wypoon commented on PR #11661: URL: https://github.com/apache/iceberg/pull/11661#issuecomment-2932993513
@pvary I reran `VectorizedReadDictionaryEncodedFlatParquetDataBenchmark` on my branch before and after the commits in this PR. Before: ``` Benchmark Mode Cnt Score Error Units VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readBigDecimalsIcebergVectorized5k ss 5 13.379 ± 0.149 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readBigDecimalsSparkVectorized5k ss 5 14.129 ± 1.095 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDatesIcebergVectorized5k ss 5 5.138 ± 0.153 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDatesSparkVectorized5k ss 5 4.236 ± 0.694 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDecimalsIcebergVectorized5k ss 5 5.015 ± 0.260 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDecimalsSparkVectorized5k ss 5 5.955 ± 0.458 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDoublesIcebergVectorized5k ss 5 5.204 ± 0.077 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDoublesSparkVectorized5k ss 5 5.162 ± 1.460 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readFloatsIcebergVectorized5k ss 5 4.763 ± 0.048 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readFloatsSparkVectorized5k ss 5 4.471 ± 1.163 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readIntegersIcebergVectorized5k ss 5 5.726 ± 0.150 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readIntegersSparkVectorized5k ss 5 4.220 ± 0.639 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readLongsIcebergVectorized5k ss 5 4.482 ± 0.289 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readLongsSparkVectorized5k ss 5 4.594 ± 0.191 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readStringsIcebergVectorized5k ss 5 6.401 ± 0.193 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readStringsSparkVectorized5k ss 5 7.955 ± 0.647 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readTimestampsIcebergVectorized5k ss 5 5.548 ± 0.119 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readTimestampsSparkVectorized5k ss 5 4.990 ± 0.312 s/op ``` After: ``` Benchmark Mode Cnt Score Error Units VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readBigDecimalsIcebergVectorized5k ss 5 14.353 ± 1.208 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readBigDecimalsSparkVectorized5k ss 5 13.920 ± 0.707 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDatesIcebergVectorized5k ss 5 5.187 ± 1.328 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDatesSparkVectorized5k ss 5 5.050 ± 1.464 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDecimalsIcebergVectorized5k ss 5 5.383 ± 0.140 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDecimalsSparkVectorized5k ss 5 6.532 ± 0.500 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDoublesIcebergVectorized5k ss 5 5.938 ± 1.266 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readDoublesSparkVectorized5k ss 5 4.404 ± 0.445 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readFloatsIcebergVectorized5k ss 5 5.194 ± 0.273 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readFloatsSparkVectorized5k ss 5 4.979 ± 2.074 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readIntegersIcebergVectorized5k ss 5 5.330 ± 0.356 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readIntegersSparkVectorized5k ss 5 5.354 ± 0.411 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readLongsIcebergVectorized5k ss 5 5.131 ± 0.208 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readLongsSparkVectorized5k ss 5 5.589 ± 0.657 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readStringsIcebergVectorized5k ss 5 6.881 ± 0.180 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readStringsSparkVectorized5k ss 5 7.440 ± 0.966 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readTimestampsIcebergVectorized5k ss 5 4.445 ± 0.128 s/op VectorizedReadDictionaryEncodedFlatParquetDataBenchmark.readTimestampsSparkVectorized5k ss 5 5.256 ± 0.910 s/op ``` The numbers are roughly the same, some slightly worse and some slightly better after. I think it's basically noise. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org