blob: 6f52338a78df9e1a1a4fabca075e40511a8947c9 [file] [log] [blame]
================================================================================================
SQL Single Numeric Column Scan
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Single BOOLEAN Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 12773 12910 194 1.2 812.1 1.0X
SQL Json 9907 10034 178 1.6 629.9 1.3X
SQL Parquet Vectorized: DataPageV1 135 164 24 116.1 8.6 94.3X
SQL Parquet Vectorized: DataPageV2 100 110 8 157.6 6.3 127.9X
SQL Parquet MR: DataPageV1 2176 2196 29 7.2 138.3 5.9X
SQL Parquet MR: DataPageV2 1974 1995 30 8.0 125.5 6.5X
SQL ORC Vectorized 203 215 9 77.4 12.9 62.9X
SQL ORC MR 1897 1909 18 8.3 120.6 6.7X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Parquet Reader Single BOOLEAN Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
ParquetReader Vectorized: DataPageV1 78 79 2 201.6 5.0 1.0X
ParquetReader Vectorized: DataPageV2 54 56 2 291.5 3.4 1.4X
ParquetReader Vectorized -> Row: DataPageV1 58 61 4 273.1 3.7 1.4X
ParquetReader Vectorized -> Row: DataPageV2 34 36 2 459.2 2.2 2.3X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Single TINYINT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 15279 15317 53 1.0 971.4 1.0X
SQL Json 11724 11763 54 1.3 745.4 1.3X
SQL Parquet Vectorized: DataPageV1 136 149 10 115.5 8.7 112.2X
SQL Parquet Vectorized: DataPageV2 134 145 9 117.2 8.5 113.8X
SQL Parquet MR: DataPageV1 2340 2412 101 6.7 148.8 6.5X
SQL Parquet MR: DataPageV2 2356 2359 5 6.7 149.8 6.5X
SQL ORC Vectorized 200 216 11 78.5 12.7 76.2X
SQL ORC MR 1808 1814 9 8.7 114.9 8.5X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Parquet Reader Single TINYINT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
ParquetReader Vectorized: DataPageV1 134 141 5 117.3 8.5 1.0X
ParquetReader Vectorized: DataPageV2 131 140 7 119.7 8.4 1.0X
ParquetReader Vectorized -> Row: DataPageV1 83 89 5 188.4 5.3 1.6X
ParquetReader Vectorized -> Row: DataPageV2 83 88 4 188.4 5.3 1.6X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Single SMALLINT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 16066 16127 87 1.0 1021.4 1.0X
SQL Json 12056 12123 94 1.3 766.5 1.3X
SQL Parquet Vectorized: DataPageV1 159 174 9 98.7 10.1 100.8X
SQL Parquet Vectorized: DataPageV2 235 253 9 66.8 15.0 68.2X
SQL Parquet MR: DataPageV1 2563 2571 11 6.1 163.0 6.3X
SQL Parquet MR: DataPageV2 2457 2494 51 6.4 156.2 6.5X
SQL ORC Vectorized 239 257 15 65.9 15.2 67.3X
SQL ORC MR 2067 2071 4 7.6 131.4 7.8X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Parquet Reader Single SMALLINT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
ParquetReader Vectorized: DataPageV1 247 257 7 63.7 15.7 1.0X
ParquetReader Vectorized: DataPageV2 323 337 16 48.7 20.5 0.8X
ParquetReader Vectorized -> Row: DataPageV1 244 250 5 64.5 15.5 1.0X
ParquetReader Vectorized -> Row: DataPageV2 306 315 10 51.5 19.4 0.8X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Single INT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 17633 17752 168 0.9 1121.1 1.0X
SQL Json 12979 13019 57 1.2 825.2 1.4X
SQL Parquet Vectorized: DataPageV1 158 167 7 99.5 10.0 111.6X
SQL Parquet Vectorized: DataPageV2 335 346 12 46.9 21.3 52.6X
SQL Parquet MR: DataPageV1 2597 2601 6 6.1 165.1 6.8X
SQL Parquet MR: DataPageV2 2497 2529 45 6.3 158.7 7.1X
SQL ORC Vectorized 285 294 8 55.2 18.1 61.9X
SQL ORC MR 2022 2107 120 7.8 128.6 8.7X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Parquet Reader Single INT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
ParquetReader Vectorized: DataPageV1 243 265 15 64.7 15.4 1.0X
ParquetReader Vectorized: DataPageV2 416 420 5 37.8 26.5 0.6X
ParquetReader Vectorized -> Row: DataPageV1 282 288 6 55.8 17.9 0.9X
ParquetReader Vectorized -> Row: DataPageV2 445 456 11 35.4 28.3 0.5X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Single BIGINT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 23061 23236 247 0.7 1466.2 1.0X
SQL Json 16191 16295 147 1.0 1029.4 1.4X
SQL Parquet Vectorized: DataPageV1 230 241 8 68.5 14.6 100.5X
SQL Parquet Vectorized: DataPageV2 465 474 17 33.8 29.5 49.6X
SQL Parquet MR: DataPageV1 2863 2949 122 5.5 182.0 8.1X
SQL Parquet MR: DataPageV2 2556 2578 31 6.2 162.5 9.0X
SQL ORC Vectorized 400 421 25 39.3 25.4 57.6X
SQL ORC MR 2333 2352 26 6.7 148.4 9.9X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Parquet Reader Single BIGINT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
ParquetReader Vectorized: DataPageV1 343 349 7 45.8 21.8 1.0X
ParquetReader Vectorized: DataPageV2 577 590 24 27.2 36.7 0.6X
ParquetReader Vectorized -> Row: DataPageV1 390 402 14 40.3 24.8 0.9X
ParquetReader Vectorized -> Row: DataPageV2 620 634 10 25.4 39.4 0.6X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Single FLOAT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 18313 18370 81 0.9 1164.3 1.0X
SQL Json 15439 15522 118 1.0 981.6 1.2X
SQL Parquet Vectorized: DataPageV1 190 219 28 82.9 12.1 96.5X
SQL Parquet Vectorized: DataPageV2 166 197 21 94.6 10.6 110.2X
SQL Parquet MR: DataPageV1 2588 2667 111 6.1 164.5 7.1X
SQL Parquet MR: DataPageV2 2350 2438 124 6.7 149.4 7.8X
SQL ORC Vectorized 446 476 23 35.3 28.3 41.1X
SQL ORC MR 2280 2305 36 6.9 145.0 8.0X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Parquet Reader Single FLOAT Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
ParquetReader Vectorized: DataPageV1 287 291 4 54.8 18.2 1.0X
ParquetReader Vectorized: DataPageV2 271 289 9 57.9 17.3 1.1X
ParquetReader Vectorized -> Row: DataPageV1 260 274 11 60.5 16.5 1.1X
ParquetReader Vectorized -> Row: DataPageV2 258 274 15 61.0 16.4 1.1X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Single DOUBLE Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 23580 23698 167 0.7 1499.2 1.0X
SQL Json 20420 20476 79 0.8 1298.2 1.2X
SQL Parquet Vectorized: DataPageV1 259 297 51 60.6 16.5 90.9X
SQL Parquet Vectorized: DataPageV2 246 280 26 64.1 15.6 96.0X
SQL Parquet MR: DataPageV1 2743 2795 74 5.7 174.4 8.6X
SQL Parquet MR: DataPageV2 2495 2512 24 6.3 158.6 9.5X
SQL ORC Vectorized 505 554 34 31.1 32.1 46.7X
SQL ORC MR 2245 2269 33 7.0 142.7 10.5X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Parquet Reader Single DOUBLE Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------
ParquetReader Vectorized: DataPageV1 362 369 7 43.5 23.0 1.0X
ParquetReader Vectorized: DataPageV2 358 368 6 43.9 22.8 1.0X
ParquetReader Vectorized -> Row: DataPageV1 359 363 5 43.9 22.8 1.0X
ParquetReader Vectorized -> Row: DataPageV2 359 362 3 43.8 22.8 1.0X
================================================================================================
SQL Single Numeric Column Scan in Struct
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
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 2949 2960 15 5.3 187.5 1.0X
SQL ORC Vectorized (Nested Column Disabled) 2814 2861 67 5.6 178.9 1.0X
SQL ORC Vectorized (Nested Column Enabled) 316 337 33 49.8 20.1 9.3X
SQL Parquet MR: DataPageV1 2954 2986 44 5.3 187.8 1.0X
SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3668 3701 46 4.3 233.2 0.8X
SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 188 262 91 83.9 11.9 15.7X
SQL Parquet MR: DataPageV2 2726 2795 97 5.8 173.3 1.1X
SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3511 3551 56 4.5 223.2 0.8X
SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 185 199 10 84.9 11.8 15.9X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
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 2823 2929 150 5.6 179.5 1.0X
SQL ORC Vectorized (Nested Column Disabled) 2836 2895 83 5.5 180.3 1.0X
SQL ORC Vectorized (Nested Column Enabled) 465 519 48 33.8 29.5 6.1X
SQL Parquet MR: DataPageV1 3441 3457 22 4.6 218.8 0.8X
SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3948 4013 92 4.0 251.0 0.7X
SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 181 220 21 86.8 11.5 15.6X
SQL Parquet MR: DataPageV2 2928 2995 95 5.4 186.2 1.0X
SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3577 3621 62 4.4 227.4 0.8X
SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 432 459 31 36.4 27.4 6.5X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
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 3029 3032 4 5.2 192.6 1.0X
SQL ORC Vectorized (Nested Column Disabled) 2961 3007 64 5.3 188.3 1.0X
SQL ORC Vectorized (Nested Column Enabled) 488 527 30 32.2 31.0 6.2X
SQL Parquet MR: DataPageV1 3403 3434 45 4.6 216.3 0.9X
SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4088 4266 252 3.8 259.9 0.7X
SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 209 227 17 75.3 13.3 14.5X
SQL Parquet MR: DataPageV2 3189 3192 5 4.9 202.7 0.9X
SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3820 3843 32 4.1 242.9 0.8X
SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 505 528 17 31.2 32.1 6.0X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
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 3153 3203 69 5.0 200.5 1.0X
SQL ORC Vectorized (Nested Column Disabled) 3002 3023 29 5.2 190.9 1.1X
SQL ORC Vectorized (Nested Column Enabled) 626 651 27 25.1 39.8 5.0X
SQL Parquet MR: DataPageV1 3701 3705 5 4.2 235.3 0.9X
SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4168 4189 30 3.8 265.0 0.8X
SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 296 316 36 53.2 18.8 10.7X
SQL Parquet MR: DataPageV2 3076 3179 145 5.1 195.6 1.0X
SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3769 3869 141 4.2 239.6 0.8X
SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 485 500 13 32.4 30.9 6.5X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
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 3035 3113 111 5.2 192.9 1.0X
SQL ORC Vectorized (Nested Column Disabled) 3214 3249 49 4.9 204.4 0.9X
SQL ORC Vectorized (Nested Column Enabled) 706 713 7 22.3 44.9 4.3X
SQL Parquet MR: DataPageV1 3190 3225 50 4.9 202.8 1.0X
SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 3711 3795 119 4.2 235.9 0.8X
SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 167 179 7 94.0 10.6 18.1X
SQL Parquet MR: DataPageV2 2722 2725 4 5.8 173.1 1.1X
SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3357 3429 102 4.7 213.4 0.9X
SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 176 194 19 89.2 11.2 17.2X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
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 3146 3157 15 5.0 200.0 1.0X
SQL ORC Vectorized (Nested Column Disabled) 3059 3089 42 5.1 194.5 1.0X
SQL ORC Vectorized (Nested Column Enabled) 729 742 17 21.6 46.4 4.3X
SQL Parquet MR: DataPageV1 3409 3429 28 4.6 216.7 0.9X
SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 4156 4193 51 3.8 264.2 0.8X
SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 238 288 35 66.0 15.2 13.2X
SQL Parquet MR: DataPageV2 3147 3198 72 5.0 200.1 1.0X
SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 3725 3737 18 4.2 236.8 0.8X
SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 265 271 4 59.4 16.8 11.9X
================================================================================================
SQL Nested Column Scan
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
SQL Nested Column Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------------
SQL ORC MR 22187 22637 480 0.0 21158.8 1.0X
SQL ORC Vectorized (Nested Column Disabled) 21946 22361 192 0.0 20929.5 1.0X
SQL ORC Vectorized (Nested Column Enabled) 10302 10575 237 0.1 9824.5 2.2X
SQL Parquet MR: DataPageV1 14303 14615 186 0.1 13640.4 1.6X
SQL Parquet Vectorized: DataPageV1 (Nested Column Disabled) 15471 15847 183 0.1 14754.1 1.4X
SQL Parquet Vectorized: DataPageV1 (Nested Column Enabled) 10542 10794 190 0.1 10053.7 2.1X
SQL Parquet MR: DataPageV2 15047 15436 456 0.1 14350.0 1.5X
SQL Parquet Vectorized: DataPageV2 (Nested Column Disabled) 15663 15953 304 0.1 14937.9 1.4X
SQL Parquet Vectorized: DataPageV2 (Nested Column Enabled) 7861 8123 237 0.1 7497.2 2.8X
================================================================================================
Int and String Scan
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Int and String Scan: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 16721 16823 143 0.6 1594.7 1.0X
SQL Json 14569 14617 68 0.7 1389.4 1.1X
SQL Parquet Vectorized: DataPageV1 2554 2641 122 4.1 243.6 6.5X
SQL Parquet Vectorized: DataPageV2 2821 2881 85 3.7 269.1 5.9X
SQL Parquet MR: DataPageV1 5548 5563 20 1.9 529.1 3.0X
SQL Parquet MR: DataPageV2 5504 5544 56 1.9 524.9 3.0X
SQL ORC Vectorized 2580 2598 25 4.1 246.1 6.5X
SQL ORC MR 4902 5036 190 2.1 467.5 3.4X
================================================================================================
Repeated String Scan
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Repeated String: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 9365 9468 145 1.1 893.2 1.0X
SQL Json 8506 8528 31 1.2 811.2 1.1X
SQL Parquet Vectorized: DataPageV1 669 692 24 15.7 63.8 14.0X
SQL Parquet Vectorized: DataPageV2 674 707 31 15.5 64.3 13.9X
SQL Parquet MR: DataPageV1 2316 2392 109 4.5 220.8 4.0X
SQL Parquet MR: DataPageV2 2177 2212 49 4.8 207.6 4.3X
SQL ORC Vectorized 564 600 44 18.6 53.8 16.6X
SQL ORC MR 2412 2427 21 4.3 230.1 3.9X
================================================================================================
Partitioned Table Scan
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Partitioned Table: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
---------------------------------------------------------------------------------------------------------------------------------
Data column - CSV 23111 23239 181 0.7 1469.3 1.0X
Data column - Json 15498 15602 147 1.0 985.3 1.5X
Data column - Parquet Vectorized: DataPageV1 231 244 8 68.2 14.7 100.2X
Data column - Parquet Vectorized: DataPageV2 521 532 16 30.2 33.1 44.4X
Data column - Parquet MR: DataPageV1 3561 3632 100 4.4 226.4 6.5X
Data column - Parquet MR: DataPageV2 3222 3263 59 4.9 204.8 7.2X
Data column - ORC Vectorized 478 505 33 32.9 30.4 48.4X
Data column - ORC MR 3118 3208 128 5.0 198.2 7.4X
Partition column - CSV 7929 8099 241 2.0 504.1 2.9X
Partition column - Json 12790 12807 24 1.2 813.2 1.8X
Partition column - Parquet Vectorized: DataPageV1 62 67 6 252.0 4.0 370.3X
Partition column - Parquet Vectorized: DataPageV2 61 68 7 257.6 3.9 378.5X
Partition column - Parquet MR: DataPageV1 1603 1654 72 9.8 101.9 14.4X
Partition column - Parquet MR: DataPageV2 1611 1654 60 9.8 102.4 14.3X
Partition column - ORC Vectorized 62 67 7 254.7 3.9 374.3X
Partition column - ORC MR 1811 1860 69 8.7 115.2 12.8X
Both columns - CSV 23937 24052 162 0.7 1521.9 1.0X
Both columns - Json 16604 16619 21 0.9 1055.7 1.4X
Both columns - Parquet Vectorized: DataPageV1 321 343 20 49.1 20.4 72.1X
Both columns - Parquet Vectorized: DataPageV2 620 650 22 25.4 39.4 37.3X
Both columns - Parquet MR: DataPageV1 3546 3668 173 4.4 225.5 6.5X
Both columns - Parquet MR: DataPageV2 3460 3466 9 4.5 220.0 6.7X
Both columns - ORC Vectorized 516 527 12 30.5 32.8 44.8X
Both columns - ORC MR 3122 3163 57 5.0 198.5 7.4X
================================================================================================
String with Nulls Scan
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
String with Nulls Scan (0.0%): Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 11666 11668 3 0.9 1112.5 1.0X
SQL Json 13075 13170 134 0.8 1246.9 0.9X
SQL Parquet Vectorized: DataPageV1 1755 1771 23 6.0 167.4 6.6X
SQL Parquet Vectorized: DataPageV2 1983 2021 53 5.3 189.2 5.9X
SQL Parquet MR: DataPageV1 5326 5384 82 2.0 508.0 2.2X
SQL Parquet MR: DataPageV2 5471 5512 58 1.9 521.7 2.1X
ParquetReader Vectorized: DataPageV1 1290 1336 64 8.1 123.1 9.0X
ParquetReader Vectorized: DataPageV2 1510 1516 9 6.9 144.0 7.7X
SQL ORC Vectorized 1271 1301 43 8.3 121.2 9.2X
SQL ORC MR 3890 3899 13 2.7 371.0 3.0X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
String with Nulls Scan (50.0%): Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 8582 8845 372 1.2 818.4 1.0X
SQL Json 10106 10115 12 1.0 963.8 0.8X
SQL Parquet Vectorized: DataPageV1 1337 1389 75 7.8 127.5 6.4X
SQL Parquet Vectorized: DataPageV2 1494 1521 38 7.0 142.5 5.7X
SQL Parquet MR: DataPageV1 3965 3984 27 2.6 378.1 2.2X
SQL Parquet MR: DataPageV2 3911 3920 13 2.7 373.0 2.2X
ParquetReader Vectorized: DataPageV1 1311 1354 62 8.0 125.0 6.5X
ParquetReader Vectorized: DataPageV2 1554 1556 4 6.7 148.2 5.5X
SQL ORC Vectorized 1385 1389 6 7.6 132.1 6.2X
SQL ORC MR 3896 3904 11 2.7 371.5 2.2X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
String with Nulls Scan (95.0%): Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 5974 6029 78 1.8 569.8 1.0X
SQL Json 6444 6451 10 1.6 614.5 0.9X
SQL Parquet Vectorized: DataPageV1 356 382 28 29.5 33.9 16.8X
SQL Parquet Vectorized: DataPageV2 397 413 14 26.4 37.9 15.0X
SQL Parquet MR: DataPageV1 2602 2671 98 4.0 248.2 2.3X
SQL Parquet MR: DataPageV2 2328 2332 5 4.5 222.1 2.6X
ParquetReader Vectorized: DataPageV1 362 366 2 29.0 34.5 16.5X
ParquetReader Vectorized: DataPageV2 397 407 13 26.4 37.9 15.0X
SQL ORC Vectorized 457 480 27 23.0 43.6 13.1X
SQL ORC MR 1981 1990 13 5.3 188.9 3.0X
================================================================================================
Single Column Scan From Wide Columns
================================================================================================
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Single Column Scan from 10 columns: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 2378 2441 89 0.4 2267.9 1.0X
SQL Json 3840 3919 112 0.3 3662.1 0.6X
SQL Parquet Vectorized: DataPageV1 50 56 6 21.0 47.7 47.6X
SQL Parquet Vectorized: DataPageV2 66 72 6 15.9 62.9 36.1X
SQL Parquet MR: DataPageV1 241 255 11 4.3 230.0 9.9X
SQL Parquet MR: DataPageV2 218 231 9 4.8 207.9 10.9X
SQL ORC Vectorized 59 66 6 17.7 56.6 40.1X
SQL ORC MR 203 208 5 5.2 193.8 11.7X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Single Column Scan from 50 columns: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 5236 5239 4 0.2 4993.6 1.0X
SQL Json 16124 16203 112 0.1 15376.6 0.3X
SQL Parquet Vectorized: DataPageV1 68 74 6 15.4 64.9 77.0X
SQL Parquet Vectorized: DataPageV2 82 87 5 12.8 78.2 63.9X
SQL Parquet MR: DataPageV1 260 271 9 4.0 248.0 20.1X
SQL Parquet MR: DataPageV2 235 250 10 4.5 223.9 22.3X
SQL ORC Vectorized 81 89 7 13.0 77.0 64.8X
SQL ORC MR 226 235 7 4.6 215.2 23.2X
OpenJDK 64-Bit Server VM 17.0.2+8-LTS on Linux 5.13.0-1021-azure
Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Single Column Scan from 100 columns: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
SQL CSV 8774 8858 119 0.1 8367.3 1.0X
SQL Json 29846 30043 278 0.0 28463.7 0.3X
SQL Parquet Vectorized: DataPageV1 108 114 7 9.7 103.0 81.2X
SQL Parquet Vectorized: DataPageV2 118 124 5 8.9 112.5 74.3X
SQL Parquet MR: DataPageV1 297 315 12 3.5 283.7 29.5X
SQL Parquet MR: DataPageV2 278 293 12 3.8 265.1 31.6X
SQL ORC Vectorized 109 116 5 9.6 104.3 80.2X
SQL ORC MR 268 278 7 3.9 256.0 32.7X