Tim Armstrong has posted comments on this change. Change subject: IMPALA-4923: reduce memory transfer for selective scans ......................................................................
Patch Set 3: (3 comments) http://gerrit.cloudera.org:8080/#/c/6949/3/be/src/exec/parquet-scratch-tuple-batch.h File be/src/exec/parquet-scratch-tuple-batch.h: Line 102: ++num_output_batches; > I think the idea was right, but the code wasn't quite implementing the idea I looked again and realised we don't even need to track num_output_batches. num_tuples_transferred > num_to_commit implies that tuples were transferred to a previous batch. It's enough of a simplification it seems worth doing. http://gerrit.cloudera.org:8080/#/c/6949/4/be/src/exec/parquet-scratch-tuple-batch.h File be/src/exec/parquet-scratch-tuple-batch.h: PS4, Line 142: Tuple* uncompacted_tuple = row->GetTuple(0); > FOREACH_ROW() ? I looked earlier - the rows aren't committed to the batch yet so FOREACH_ROW doesn't work for this case. PS4, Line 155: } > not needed ? Yep, and also the ones below. -- To view, visit http://gerrit.cloudera.org:8080/6949 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I3773dc63c498e295a2c1386a15c5e69205e747ea Gerrit-PatchSet: 3 Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-Owner: Tim Armstrong <[email protected]> Gerrit-Reviewer: Alex Behm <[email protected]> Gerrit-Reviewer: Dan Hecht <[email protected]> Gerrit-Reviewer: Michael Ho <[email protected]> Gerrit-Reviewer: Mostafa Mokhtar <[email protected]> Gerrit-Reviewer: Tim Armstrong <[email protected]> Gerrit-HasComments: Yes
