This is an automated email from the ASF dual-hosted git repository.
yangjie01 pushed a commit to branch branch-3.5
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/branch-3.5 by this push:
new 2c3500c71bf8 [SPARK-51549][BUILD][3.5] Bump Parquet 1.15.1
2c3500c71bf8 is described below
commit 2c3500c71bf8ed348a7acbfedfe40ce4d8b0309b
Author: yumw...@ebay.com <yumw...@ebay.com>
AuthorDate: Wed Apr 9 12:43:38 2025 +0800
[SPARK-51549][BUILD][3.5] Bump Parquet 1.15.1
### What changes were proposed in this pull request?
Bump Parquet to 1.15.1.
### Why are the changes needed?
To fix critical CVE: https://www.cve.org/CVERecord?id=CVE-2025-30065
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Pass GHA.
### Was this patch authored or co-authored using generative AI tooling?
No.
Closes #50528 from wangyum/parquet-branch-3.5.
Lead-authored-by: yumw...@ebay.com <yumw...@ebay.com>
Co-authored-by: Fokko <fo...@apache.org>
Co-authored-by: Fokko Driesprong <fo...@tabular.io>
Co-authored-by: panbingkun <panbing...@baidu.com>
Co-authored-by: Fokko Driesprong <fo...@apache.org>
Co-authored-by: Cheng Pan <cheng...@apache.org>
Signed-off-by: yangjie01 <yangji...@baidu.com>
---
dev/deps/spark-deps-hadoop-3-hive-2.3 | 12 +-
pom.xml | 8 +-
.../BuiltInDataSourceWriteBenchmark-results.txt | 70 +--
.../benchmarks/DataSourceReadBenchmark-results.txt | 634 ++++++++++-----------
.../spark/sql/InjectRuntimeFilterSuite.scala | 4 +-
.../parquet/ParquetVectorizedSuite.scala | 2 +-
.../apache/spark/sql/hive/StatisticsSuite.scala | 2 +-
7 files changed, 369 insertions(+), 363 deletions(-)
diff --git a/dev/deps/spark-deps-hadoop-3-hive-2.3
b/dev/deps/spark-deps-hadoop-3-hive-2.3
index 4feea62dfe64..891833f1fdf5 100644
--- a/dev/deps/spark-deps-hadoop-3-hive-2.3
+++ b/dev/deps/spark-deps-hadoop-3-hive-2.3
@@ -218,12 +218,12 @@ orc-shims/1.9.5//orc-shims-1.9.5.jar
oro/2.0.8//oro-2.0.8.jar
osgi-resource-locator/1.0.3//osgi-resource-locator-1.0.3.jar
paranamer/2.8//paranamer-2.8.jar
-parquet-column/1.13.1//parquet-column-1.13.1.jar
-parquet-common/1.13.1//parquet-common-1.13.1.jar
-parquet-encoding/1.13.1//parquet-encoding-1.13.1.jar
-parquet-format-structures/1.13.1//parquet-format-structures-1.13.1.jar
-parquet-hadoop/1.13.1//parquet-hadoop-1.13.1.jar
-parquet-jackson/1.13.1//parquet-jackson-1.13.1.jar
+parquet-column/1.15.1//parquet-column-1.15.1.jar
+parquet-common/1.15.1//parquet-common-1.15.1.jar
+parquet-encoding/1.15.1//parquet-encoding-1.15.1.jar
+parquet-format-structures/1.15.1//parquet-format-structures-1.15.1.jar
+parquet-hadoop/1.15.1//parquet-hadoop-1.15.1.jar
+parquet-jackson/1.15.1//parquet-jackson-1.15.1.jar
pickle/1.3//pickle-1.3.jar
py4j/0.10.9.7//py4j-0.10.9.7.jar
remotetea-oncrpc/1.1.2//remotetea-oncrpc-1.1.2.jar
diff --git a/pom.xml b/pom.xml
index 33742c2ba95e..f33baa04f500 100644
--- a/pom.xml
+++ b/pom.xml
@@ -140,7 +140,7 @@
<kafka.version>3.4.1</kafka.version>
<!-- After 10.15.1.3, the minimum required version is JDK9 -->
<derby.version>10.14.2.0</derby.version>
- <parquet.version>1.13.1</parquet.version>
+ <parquet.version>1.15.1</parquet.version>
<orc.version>1.9.5</orc.version>
<orc.classifier>shaded-protobuf</orc.classifier>
<jetty.version>9.4.56.v20240826</jetty.version>
@@ -2663,6 +2663,12 @@
<version>${parquet.version}</version>
<scope>${parquet.test.deps.scope}</scope>
<classifier>tests</classifier>
+ <exclusions>
+ <exclusion>
+ <groupId>com.h2database</groupId>
+ <artifactId>h2</artifactId>
+ </exclusion>
+ </exclusions>
</dependency>
<dependency>
<groupId>org.apache.parquet</groupId>
diff --git a/sql/core/benchmarks/BuiltInDataSourceWriteBenchmark-results.txt
b/sql/core/benchmarks/BuiltInDataSourceWriteBenchmark-results.txt
index 4863a737785d..135f96d728a1 100644
--- a/sql/core/benchmarks/BuiltInDataSourceWriteBenchmark-results.txt
+++ b/sql/core/benchmarks/BuiltInDataSourceWriteBenchmark-results.txt
@@ -2,69 +2,69 @@
Parquet writer benchmark
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet(PARQUET_1_0) writer benchmark: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-Output Single Int Column 2724 2758
49 5.8 173.2 1.0X
-Output Single Double Column 2816 2829
20 5.6 179.0 1.0X
-Output Int and String Column 8999 9080
115 1.7 572.1 0.3X
-Output Partitions 5003 5086
117 3.1 318.1 0.5X
-Output Buckets 6911 6956
64 2.3 439.4 0.4X
+Output Single Int Column 1685 1742
81 9.3 107.1 1.0X
+Output Single Double Column 1675 1774
139 9.4 106.5 1.0X
+Output Int and String Column 5038 5126
125 3.1 320.3 0.3X
+Output Partitions 2904 2927
33 5.4 184.6 0.6X
+Output Buckets 4051 4058
10 3.9 257.6 0.4X
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet(PARQUET_2_0) writer benchmark: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-Output Single Int Column 2761 2806
64 5.7 175.5 1.0X
-Output Single Double Column 2652 2678
37 5.9 168.6 1.0X
-Output Int and String Column 8377 8518
199 1.9 532.6 0.3X
-Output Partitions 4865 4914
70 3.2 309.3 0.6X
-Output Buckets 6622 6664
59 2.4 421.0 0.4X
+Output Single Int Column 1545 1551
9 10.2 98.2 1.0X
+Output Single Double Column 1605 1629
34 9.8 102.0 1.0X
+Output Int and String Column 5077 5107
42 3.1 322.8 0.3X
+Output Partitions 2819 2822
3 5.6 179.2 0.5X
+Output Buckets 3911 3911
0 4.0 248.7 0.4X
================================================================================================
ORC writer benchmark
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
ORC writer benchmark: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-Output Single Int Column 1575 1627
74 10.0 100.1 1.0X
-Output Single Double Column 2021 2087
94 7.8 128.5 0.8X
-Output Int and String Column 6533 6800
377 2.4 415.4 0.2X
-Output Partitions 3577 3635
82 4.4 227.4 0.4X
-Output Buckets 4895 4923
41 3.2 311.2 0.3X
+Output Single Int Column 944 974
32 16.7 60.0 1.0X
+Output Single Double Column 1514 1518
6 10.4 96.3 0.6X
+Output Int and String Column 4797 4801
6 3.3 305.0 0.2X
+Output Partitions 2270 2272
3 6.9 144.3 0.4X
+Output Buckets 3201 3222
30 4.9 203.5 0.3X
================================================================================================
JSON writer benchmark
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
JSON writer benchmark: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-Output Single Int Column 2415 2465
71 6.5 153.6 1.0X
-Output Single Double Column 3690 3856
236 4.3 234.6 0.7X
-Output Int and String Column 6922 6930
12 2.3 440.1 0.3X
-Output Partitions 4619 4622
4 3.4 293.7 0.5X
-Output Buckets 6674 6756
116 2.4 424.3 0.4X
+Output Single Int Column 1659 1671
17 9.5 105.4 1.0X
+Output Single Double Column 2260 2262
4 7.0 143.7 0.7X
+Output Int and String Column 4963 4964
2 3.2 315.5 0.3X
+Output Partitions 2912 2915
3 5.4 185.2 0.6X
+Output Buckets 3868 3870
3 4.1 245.9 0.4X
================================================================================================
CSV writer benchmark
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
CSV writer benchmark: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-Output Single Int Column 4276 4368
130 3.7 271.8 1.0X
-Output Single Double Column 5273 5346
104 3.0 335.2 0.8X
-Output Int and String Column 8999 9139
199 1.7 572.1 0.5X
-Output Partitions 6466 6526
85 2.4 411.1 0.7X
-Output Buckets 8844 8878
48 1.8 562.3 0.5X
+Output Single Int Column 2603 2606
4 6.0 165.5 1.0X
+Output Single Double Column 2887 2888
1 5.4 183.6 0.9X
+Output Int and String Column 6464 6492
40 2.4 411.0 0.4X
+Output Partitions 3844 3896
73 4.1 244.4 0.7X
+Output Buckets 5662 5671
13 2.8 360.0 0.5X
diff --git a/sql/core/benchmarks/DataSourceReadBenchmark-results.txt
b/sql/core/benchmarks/DataSourceReadBenchmark-results.txt
index ceed213ef85c..d60a04fb8bc3 100644
--- a/sql/core/benchmarks/DataSourceReadBenchmark-results.txt
+++ b/sql/core/benchmarks/DataSourceReadBenchmark-results.txt
@@ -2,430 +2,430 @@
SQL Single Numeric Column Scan
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single BOOLEAN Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 13143 13363
311 1.2 835.6 1.0X
-SQL Json 7721 7787
93 2.0 490.9 1.7X
-SQL Parquet Vectorized: DataPageV1 110 128
18 143.6 7.0 120.0X
-SQL Parquet Vectorized: DataPageV2 90 103
14 175.1 5.7 146.3X
-SQL Parquet MR: DataPageV1 1785 1810
35 8.8 113.5 7.4X
-SQL Parquet MR: DataPageV2 1554 1557
5 10.1 98.8 8.5X
-SQL ORC Vectorized 175 180
4 89.9 11.1 75.2X
-SQL ORC MR 1585 1604
27 9.9 100.8 8.3X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 10902 10941
56 1.4 693.1 1.0X
+SQL Json 9892 9929
51 1.6 628.9 1.1X
+SQL Parquet Vectorized: DataPageV1 74 83
10 211.7 4.7 146.8X
+SQL Parquet Vectorized: DataPageV2 56 63
5 279.4 3.6 193.7X
+SQL Parquet MR: DataPageV1 2684 2697
19 5.9 170.7 4.1X
+SQL Parquet MR: DataPageV2 2596 2611
22 6.1 165.1 4.2X
+SQL ORC Vectorized 108 112
4 146.3 6.8 101.4X
+SQL ORC MR 2510 2513
4 6.3 159.6 4.3X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet Reader Single BOOLEAN Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
-ParquetReader Vectorized: DataPageV1 116 117
2 136.1 7.3 1.0X
-ParquetReader Vectorized: DataPageV2 110 112
3 142.9 7.0 1.0X
-ParquetReader Vectorized -> Row: DataPageV1 45 46
1 347.0 2.9 2.5X
-ParquetReader Vectorized -> Row: DataPageV2 40 40
1 394.4 2.5 2.9X
+ParquetReader Vectorized: DataPageV1 78 80
2 200.5 5.0 1.0X
+ParquetReader Vectorized: DataPageV2 71 72
2 222.0 4.5 1.1X
+ParquetReader Vectorized -> Row: DataPageV1 31 31
1 512.6 2.0 2.6X
+ParquetReader Vectorized -> Row: DataPageV2 24 25
1 652.5 1.5 3.3X
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single TINYINT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 15823 15829
8 1.0 1006.0 1.0X
-SQL Json 8823 8824
1 1.8 560.9 1.8X
-SQL Parquet Vectorized: DataPageV1 142 149
7 110.9 9.0 111.5X
-SQL Parquet Vectorized: DataPageV2 140 145
11 112.1 8.9 112.8X
-SQL Parquet MR: DataPageV1 1965 1979
20 8.0 124.9 8.1X
-SQL Parquet MR: DataPageV2 1833 1837
7 8.6 116.5 8.6X
-SQL ORC Vectorized 147 153
7 106.8 9.4 107.4X
-SQL ORC MR 1437 1438
2 10.9 91.3 11.0X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 12331 12414
118 1.3 784.0 1.0X
+SQL Json 10932 10933
1 1.4 695.1 1.1X
+SQL Parquet Vectorized: DataPageV1 86 93
5 183.6 5.4 143.9X
+SQL Parquet Vectorized: DataPageV2 85 91
7 185.0 5.4 145.1X
+SQL Parquet MR: DataPageV1 2714 2736
31 5.8 172.6 4.5X
+SQL Parquet MR: DataPageV2 2597 2605
12 6.1 165.1 4.7X
+SQL ORC Vectorized 94 99
7 168.1 6.0 131.7X
+SQL ORC MR 2546 2554
13 6.2 161.8 4.8X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet Reader Single TINYINT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
-ParquetReader Vectorized: DataPageV1 194 197
6 81.0 12.3 1.0X
-ParquetReader Vectorized: DataPageV2 194 196
3 80.9 12.4 1.0X
-ParquetReader Vectorized -> Row: DataPageV1 183 183
0 86.2 11.6 1.1X
-ParquetReader Vectorized -> Row: DataPageV2 182 183
0 86.5 11.6 1.1X
+ParquetReader Vectorized: DataPageV1 121 122
2 130.4 7.7 1.0X
+ParquetReader Vectorized: DataPageV2 121 122
2 130.4 7.7 1.0X
+ParquetReader Vectorized -> Row: DataPageV1 112 113
2 140.0 7.1 1.1X
+ParquetReader Vectorized -> Row: DataPageV2 112 114
2 139.9 7.1 1.1X
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single SMALLINT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 16581 16592
16 0.9 1054.2 1.0X
-SQL Json 9305 9308
5 1.7 591.6 1.8X
-SQL Parquet Vectorized: DataPageV1 200 227
68 78.8 12.7 83.1X
-SQL Parquet Vectorized: DataPageV2 179 187
11 87.7 11.4 92.5X
-SQL Parquet MR: DataPageV1 2270 2282
18 6.9 144.3 7.3X
-SQL Parquet MR: DataPageV2 1945 1947
3 8.1 123.7 8.5X
-SQL ORC Vectorized 176 180
3 89.2 11.2 94.0X
-SQL ORC MR 1647 1649
3 9.6 104.7 10.1X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 13206 13214
11 1.2 839.6 1.0X
+SQL Json 11219 11241
31 1.4 713.3 1.2X
+SQL Parquet Vectorized: DataPageV1 135 152
34 116.2 8.6 97.6X
+SQL Parquet Vectorized: DataPageV2 131 136
4 120.2 8.3 100.9X
+SQL Parquet MR: DataPageV1 3004 3019
22 5.2 191.0 4.4X
+SQL Parquet MR: DataPageV2 2737 2742
8 5.7 174.0 4.8X
+SQL ORC Vectorized 123 125
3 127.9 7.8 107.4X
+SQL ORC MR 2720 2731
15 5.8 173.0 4.9X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet Reader Single SMALLINT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
-ParquetReader Vectorized: DataPageV1 244 247
5 64.5 15.5 1.0X
-ParquetReader Vectorized: DataPageV2 266 268
4 59.1 16.9 0.9X
-ParquetReader Vectorized -> Row: DataPageV1 229 231
5 68.8 14.5 1.1X
-ParquetReader Vectorized -> Row: DataPageV2 250 251
1 62.9 15.9 1.0X
+ParquetReader Vectorized: DataPageV1 154 158
4 102.0 9.8 1.0X
+ParquetReader Vectorized: DataPageV2 180 182
2 87.4 11.4 0.9X
+ParquetReader Vectorized -> Row: DataPageV1 154 156
3 102.0 9.8 1.0X
+ParquetReader Vectorized -> Row: DataPageV2 179 181
3 88.1 11.4 0.9X
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single INT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 18059 18090
44 0.9 1148.2 1.0X
-SQL Json 9790 9791
1 1.6 622.5 1.8X
-SQL Parquet Vectorized: DataPageV1 144 150
7 109.2 9.2 125.4X
-SQL Parquet Vectorized: DataPageV2 260 266
13 60.6 16.5 69.6X
-SQL Parquet MR: DataPageV1 2241 2263
31 7.0 142.5 8.1X
-SQL Parquet MR: DataPageV2 1984 1991
10 7.9 126.2 9.1X
-SQL ORC Vectorized 242 249
7 64.9 15.4 74.6X
-SQL ORC MR 1693 1700
9 9.3 107.7 10.7X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 14589 14630
58 1.1 927.6 1.0X
+SQL Json 11507 11510
5 1.4 731.6 1.3X
+SQL Parquet Vectorized: DataPageV1 99 103
4 158.7 6.3 147.2X
+SQL Parquet Vectorized: DataPageV2 173 178
4 90.8 11.0 84.2X
+SQL Parquet MR: DataPageV1 3114 3133
27 5.1 198.0 4.7X
+SQL Parquet MR: DataPageV2 2857 2875
26 5.5 181.6 5.1X
+SQL ORC Vectorized 163 167
6 96.3 10.4 89.3X
+SQL ORC MR 2602 2637
50 6.0 165.4 5.6X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet Reader Single INT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
-ParquetReader Vectorized: DataPageV1 247 253
9 63.7 15.7 1.0X
-ParquetReader Vectorized: DataPageV2 360 365
6 43.7 22.9 0.7X
-ParquetReader Vectorized -> Row: DataPageV1 212 220
9 74.1 13.5 1.2X
-ParquetReader Vectorized -> Row: DataPageV2 327 329
3 48.0 20.8 0.8X
+ParquetReader Vectorized: DataPageV1 161 163
3 97.9 10.2 1.0X
+ParquetReader Vectorized: DataPageV2 249 253
5 63.2 15.8 0.6X
+ParquetReader Vectorized -> Row: DataPageV1 140 143
3 112.7 8.9 1.2X
+ParquetReader Vectorized -> Row: DataPageV2 223 225
4 70.6 14.2 0.7X
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single BIGINT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 23621 23622
2 0.7 1501.8 1.0X
-SQL Json 12398 12402
5 1.3 788.3 1.9X
-SQL Parquet Vectorized: DataPageV1 219 226
10 71.8 13.9 107.9X
-SQL Parquet Vectorized: DataPageV2 379 385
9 41.5 24.1 62.4X
-SQL Parquet MR: DataPageV1 2319 2338
27 6.8 147.5 10.2X
-SQL Parquet MR: DataPageV2 2066 2079
19 7.6 131.4 11.4X
-SQL ORC Vectorized 298 341
93 52.8 19.0 79.2X
-SQL ORC MR 1844 1844
0 8.5 117.2 12.8X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 19336 19362
36 0.8 1229.3 1.0X
+SQL Json 12561 12562
1 1.3 798.6 1.5X
+SQL Parquet Vectorized: DataPageV1 135 149
38 116.7 8.6 143.4X
+SQL Parquet Vectorized: DataPageV2 263 268
3 59.9 16.7 73.6X
+SQL Parquet MR: DataPageV1 3362 3369
9 4.7 213.8 5.8X
+SQL Parquet MR: DataPageV2 3101 3101
0 5.1 197.2 6.2X
+SQL ORC Vectorized 201 205
4 78.2 12.8 96.1X
+SQL ORC MR 2685 2694
13 5.9 170.7 7.2X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet Reader Single BIGINT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
-ParquetReader Vectorized: DataPageV1 294 327
77 53.4 18.7 1.0X
-ParquetReader Vectorized: DataPageV2 471 479
15 33.4 30.0 0.6X
-ParquetReader Vectorized -> Row: DataPageV1 276 278
4 57.0 17.5 1.1X
-ParquetReader Vectorized -> Row: DataPageV2 454 460
11 34.6 28.9 0.6X
+ParquetReader Vectorized: DataPageV1 179 187
6 87.7 11.4 1.0X
+ParquetReader Vectorized: DataPageV2 320 327
7 49.2 20.3 0.6X
+ParquetReader Vectorized -> Row: DataPageV1 178 184
6 88.4 11.3 1.0X
+ParquetReader Vectorized -> Row: DataPageV2 314 323
7 50.0 20.0 0.6X
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single FLOAT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 19058 19073
21 0.8 1211.7 1.0X
-SQL Json 12557 12578
29 1.3 798.4 1.5X
-SQL Parquet Vectorized: DataPageV1 145 150
6 108.7 9.2 131.8X
-SQL Parquet Vectorized: DataPageV2 145 151
9 108.7 9.2 131.7X
-SQL Parquet MR: DataPageV1 2197 2199
3 7.2 139.7 8.7X
-SQL Parquet MR: DataPageV2 2051 2060
13 7.7 130.4 9.3X
-SQL ORC Vectorized 314 318
3 50.0 20.0 60.6X
-SQL ORC MR 1737 1742
6 9.1 110.5 11.0X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 15310 15378
96 1.0 973.4 1.0X
+SQL Json 13289 13289
0 1.2 844.9 1.2X
+SQL Parquet Vectorized: DataPageV1 94 97
4 167.3 6.0 162.9X
+SQL Parquet Vectorized: DataPageV2 93 97
4 168.4 5.9 163.9X
+SQL Parquet MR: DataPageV1 3260 3284
34 4.8 207.3 4.7X
+SQL Parquet MR: DataPageV2 3081 3081
0 5.1 195.9 5.0X
+SQL ORC Vectorized 232 241
9 67.9 14.7 66.1X
+SQL ORC MR 2768 2774
9 5.7 176.0 5.5X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet Reader Single FLOAT Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
-ParquetReader Vectorized: DataPageV1 241 244
6 65.4 15.3 1.0X
-ParquetReader Vectorized: DataPageV2 240 243
5 65.4 15.3 1.0X
-ParquetReader Vectorized -> Row: DataPageV1 213 214
3 74.0 13.5 1.1X
-ParquetReader Vectorized -> Row: DataPageV2 212 217
8 74.1 13.5 1.1X
+ParquetReader Vectorized: DataPageV1 138 141
3 114.1 8.8 1.0X
+ParquetReader Vectorized: DataPageV2 138 142
4 113.8 8.8 1.0X
+ParquetReader Vectorized -> Row: DataPageV1 137 139
2 114.6 8.7 1.0X
+ParquetReader Vectorized -> Row: DataPageV2 138 139
3 114.3 8.8 1.0X
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single DOUBLE Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 24573 24625
74 0.6 1562.3 1.0X
-SQL Json 16677 16680
5 0.9 1060.3 1.5X
-SQL Parquet Vectorized: DataPageV1 209 216
10 75.3 13.3 117.6X
-SQL Parquet Vectorized: DataPageV2 208 217
9 75.4 13.3 117.9X
-SQL Parquet MR: DataPageV1 2287 2303
23 6.9 145.4 10.7X
-SQL Parquet MR: DataPageV2 2153 2182
42 7.3 136.9 11.4X
-SQL ORC Vectorized 397 401
4 39.6 25.2 61.9X
-SQL ORC MR 1857 1875
25 8.5 118.1 13.2X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 20013 20059
65 0.8 1272.4 1.0X
+SQL Json 16777 16780
5 0.9 1066.6 1.2X
+SQL Parquet Vectorized: DataPageV1 128 133
6 122.8 8.1 156.3X
+SQL Parquet Vectorized: DataPageV2 128 134
6 122.8 8.1 156.3X
+SQL Parquet MR: DataPageV1 3431 3452
30 4.6 218.1 5.8X
+SQL Parquet MR: DataPageV2 3325 3340
21 4.7 211.4 6.0X
+SQL ORC Vectorized 303 312
7 52.0 19.2 66.1X
+SQL ORC MR 2917 2919
3 5.4 185.4 6.9X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Parquet Reader Single DOUBLE Column Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
-ParquetReader Vectorized: DataPageV1 291 299
12 54.0 18.5 1.0X
-ParquetReader Vectorized: DataPageV2 291 301
13 54.0 18.5 1.0X
-ParquetReader Vectorized -> Row: DataPageV1 274 278
5 57.3 17.4 1.1X
-ParquetReader Vectorized -> Row: DataPageV2 274 275
4 57.5 17.4 1.1X
+ParquetReader Vectorized: DataPageV1 179 185
7 87.9 11.4 1.0X
+ParquetReader Vectorized: DataPageV2 180 188
6 87.2 11.5 1.0X
+ParquetReader Vectorized -> Row: DataPageV1 180 185
6 87.5 11.4 1.0X
+ParquetReader Vectorized -> Row: DataPageV2 178 184
7 88.5 11.3 1.0X
================================================================================================
SQL Single Numeric Column Scan in Struct
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single TINYINT Column Scan in Struct: Best Time(ms)
Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
-SQL ORC MR 2294
2370 108 6.9 145.8 1.0X
-SQL ORC Vectorized (Nested Column Disabled) 2228
2236 10 7.1 141.7 1.0X
-SQL ORC Vectorized (Nested Column Enabled) 287
289 1 54.7 18.3 8.0X
-SQL Parquet MR: DataPageV1 2342
2352 14 6.7 148.9 1.0X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 2753
2758 7 5.7 175.0 0.8X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 304
309 8 51.7 19.3 7.5X
-SQL Parquet MR: DataPageV2 2216
2220 6 7.1 140.9 1.0X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 2625
2625 1 6.0 166.9 0.9X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 305
312 13 51.6 19.4 7.5X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL ORC MR 3085
3089 6 5.1 196.1 1.0X
+SQL ORC Vectorized (Nested Column Disabled) 3100
3109 13 5.1 197.1 1.0X
+SQL ORC Vectorized (Nested Column Enabled) 200
202 3 78.7 12.7 15.4X
+SQL Parquet MR: DataPageV1 3445
3454 13 4.6 219.0 0.9X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3929
3939 13 4.0 249.8 0.8X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 217
221 5 72.6 13.8 14.2X
+SQL Parquet MR: DataPageV2 3348
3362 20 4.7 212.9 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3802
3806 6 4.1 241.7 0.8X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 218
222 3 72.2 13.8 14.2X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single SMALLINT Column Scan in Struct: Best Time(ms)
Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
-SQL ORC MR 2292
2304 17 6.9 145.7 1.0X
-SQL ORC Vectorized (Nested Column Disabled) 2295
2306 16 6.9 145.9 1.0X
-SQL ORC Vectorized (Nested Column Enabled) 324
329 7 48.6 20.6 7.1X
-SQL Parquet MR: DataPageV1 2541
2547 9 6.2 161.5 0.9X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3086
3088 2 5.1 196.2 0.7X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 298
305 9 52.8 18.9 7.7X
-SQL Parquet MR: DataPageV2 2334
2339 8 6.7 148.4 1.0X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 2800
2803 4 5.6 178.0 0.8X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 416
419 3 37.8 26.5 5.5X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL ORC MR 3226
3263 52 4.9 205.1 1.0X
+SQL ORC Vectorized (Nested Column Disabled) 3305
3310 8 4.8 210.1 1.0X
+SQL ORC Vectorized (Nested Column Enabled) 238
241 4 66.2 15.1 13.6X
+SQL Parquet MR: DataPageV1 3631
3634 4 4.3 230.9 0.9X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4054
4067 18 3.9 257.8 0.8X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 218
223 6 72.1 13.9 14.8X
+SQL Parquet MR: DataPageV2 3401
3409 12 4.6 216.2 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3781
3797 21 4.2 240.4 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 309
313 5 50.9 19.6 10.4X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single INT Column Scan in Struct: Best Time(ms)
Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
-SQL ORC MR 2351
2364 19 6.7 149.5 1.0X
-SQL ORC Vectorized (Nested Column Disabled) 2344
2358 20 6.7 149.0 1.0X
-SQL ORC Vectorized (Nested Column Enabled) 402
406 4 39.1 25.6 5.8X
-SQL Parquet MR: DataPageV1 2572
2574 3 6.1 163.5 0.9X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3087
3088 2 5.1 196.3 0.8X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 282
292 15 55.7 17.9 8.3X
-SQL Parquet MR: DataPageV2 2390
2418 40 6.6 152.0 1.0X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 2846
2870 35 5.5 180.9 0.8X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 454
461 9 34.6 28.9 5.2X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL ORC MR 3290
3318 41 4.8 209.1 1.0X
+SQL ORC Vectorized (Nested Column Disabled) 3346
3411 92 4.7 212.8 1.0X
+SQL ORC Vectorized (Nested Column Enabled) 282
286 2 55.7 17.9 11.7X
+SQL Parquet MR: DataPageV1 3781
3858 110 4.2 240.4 0.9X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4204
4212 11 3.7 267.3 0.8X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 204
208 4 77.2 12.9 16.2X
+SQL Parquet MR: DataPageV2 3596
3596 1 4.4 228.6 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 4096
4099 4 3.8 260.4 0.8X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 329
336 4 47.7 20.9 10.0X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single BIGINT Column Scan in Struct: Best Time(ms)
Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
-SQL ORC MR 2470
2472 2 6.4 157.1 1.0X
-SQL ORC Vectorized (Nested Column Disabled) 2454
2462 12 6.4 156.0 1.0X
-SQL ORC Vectorized (Nested Column Enabled) 446
452 10 35.3 28.4 5.5X
-SQL Parquet MR: DataPageV1 2668
2679 15 5.9 169.7 0.9X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3169
3171 3 5.0 201.5 0.8X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 353
378 14 44.6 22.4 7.0X
-SQL Parquet MR: DataPageV2 2466
2474 11 6.4 156.8 1.0X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 2898
2898 1 5.4 184.2 0.9X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 565
570 6 27.8 36.0 4.4X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL ORC MR 3332
3356 34 4.7 211.8 1.0X
+SQL ORC Vectorized (Nested Column Disabled) 3396
3399 5 4.6 215.9 1.0X
+SQL ORC Vectorized (Nested Column Enabled) 324
334 6 48.5 20.6 10.3X
+SQL Parquet MR: DataPageV1 3811
3815 5 4.1 242.3 0.9X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4217
4219 2 3.7 268.1 0.8X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 238
245 9 66.2 15.1 14.0X
+SQL Parquet MR: DataPageV2 3598
3611 19 4.4 228.8 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3915
3917 3 4.0 248.9 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 394
402 6 39.9 25.1 8.4X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single FLOAT Column Scan in Struct: Best Time(ms)
Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
-SQL ORC MR 2435
2449 20 6.5 154.8 1.0X
-SQL ORC Vectorized (Nested Column Disabled) 2458
2467 13 6.4 156.3 1.0X
-SQL ORC Vectorized (Nested Column Enabled) 444
458 8 35.4 28.2 5.5X
-SQL Parquet MR: DataPageV1 2548
2640 130 6.2 162.0 1.0X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 2968
2971 5 5.3 188.7 0.8X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 276
295 28 57.0 17.6 8.8X
-SQL Parquet MR: DataPageV2 2402
2406 5 6.5 152.7 1.0X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 2811
2828 24 5.6 178.7 0.9X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 278
285 7 56.6 17.7 8.8X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL ORC MR 3399
3434 49 4.6 216.1 1.0X
+SQL ORC Vectorized (Nested Column Disabled) 3452
3461 13 4.6 219.5 1.0X
+SQL ORC Vectorized (Nested Column Enabled) 336
358 17 46.8 21.4 10.1X
+SQL Parquet MR: DataPageV1 3668
3675 10 4.3 233.2 0.9X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4022
4023 2 3.9 255.7 0.8X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 191
196 4 82.3 12.1 17.8X
+SQL Parquet MR: DataPageV2 3505
3513 10 4.5 222.9 1.0X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3782
3785 4 4.2 240.4 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 191
196 3 82.2 12.2 17.8X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Single DOUBLE Column Scan in Struct: Best Time(ms)
Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
-SQL ORC MR 2540
2542 3 6.2 161.5 1.0X
-SQL ORC Vectorized (Nested Column Disabled) 2531
2541 14 6.2 160.9 1.0X
-SQL ORC Vectorized (Nested Column Enabled) 542
546 5 29.0 34.4 4.7X
-SQL Parquet MR: DataPageV1 2643
2674 44 6.0 168.0 1.0X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3178
3195 23 4.9 202.1 0.8X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 348
353 7 45.1 22.1 7.3X
-SQL Parquet MR: DataPageV2 2525
2546 30 6.2 160.5 1.0X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3005
3009 5 5.2 191.0 0.8X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 344
356 14 45.8 21.9 7.4X
+SQL ORC MR 3454
3507 75 4.6 219.6 1.0X
+SQL ORC Vectorized (Nested Column Disabled) 3408
3484 107 4.6 216.7 1.0X
+SQL ORC Vectorized (Nested Column Enabled) 425
442 11 37.0 27.0 8.1X
+SQL Parquet MR: DataPageV1 3689
3698 12 4.3 234.6 0.9X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4190
4191 0 3.8 266.4 0.8X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 240
247 10 65.5 15.3 14.4X
+SQL Parquet MR: DataPageV2 3625
3626 2 4.3 230.4 1.0X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 4019
4034 21 3.9 255.5 0.9X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 239
244 4 65.7 15.2 14.4X
================================================================================================
SQL Nested Column Scan
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
SQL Nested Column Scan: Best Time(ms)
Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
-SQL ORC MR 23536
23843 176 0.0 22445.8 1.0X
-SQL ORC Vectorized (Nested Column Disabled) 23036
23334 127 0.0 21969.1 1.0X
-SQL ORC Vectorized (Nested Column Enabled) 8504
8623 122 0.1 8110.1 2.8X
-SQL Parquet MR: DataPageV1 13540
13645 122 0.1 12913.0 1.7X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 14310
14430 123 0.1 13647.3 1.6X
-SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 9081
9458 677 0.1 8660.8 2.6X
-SQL Parquet MR: DataPageV2 16024
16350 380 0.1 15281.4 1.5X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 16714
16847 111 0.1 15939.8 1.4X
-SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 7080
7782 744 0.1 6752.4 3.3X
+SQL ORC MR 16096
16236 120 0.1 15350.8 1.0X
+SQL ORC Vectorized (Nested Column Disabled) 16132
16214 46 0.1 15384.7 1.0X
+SQL ORC Vectorized (Nested Column Enabled) 7627
7715 95 0.1 7273.5 2.1X
+SQL Parquet MR: DataPageV1 9442
9586 94 0.1 9004.3 1.7X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 9959
10050 69 0.1 9498.1 1.6X
+SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 6237
6322 65 0.2 5948.5 2.6X
+SQL Parquet MR: DataPageV2 10874
10952 81 0.1 10370.4 1.5X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 11315
11411 86 0.1 10790.7 1.4X
+SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 5748
5833 74 0.2 5481.3 2.8X
================================================================================================
Int and String Scan
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Int and String Scan: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 17078 17133
79 0.6 1628.7 1.0X
-SQL Json 11708 11723
21 0.9 1116.5 1.5X
-SQL Parquet Vectorized: DataPageV1 2002 2002
1 5.2 190.9 8.5X
-SQL Parquet Vectorized: DataPageV2 2313 2325
16 4.5 220.6 7.4X
-SQL Parquet MR: DataPageV1 4157 4170
18 2.5 396.5 4.1X
-SQL Parquet MR: DataPageV2 4052 4067
22 2.6 386.4 4.2X
-SQL ORC Vectorized 1971 1989
25 5.3 188.0 8.7X
-SQL ORC MR 3646 3648
3 2.9 347.7 4.7X
+SQL CSV 14236 14257
30 0.7 1357.7 1.0X
+SQL Json 12705 12713
12 0.8 1211.7 1.1X
+SQL Parquet Vectorized: DataPageV1 1697 1717
28 6.2 161.9 8.4X
+SQL Parquet Vectorized: DataPageV2 1866 1874
11 5.6 178.0 7.6X
+SQL Parquet MR: DataPageV1 4766 4773
9 2.2 454.6 3.0X
+SQL Parquet MR: DataPageV2 4695 4699
6 2.2 447.7 3.0X
+SQL ORC Vectorized 1618 1622
6 6.5 154.3 8.8X
+SQL ORC MR 4264 4295
43 2.5 406.7 3.3X
================================================================================================
Repeated String Scan
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Repeated String: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 9526 9547
30 1.1 908.5 1.0X
-SQL Json 6867 6883
23 1.5 654.9 1.4X
-SQL Parquet Vectorized: DataPageV1 728 738
15 14.4 69.4 13.1X
-SQL Parquet Vectorized: DataPageV2 702 714
12 14.9 67.0 13.6X
-SQL Parquet MR: DataPageV1 1877 1887
14 5.6 179.1 5.1X
-SQL Parquet MR: DataPageV2 1821 1827
8 5.8 173.7 5.2X
-SQL ORC Vectorized 422 426
4 24.9 40.2 22.6X
-SQL ORC MR 1838 1849
15 5.7 175.3 5.2X
+SQL CSV 7618 7632
20 1.4 726.5 1.0X
+SQL Json 8269 8279
14 1.3 788.6 0.9X
+SQL Parquet Vectorized: DataPageV1 535 541
7 19.6 51.1 14.2X
+SQL Parquet Vectorized: DataPageV2 540 544
7 19.4 51.5 14.1X
+SQL Parquet MR: DataPageV1 2437 2446
12 4.3 232.4 3.1X
+SQL Parquet MR: DataPageV2 2403 2407
6 4.4 229.2 3.2X
+SQL ORC Vectorized 335 350
16 31.3 32.0 22.7X
+SQL ORC MR 2492 2494
2 4.2 237.7 3.1X
================================================================================================
Partitioned Table Scan
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Partitioned Table: Best Time(ms) Avg
Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------------
-Data column - CSV 23701
23707 10 0.7 1506.9 1.0X
-Data column - Json 12457
12521 90 1.3 792.0 1.9X
-Data column - Parquet Vectorized: DataPageV1 209
219 11 75.3 13.3 113.5X
-Data column - Parquet Vectorized: DataPageV2 424
431 7 37.1 27.0 55.9X
-Data column - Parquet MR: DataPageV1 2711
2715 6 5.8 172.4 8.7X
-Data column - Parquet MR: DataPageV2 2467
2471 6 6.4 156.8 9.6X
-Data column - ORC Vectorized 299
306 8 52.6 19.0 79.3X
-Data column - ORC MR 2139
2146 9 7.4 136.0 11.1X
-Partition column - CSV 6516
6656 198 2.4 414.3 3.6X
-Partition column - Json 9845
9849 5 1.6 625.9 2.4X
-Partition column - Parquet Vectorized: DataPageV1 43
49 8 361.9 2.8 545.3X
-Partition column - Parquet Vectorized: DataPageV2 43
49 9 367.2 2.7 553.3X
-Partition column - Parquet MR: DataPageV1 1380
1389 14 11.4 87.7 17.2X
-Partition column - Parquet MR: DataPageV2 1374
1381 11 11.5 87.3 17.3X
-Partition column - ORC Vectorized 46
52 11 344.4 2.9 519.0X
-Partition column - ORC MR 1378
1378 0 11.4 87.6 17.2X
-Both columns - CSV 23758
23771 17 0.7 1510.5 1.0X
-Both columns - Json 13246
13293 67 1.2 842.1 1.8X
-Both columns - Parquet Vectorized: DataPageV1 248
261 16 63.3 15.8 95.4X
-Both columns - Parquet Vectorized: DataPageV2 469
480 12 33.5 29.8 50.5X
-Both columns - Parquet MR: DataPageV1 2779
2786 10 5.7 176.7 8.5X
-Both columns - Parquet MR: DataPageV2 2533
2548 21 6.2 161.1 9.4X
-Both columns - ORC Vectorized 338
340 3 46.5 21.5 70.1X
-Both columns - ORC MR 2210
2210 0 7.1 140.5 10.7X
+Data column - CSV 19445
19531 121 0.8 1236.3 1.0X
+Data column - Json 12628
12630 3 1.2 802.9 1.5X
+Data column - Parquet Vectorized: DataPageV1 130
134 4 120.8 8.3 149.4X
+Data column - Parquet Vectorized: DataPageV2 289
295 5 54.3 18.4 67.2X
+Data column - Parquet MR: DataPageV1 3652
3664 16 4.3 232.2 5.3X
+Data column - Parquet MR: DataPageV2 3400
3407 10 4.6 216.2 5.7X
+Data column - ORC Vectorized 206
210 4 76.2 13.1 94.3X
+Data column - ORC MR 3205
3373 238 4.9 203.8 6.1X
+Partition column - CSV 4973
4978 7 3.2 316.1 3.9X
+Partition column - Json 10793
10807 20 1.5 686.2 1.8X
+Partition column - Parquet Vectorized: DataPageV1 31
34 5 504.0 2.0 623.0X
+Partition column - Parquet Vectorized: DataPageV2 31
33 4 512.8 2.0 633.9X
+Partition column - Parquet MR: DataPageV1 2064
2068 5 7.6 131.2 9.4X
+Partition column - Parquet MR: DataPageV2 2073
2082 13 7.6 131.8 9.4X
+Partition column - ORC Vectorized 33
36 5 483.6 2.1 597.8X
+Partition column - ORC MR 2083
2090 9 7.6 132.4 9.3X
+Both columns - CSV 19572
19679 152 0.8 1244.3 1.0X
+Both columns - Json 14661
14689 39 1.1 932.1 1.3X
+Both columns - Parquet Vectorized: DataPageV1 146
159 12 107.8 9.3 133.3X
+Both columns - Parquet Vectorized: DataPageV2 308
316 10 51.0 19.6 63.1X
+Both columns - Parquet MR: DataPageV1 3684
3690 8 4.3 234.2 5.3X
+Both columns - Parquet MR: DataPageV2 3393
3409 23 4.6 215.7 5.7X
+Both columns - ORC Vectorized 225
234 9 70.0 14.3 86.5X
+Both columns - ORC MR 3141
3154 19 5.0 199.7 6.2X
================================================================================================
String with Nulls Scan
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
String with Nulls Scan (0.0%): Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 11462 11576
162 0.9 1093.1 1.0X
-SQL Json 10487 10489
3 1.0 1000.1 1.1X
-SQL Parquet Vectorized: DataPageV1 1321 1327
9 7.9 126.0 8.7X
-SQL Parquet Vectorized: DataPageV2 1689 1691
3 6.2 161.1 6.8X
-SQL Parquet MR: DataPageV1 3489 3505
22 3.0 332.8 3.3X
-SQL Parquet MR: DataPageV2 4243 4246
4 2.5 404.6 2.7X
-ParquetReader Vectorized: DataPageV1 959 964
6 10.9 91.5 11.9X
-ParquetReader Vectorized: DataPageV2 1341 1345
5 7.8 127.9 8.5X
-SQL ORC Vectorized 962 979
15 10.9 91.8 11.9X
-SQL ORC MR 3227 3241
20 3.2 307.7 3.6X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 9905 9935
42 1.1 944.6 1.0X
+SQL Json 13262 13269
10 0.8 1264.7 0.7X
+SQL Parquet Vectorized: DataPageV1 1062 1069
9 9.9 101.3 9.3X
+SQL Parquet Vectorized: DataPageV2 1363 1378
21 7.7 130.0 7.3X
+SQL Parquet MR: DataPageV1 4236 4237
2 2.5 403.9 2.3X
+SQL Parquet MR: DataPageV2 4773 4776
5 2.2 455.1 2.1X
+ParquetReader Vectorized: DataPageV1 738 741
3 14.2 70.4 13.4X
+ParquetReader Vectorized: DataPageV2 1000 1001
2 10.5 95.4 9.9X
+SQL ORC Vectorized 845 850
6 12.4 80.6 11.7X
+SQL ORC MR 3833 3850
24 2.7 365.5 2.6X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
String with Nulls Scan (50.0%): Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 8004 8008
6 1.3 763.3 1.0X
-SQL Json 7827 7843
22 1.3 746.5 1.0X
-SQL Parquet Vectorized: DataPageV1 1026 1038
17 10.2 97.8 7.8X
-SQL Parquet Vectorized: DataPageV2 1265 1276
15 8.3 120.7 6.3X
-SQL Parquet MR: DataPageV1 2738 2749
16 3.8 261.1 2.9X
-SQL Parquet MR: DataPageV2 3219 3227
12 3.3 306.9 2.5X
-ParquetReader Vectorized: DataPageV1 934 938
5 11.2 89.0 8.6X
-ParquetReader Vectorized: DataPageV2 1192 1196
6 8.8 113.7 6.7X
-SQL ORC Vectorized 1207 1207
1 8.7 115.1 6.6X
-SQL ORC MR 3020 3021
1 3.5 288.0 2.7X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 6270 6280
13 1.7 598.0 1.0X
+SQL Json 10908 10911
4 1.0 1040.3 0.6X
+SQL Parquet Vectorized: DataPageV1 799 801
3 13.1 76.2 7.8X
+SQL Parquet Vectorized: DataPageV2 921 933
11 11.4 87.8 6.8X
+SQL Parquet MR: DataPageV1 3460 3556
136 3.0 330.0 1.8X
+SQL Parquet MR: DataPageV2 3882 3899
23 2.7 370.2 1.6X
+ParquetReader Vectorized: DataPageV1 715 721
7 14.7 68.2 8.8X
+ParquetReader Vectorized: DataPageV2 849 858
9 12.3 81.0 7.4X
+SQL ORC Vectorized 925 930
5 11.3 88.2 6.8X
+SQL ORC MR 3654 3656
3 2.9 348.5 1.7X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
String with Nulls Scan (95.0%): Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 5075 5082
10 2.1 484.0 1.0X
-SQL Json 4602 4604
3 2.3 438.9 1.1X
-SQL Parquet Vectorized: DataPageV1 228 232
8 46.0 21.8 22.3X
-SQL Parquet Vectorized: DataPageV2 281 287
9 37.3 26.8 18.1X
-SQL Parquet MR: DataPageV1 1868 1875
10 5.6 178.1 2.7X
-SQL Parquet MR: DataPageV2 1798 1803
8 5.8 171.4 2.8X
-ParquetReader Vectorized: DataPageV1 241 242
2 43.6 22.9 21.1X
-ParquetReader Vectorized: DataPageV2 290 291
3 36.2 27.6 17.5X
-SQL ORC Vectorized 404 411
5 25.9 38.5 12.6X
-SQL ORC MR 1584 1585
2 6.6 151.1 3.2X
+SQL CSV 4086 4093
10 2.6 389.7 1.0X
+SQL Json 7907 7919
17 1.3 754.1 0.5X
+SQL Parquet Vectorized: DataPageV1 161 164
4 65.2 15.3 25.4X
+SQL Parquet Vectorized: DataPageV2 184 188
6 57.0 17.5 22.2X
+SQL Parquet MR: DataPageV1 2675 2677
2 3.9 255.2 1.5X
+SQL Parquet MR: DataPageV2 2688 2692
6 3.9 256.3 1.5X
+ParquetReader Vectorized: DataPageV1 169 170
2 62.0 16.1 24.1X
+ParquetReader Vectorized: DataPageV2 193 194
2 54.3 18.4 21.2X
+SQL ORC Vectorized 301 303
2 34.9 28.7 13.6X
+SQL ORC MR 2547 2550
4 4.1 242.9 1.6X
================================================================================================
Single Column Scan From Wide Columns
================================================================================================
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Single Column Scan from 10 columns: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 2596 2603
10 0.4 2475.4 1.0X
-SQL Json 2935 2961
36 0.4 2799.5 0.9X
-SQL Parquet Vectorized: DataPageV1 45 49
7 23.5 42.5 58.3X
-SQL Parquet Vectorized: DataPageV2 60 65
7 17.5 57.2 43.3X
-SQL Parquet MR: DataPageV1 200 207
8 5.3 190.4 13.0X
-SQL Parquet MR: DataPageV2 184 190
5 5.7 175.5 14.1X
-SQL ORC Vectorized 52 58
7 20.3 49.2 50.3X
-SQL ORC MR 155 159
4 6.8 147.7 16.8X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 1996 1998
3 0.5 1903.2 1.0X
+SQL Json 2477 2503
36 0.4 2362.4 0.8X
+SQL Parquet Vectorized: DataPageV1 29 34
6 35.8 28.0 68.1X
+SQL Parquet Vectorized: DataPageV2 40 42
4 26.4 37.9 50.3X
+SQL Parquet MR: DataPageV1 248 253
5 4.2 236.9 8.0X
+SQL Parquet MR: DataPageV2 230 235
7 4.6 219.1 8.7X
+SQL ORC Vectorized 35 39
6 29.8 33.5 56.8X
+SQL ORC MR 214 217
4 4.9 204.5 9.3X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Single Column Scan from 50 columns: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 7659 7670
15 0.1 7304.2 1.0X
-SQL Json 11990 12203
300 0.1 11434.9 0.6X
-SQL Parquet Vectorized: DataPageV1 63 67
6 16.7 59.8 122.1X
-SQL Parquet Vectorized: DataPageV2 75 80
8 13.9 71.9 101.6X
-SQL Parquet MR: DataPageV1 218 223
8 4.8 208.1 35.1X
-SQL Parquet MR: DataPageV2 205 211
9 5.1 195.2 37.4X
-SQL ORC Vectorized 67 73
12 15.7 63.8 114.5X
-SQL ORC MR 175 179
3 6.0 167.3 43.7X
-
-OpenJDK 64-Bit Server VM 1.8.0_362-b09 on Linux 5.15.0-1031-azure
-Intel(R) Xeon(R) Platinum 8272CL CPU @ 2.60GHz
+SQL CSV 5001 5011
13 0.2 4769.5 1.0X
+SQL Json 8590 8705
162 0.1 8192.0 0.6X
+SQL Parquet Vectorized: DataPageV1 39 44
7 26.8 37.3 127.8X
+SQL Parquet Vectorized: DataPageV2 50 55
8 21.1 47.3 100.8X
+SQL Parquet MR: DataPageV1 268 272
5 3.9 255.2 18.7X
+SQL Parquet MR: DataPageV2 246 252
6 4.3 234.2 20.4X
+SQL ORC Vectorized 47 50
5 22.3 44.8 106.6X
+SQL ORC MR 229 233
5 4.6 218.0 21.9X
+
+OpenJDK 64-Bit Server VM 1.8.0_442-b06 on Linux 6.8.0-1021-azure
+AMD EPYC 7763 64-Core Processor
Single Column Scan from 100 columns: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-SQL CSV 13640 13681
58 0.1 13008.1 1.0X
-SQL Json 22078 22212
189 0.0 21055.5 0.6X
-SQL Parquet Vectorized: DataPageV1 94 101
10 11.2 89.3 145.6X
-SQL Parquet Vectorized: DataPageV2 109 119
15 9.6 104.2 124.8X
-SQL Parquet MR: DataPageV1 255 266
15 4.1 242.9 53.6X
-SQL Parquet MR: DataPageV2 237 242
7 4.4 226.1 57.5X
-SQL ORC Vectorized 85 93
12 12.3 81.1 160.5X
-SQL ORC MR 198 204
7 5.3 188.8 68.9X
+SQL CSV 9001 9003
4 0.1 8583.9 1.0X
+SQL Json 16322 16468
206 0.1 15566.2 0.6X
+SQL Parquet Vectorized: DataPageV1 57 60
6 18.4 54.3 158.0X
+SQL Parquet Vectorized: DataPageV2 68 72
4 15.5 64.5 133.0X
+SQL Parquet MR: DataPageV1 288 295
8 3.6 274.4 31.3X
+SQL Parquet MR: DataPageV2 266 273
7 3.9 253.8 33.8X
+SQL ORC Vectorized 65 68
7 16.0 62.4 137.5X
+SQL ORC MR 238 241
5 4.4 226.5 37.9X
diff --git
a/sql/core/src/test/scala/org/apache/spark/sql/InjectRuntimeFilterSuite.scala
b/sql/core/src/test/scala/org/apache/spark/sql/InjectRuntimeFilterSuite.scala
index fedfd9ff587a..4d1795daa1fe 100644
---
a/sql/core/src/test/scala/org/apache/spark/sql/InjectRuntimeFilterSuite.scala
+++
b/sql/core/src/test/scala/org/apache/spark/sql/InjectRuntimeFilterSuite.scala
@@ -550,7 +550,7 @@ class InjectRuntimeFilterSuite extends QueryTest with
SQLTestUtils with SharedSp
withSQLConf(SQLConf.AUTO_BROADCASTJOIN_THRESHOLD.key -> "32",
SQLConf.RUNTIME_BLOOM_FILTER_CREATION_SIDE_THRESHOLD.key -> "4000") {
// Test that the max scan size rather than an individual scan size on
the filter
- // application side matters. `bf5filtered` has 14168 bytes and `bf2` has
3409 bytes.
+ // application side matters. `bf5filtered` has 15049 bytes and `bf2` has
3409 bytes.
withSQLConf(
SQLConf.RUNTIME_BLOOM_FILTER_APPLICATION_SIDE_SCAN_SIZE_THRESHOLD.key
-> "5000") {
assertRewroteWithBloomFilter("select * from " +
@@ -558,7 +558,7 @@ class InjectRuntimeFilterSuite extends QueryTest with
SQLTestUtils with SharedSp
"join bf3 on t.c5 = bf3.c3 where bf3.a3 = 5", 2)
}
withSQLConf(
- SQLConf.RUNTIME_BLOOM_FILTER_APPLICATION_SIDE_SCAN_SIZE_THRESHOLD.key
-> "15000") {
+ SQLConf.RUNTIME_BLOOM_FILTER_APPLICATION_SIDE_SCAN_SIZE_THRESHOLD.key
-> "16000") {
assertDidNotRewriteWithBloomFilter("select * from " +
"(select * from bf5filtered union all select * from bf2) t " +
"join bf3 on t.c5 = bf3.c3 where bf3.a3 = 5")
diff --git
a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorizedSuite.scala
b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorizedSuite.scala
index 123992a1a86b..e03dd22ed4e5 100644
---
a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorizedSuite.scala
+++
b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetVectorizedSuite.scala
@@ -501,7 +501,7 @@ class ParquetVectorizedSuite extends QueryTest with
ParquetTest with SharedSpark
val ty = parquetSchema.asGroupType().getType("a").asPrimitiveType()
val cd = new ColumnDescriptor(Seq("a").toArray, ty, 0, maxDef)
val repetitionLevels = Array.fill[Int](inputValues.length)(0)
- val definitionLevels = inputValues.map(v => if (v == null) 0 else 1)
+ val definitionLevels = inputValues.map(v => if (v == null) 0 else maxDef)
val memPageStore = new MemPageStore(expectedValues.length)
diff --git
a/sql/hive/src/test/scala/org/apache/spark/sql/hive/StatisticsSuite.scala
b/sql/hive/src/test/scala/org/apache/spark/sql/hive/StatisticsSuite.scala
index 507c482525c5..77bb68d2506f 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/StatisticsSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/StatisticsSuite.scala
@@ -1536,7 +1536,7 @@ class StatisticsSuite extends
StatisticsCollectionTestBase with TestHiveSingleto
Seq(tbl, ext_tbl).foreach { tblName =>
sql(s"INSERT INTO $tblName VALUES (1, 'a', '2019-12-13')")
- val expectedSize = 657
+ val expectedSize = 690
// analyze table
sql(s"ANALYZE TABLE $tblName COMPUTE STATISTICS NOSCAN")
var tableStats = getTableStats(tblName)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org
