

Java HotSpot(TM) 64-Bit Server VM 1.8.0_92-b14 on Mac OS X 10.11.6
Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz
1 wide x 100000 rows (read in-mem) 37 / 43 2.7 373.6 1.0X
1 wide x 100000 rows (exec in-mem) 47 / 54 2.1 472.7 0.8X
1 wide x 100000 rows (read parquet) 132 / 145 0.8 1316.5 0.3X
1 wide x 100000 rows (write parquet) 205 / 232 0.5 2046.3 0.2X
100 wide x 1000 rows (read in-mem) 68 / 79 1.5 676.3 0.6X
100 wide x 1000 rows (exec in-mem) 88 / 97 1.1 882.2 0.4X
100 wide x 1000 rows (read parquet) 197 / 234 0.5 1971.8 0.2X
100 wide x 1000 rows (write parquet) 236 / 249 0.4 2359.6 0.2X
2500 wide x 40 rows (read in-mem) 77 / 85 1.3 768.0 0.5X
2500 wide x 40 rows (exec in-mem) 386 / 393 0.3 3855.2 0.1X
2500 wide x 40 rows (read parquet) 1741 / 1765 0.1 17408.3 0.0X
2500 wide x 40 rows (write parquet) 243 / 256 0.4 2425.2 0.2X


================================================================================================
deeply nested struct field read and write
================================================================================================

OpenJDK 64-Bit Server VM 1.8.0_181-b13 on Linux 3.10.0-862.3.2.el7.x86_64
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
deeply nested struct field r/w: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
1 deep x 100000 rows (read in-mem) 14 / 16 7.0 143.8 1.0X
1 deep x 100000 rows (exec in-mem) 17 / 19 5.9 169.7 0.8X
1 deep x 100000 rows (read parquet) 33 / 35 3.1 327.0 0.4X
1 deep x 100000 rows (write parquet) 79 / 84 1.3 786.9 0.2X
100 deep x 1000 rows (read in-mem) 21 / 24 4.7 211.3 0.7X
100 deep x 1000 rows (exec in-mem) 221 / 235 0.5 2214.5 0.1X
100 deep x 1000 rows (read parquet) 1928 / 1952 0.1 19277.1 0.0X
100 deep x 1000 rows (write parquet) 91 / 96 1.1 909.5 0.2X
250 deep x 400 rows (read in-mem) 57 / 61 1.8 567.1 0.3X
250 deep x 400 rows (exec in-mem) 1329 / 1385 0.1 13291.8 0.0X
250 deep x 400 rows (read parquet) 36563 / 36750 0.0 365630.2 0.0X
250 deep x 400 rows (write parquet) 126 / 130 0.8 1262.0 0.1X

Java HotSpot(TM) 64-Bit Server VM 1.8.0_92-b14 on Mac OS X 10.11.6
Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz
1 deep x 100000 rows (read in-mem) 35 / 42 2.9 350.2 1.0X
1 deep x 100000 rows (exec in-mem) 40 / 43 2.5 399.5 0.9X
1 deep x 100000 rows (read parquet) 69 / 73 1.4 691.6 0.5X
1 deep x 100000 rows (write parquet) 203 / 224 0.5 2025.9 0.2X
100 deep x 1000 rows (read in-mem) 70 / 75 1.4 703.7 0.5X
100 deep x 1000 rows (exec in-mem) 654 / 684 0.2 6539.9 0.1X
100 deep x 1000 rows (read parquet) 10503 / 10550 0.0 105030.5 0.0X
100 deep x 1000 rows (write parquet) 235 / 243 0.4 2353.2 0.1X
250 deep x 400 rows (read in-mem) 249 / 259 0.4 2492.6 0.1X
250 deep x 400 rows (exec in-mem) 3842 / 3854 0.0 38424.8 0.0X
250 deep x 400 rows (read parquet) 153080 / 153444 0.0 1530796.1 0.0X
250 deep x 400 rows (write parquet) 434 / 441 0.2 4344.6 0.1X


================================================================================================
bushy struct field read and write
================================================================================================

OpenJDK 64-Bit Server VM 1.8.0_181-b13 on Linux 3.10.0-862.3.2.el7.x86_64
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
bushy struct field r/w: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
1 x 1 deep x 100000 rows (read in-mem) 13 / 15 7.8 127.7 1.0X
1 x 1 deep x 100000 rows (exec in-mem) 15 / 17 6.6 151.5 0.8X
1 x 1 deep x 100000 rows (read parquet) 20 / 23 5.0 198.3 0.6X
1 x 1 deep x 100000 rows (write parquet) 77 / 82 1.3 770.4 0.2X
128 x 8 deep x 1000 rows (read in-mem) 12 / 14 8.2 122.5 1.0X
128 x 8 deep x 1000 rows (exec in-mem) 124 / 140 0.8 1241.2 0.1X
128 x 8 deep x 1000 rows (read parquet) 69 / 74 1.4 693.9 0.2X
128 x 8 deep x 1000 rows (write parquet) 78 / 83 1.3 777.7 0.2X
1024 x 11 deep x 100 rows (read in-mem) 25 / 29 4.1 246.1 0.5X
1024 x 11 deep x 100 rows (exec in-mem) 1197 / 1223 0.1 11974.6 0.0X
1024 x 11 deep x 100 rows (read parquet) 426 / 433 0.2 4263.7 0.0X
1024 x 11 deep x 100 rows (write parquet) 91 / 98 1.1 913.5 0.1X

Java HotSpot(TM) 64-Bit Server VM 1.8.0_92-b14 on Mac OS X 10.11.6
Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz
1 x 1 deep x 100000 rows (read in-mem) 37 / 42 2.7 370.2 1.0X
1 x 1 deep x 100000 rows (exec in-mem) 43 / 47 2.4 425.3 0.9X
1 x 1 deep x 100000 rows (read parquet) 48 / 51 2.1 478.7 0.8X
1 x 1 deep x 100000 rows (write parquet) 204 / 215 0.5 2042.0 0.2X
128 x 8 deep x 1000 rows (read in-mem) 32 / 37 3.1 318.6 1.2X
128 x 8 deep x 1000 rows (exec in-mem) 91 / 96 1.1 906.6 0.4X
128 x 8 deep x 1000 rows (read parquet) 351 / 379 0.3 3510.3 0.1X
128 x 8 deep x 1000 rows (write parquet) 199 / 203 0.5 1988.3 0.2X
1024 x 11 deep x 100 rows (read in-mem) 73 / 76 1.4 730.4 0.5X
1024 x 11 deep x 100 rows (exec in-mem) 327 / 334 0.3 3267.2 0.1X
1024 x 11 deep x 100 rows (read parquet) 2063 / 2078 0.0 20629.2 0.0X
1024 x 11 deep x 100 rows (write parquet) 248 / 266 0.4 2475.1 0.1X


================================================================================================
wide array field read and write
================================================================================================

OpenJDK 64-Bit Server VM 1.8.0_181-b13 on Linux 3.10.0-862.3.2.el7.x86_64
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
wide array field r/w: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
1 wide x 100000 rows (read in-mem) 14 / 16 7.0 143.2 1.0X
1 wide x 100000 rows (exec in-mem) 17 / 19 5.9 170.9 0.8X
1 wide x 100000 rows (read parquet) 43 / 46 2.3 434.1 0.3X
1 wide x 100000 rows (write parquet) 78 / 83 1.3 777.6 0.2X
100 wide x 1000 rows (read in-mem) 11 / 13 9.0 111.5 1.3X
100 wide x 1000 rows (exec in-mem) 13 / 15 7.8 128.3 1.1X
100 wide x 1000 rows (read parquet) 24 / 27 4.1 245.0 0.6X
100 wide x 1000 rows (write parquet) 74 / 80 1.4 740.5 0.2X
2500 wide x 40 rows (read in-mem) 11 / 13 9.1 109.5 1.3X
2500 wide x 40 rows (exec in-mem) 13 / 15 7.7 129.4 1.1X
2500 wide x 40 rows (read parquet) 24 / 26 4.1 241.3 0.6X
2500 wide x 40 rows (write parquet) 75 / 81 1.3 751.8 0.2X

Java HotSpot(TM) 64-Bit Server VM 1.8.0_92-b14 on Mac OS X 10.11.6
Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz
1 wide x 100000 rows (read in-mem) 33 / 38 3.0 328.4 1.0X
1 wide x 100000 rows (exec in-mem) 40 / 44 2.5 402.7 0.8X
1 wide x 100000 rows (read parquet) 83 / 91 1.2 826.6 0.4X
1 wide x 100000 rows (write parquet) 204 / 218 0.5 2039.1 0.2X
100 wide x 1000 rows (read in-mem) 28 / 31 3.6 277.2 1.2X
100 wide x 1000 rows (exec in-mem) 34 / 37 2.9 343.2 1.0X
100 wide x 1000 rows (read parquet) 56 / 61 1.8 556.4 0.6X
100 wide x 1000 rows (write parquet) 202 / 206 0.5 2017.3 0.2X
2500 wide x 40 rows (read in-mem) 29 / 30 3.5 286.4 1.1X
2500 wide x 40 rows (exec in-mem) 33 / 39 3.0 330.2 1.0X
2500 wide x 40 rows (read parquet) 54 / 66 1.8 544.0 0.6X
2500 wide x 40 rows (write parquet) 196 / 208 0.5 1959.2 0.2X


================================================================================================
wide map field read and write
================================================================================================

OpenJDK 64-Bit Server VM 1.8.0_181-b13 on Linux 3.10.0-862.3.2.el7.x86_64
Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
wide map field r/w: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
1 wide x 100000 rows (read in-mem) 16 / 18 6.2 162.6 1.0X
1 wide x 100000 rows (exec in-mem) 21 / 23 4.8 208.2 0.8X
1 wide x 100000 rows (read parquet) 54 / 59 1.8 543.6 0.3X
1 wide x 100000 rows (write parquet) 80 / 86 1.2 804.5 0.2X
100 wide x 1000 rows (read in-mem) 11 / 13 8.7 114.5 1.4X
100 wide x 1000 rows (exec in-mem) 14 / 16 7.0 143.5 1.1X
100 wide x 1000 rows (read parquet) 30 / 32 3.3 300.4 0.5X
100 wide x 1000 rows (write parquet) 75 / 80 1.3 749.9 0.2X
2500 wide x 40 rows (read in-mem) 13 / 15 7.8 128.1 1.3X
2500 wide x 40 rows (exec in-mem) 15 / 18 6.5 153.6 1.1X
2500 wide x 40 rows (read parquet) 30 / 33 3.3 304.4 0.5X
2500 wide x 40 rows (write parquet) 77 / 83 1.3 768.5 0.2X
1 wide x 100000 rows (read in-mem) 31 / 34 3.3 305.7 1.0X
1 wide x 100000 rows (exec in-mem) 39 / 44 2.6 390.0 0.8X
1 wide x 100000 rows (read parquet) 125 / 132 0.8 1250.5 0.2X
1 wide x 100000 rows (write parquet) 198 / 213 0.5 1979.9 0.2X
100 wide x 1000 rows (read in-mem) 21 / 23 4.7 212.7 1.4X
100 wide x 1000 rows (exec in-mem) 28 / 32 3.5 283.3 1.1X
100 wide x 1000 rows (read parquet) 68 / 73 1.5 683.0 0.4X
100 wide x 1000 rows (write parquet) 188 / 206 0.5 1882.1 0.2X
2500 wide x 40 rows (read in-mem) 25 / 28 4.0 252.2 1.2X
2500 wide x 40 rows (exec in-mem) 32 / 34 3.1 318.5 1.0X
2500 wide x 40 rows (read parquet) 69 / 73 1.4 691.5 0.4X
2500 wide x 40 rows (write parquet) 193 / 202 0.5 1932.8 0.2X


Loading

0 comments on commit 47d15e0

Please sign in to comment.