Fokko commented on PR #7301:
URL: https://github.com/apache/iceberg/pull/7301#issuecomment-1532986138
@singhpk234 I'm okay with the change, looking at the benchmark, I don't see
much difference:
With the change:
```
Benchmark
Mode Cnt Score Error Units
IcebergSourceFlatParquetDataReadBenchmark.readFileSourceNonVectorized
ss 5 4,917 ± 0,179 s/op
IcebergSourceFlatParquetDataReadBenchmark.readFileSourceVectorized
ss 5 1,806 ± 0,036 s/op
IcebergSourceFlatParquetDataReadBenchmark.readIceberg
ss 5 2,037 ± 0,022 s/op
IcebergSourceFlatParquetDataReadBenchmark.readWithProjectionFileSourceNonVectorized
ss 5 0,978 ± 0,077 s/op
IcebergSourceFlatParquetDataReadBenchmark.readWithProjectionFileSourceVectorized
ss 5 0,435 ± 0,028 s/op
IcebergSourceFlatParquetDataReadBenchmark.readWithProjectionIceberg
ss 5 0,364 ± 0,021 s/op
```
Master:
```
Benchmark
Mode Cnt Score Error Units
IcebergSourceFlatParquetDataReadBenchmark.readFileSourceNonVectorized
ss 5 4,658 ± 0,222 s/op
IcebergSourceFlatParquetDataReadBenchmark.readFileSourceVectorized
ss 5 1,774 ± 0,056 s/op
IcebergSourceFlatParquetDataReadBenchmark.readIceberg
ss 5 1,952 ± 0,060 s/op
IcebergSourceFlatParquetDataReadBenchmark.readWithProjectionFileSourceNonVectorized
ss 5 1,019 ± 0,211 s/op
IcebergSourceFlatParquetDataReadBenchmark.readWithProjectionFileSourceVectorized
ss 5 0,422 ± 0,208 s/op
IcebergSourceFlatParquetDataReadBenchmark.readWithProjectionIceberg
ss 5 0,363 ± 0,042 s/op
```
It even looks a bit faster on some of the benchmarks, but when taking the
error into account, the difference is minimal.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
