Skye Wanderman-Milne has posted comments on this change.

Change subject: IMPALA-3311: fix string data coming out of aggs in subplans
......................................................................


Patch Set 1:

(3 comments)

http://gerrit.cloudera.org:8080/#/c/2929/1/be/src/exec/partitioned-aggregation-node.cc
File be/src/exec/partitioned-aggregation-node.cc:

Line 370:   for (int i = 0; i < aggregate_evaluators_.size(); ++i) {
> This is a pretty big block of code, I think it'd be more readable if it was
I moved the string copying code to its own function. Lemme know if you think I 
should factor out more.


Line 374:     if (IsInSubplan()) {
> Makes sense. Is it worth adding a targeted perf query to check we don't reg
I'm having trouble writing a good query that really isolates this effect and 
isn't too complicated. I have a query where the effect is noticeable, but the 
subplan still dominates the time (2s vs 2.4s locally, but the non-subplan 
aggregation time doubles). I'll post what I have.


Line 380:       for (int i = first_row_idx; i < row_batch->num_rows(); ++i, 
++row) {
> You could use the new RowBatch::Iterator. I think you're doing exactly what
Done


-- 
To view, visit http://gerrit.cloudera.org:8080/2929
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Iada891504c261ba54f4eb8c9d7e4e5223668d7b9
Gerrit-PatchSet: 1
Gerrit-Project: Impala
Gerrit-Branch: cdh5-trunk
Gerrit-Owner: Skye Wanderman-Milne <[email protected]>
Gerrit-Reviewer: Dan Hecht <[email protected]>
Gerrit-Reviewer: Skye Wanderman-Milne <[email protected]>
Gerrit-Reviewer: Tim Armstrong <[email protected]>
Gerrit-HasComments: Yes

Reply via email to