Details
-
Improvement
-
Status: Resolved
-
Minor
-
Resolution: Done
-
None
-
ghx-label-6
Description
Removing the "if" at
https://github.com/apache/impala/blob/5670f96b828d57f9e36510bb9af02bcc31de775c/be/src/exec/parquet/parquet-common.h#L203
can lead to 1.5x speed up in plain decoding (type=int32, stride=16). For primitive types, the same check can be done for a whole batch, so the speedup can be gained for large batches without loosing safety. The only Parquet type where this check is needed per element is BYTE_ARRAY (typically used for STRING columns), which already has a template specialization for ParquetPlainEncoder::Decode().
Attachments
Issue Links
- Blocked
-
IMPALA-8467 ParquetPlainEncoder::Decode leads to multiple test failures in ASAN builds
- Resolved