| ================================================================================================ |
| BYTE_STREAM_SPLIT INT32 |
| ================================================================================================ |
| |
| OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure |
| AMD EPYC 9V74 80-Core Processor |
| BYTE_STREAM_SPLIT INT32: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative |
| ------------------------------------------------------------------------------------------------------------------------ |
| Spark vectorized readIntegers 2 2 0 509.3 2.0 1.0X |
| Spark vectorized readIntegersAsLongs 3 3 0 406.9 2.5 0.8X |
| Spark vectorized readIntegersAsDoubles 2 2 0 450.2 2.2 0.9X |
| parquet-mr readInteger (per-value) 8 8 0 133.8 7.5 0.3X |
| |
| |
| ================================================================================================ |
| BYTE_STREAM_SPLIT INT64 |
| ================================================================================================ |
| |
| OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure |
| AMD EPYC 9V74 80-Core Processor |
| BYTE_STREAM_SPLIT INT64: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative |
| ------------------------------------------------------------------------------------------------------------------------ |
| Spark vectorized readLongs 7 7 0 159.9 6.3 1.0X |
| Spark vectorized readLongsAsInts 7 7 0 157.6 6.3 1.0X |
| parquet-mr readLong (per-value) 13 13 1 81.1 12.3 0.5X |
| |
| |
| ================================================================================================ |
| BYTE_STREAM_SPLIT FLOAT |
| ================================================================================================ |
| |
| OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure |
| AMD EPYC 9V74 80-Core Processor |
| BYTE_STREAM_SPLIT FLOAT: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative |
| ------------------------------------------------------------------------------------------------------------------------ |
| Spark vectorized readFloats 2 2 0 486.9 2.1 1.0X |
| Spark vectorized readFloatsAsDoubles 2 3 0 433.6 2.3 0.9X |
| parquet-mr readFloat (per-value) 8 8 0 133.7 7.5 0.3X |
| |
| |
| ================================================================================================ |
| BYTE_STREAM_SPLIT DOUBLE |
| ================================================================================================ |
| |
| OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure |
| AMD EPYC 9V74 80-Core Processor |
| BYTE_STREAM_SPLIT DOUBLE: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative |
| ------------------------------------------------------------------------------------------------------------------------ |
| Spark vectorized readDoubles 6 7 1 161.4 6.2 1.0X |
| parquet-mr readDouble (per-value) 13 13 0 81.0 12.4 0.5X |
| |
| |
| ================================================================================================ |
| BYTE_STREAM_SPLIT FLBA |
| ================================================================================================ |
| |
| OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure |
| AMD EPYC 9V74 80-Core Processor |
| BYTE_STREAM_SPLIT FLBA: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative |
| --------------------------------------------------------------------------------------------------------------------------------------- |
| Spark vectorized readBinary (width=12) 24 25 3 43.6 22.9 1.0X |
| Spark per-value readBinary (old updater path, width=12) 29 29 1 36.2 27.6 0.8X |
| parquet-mr readBinary (per-value, width=12) 39 39 1 26.8 37.3 0.6X |
| |
| |