impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tim Armstrong (Code Review)" <ger...@cloudera.org>
Subject [Impala-ASF-CR] IMPALA-4939, IMPALA-4939: Decimal V2 multiplication
Date Thu, 03 Aug 2017 22:51:11 GMT
Tim Armstrong has posted comments on this change.

Change subject: IMPALA-4939, IMPALA-4939: Decimal V2 multiplication
......................................................................


Patch Set 4:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/7438/4//COMMIT_MSG
Commit Message:

Line 75: DECIMAL_V2 disabled: 2.65s
> It takes time to convert a number from 128 bits to 256 bits. Also, Boost do
So we go down this path for multiplies that overflow, or are close to overflowing?

I'd be less concerned if the multiply itself was 4x slower in isolation, but 10x end-to-end
slowdown implies that the slowdown of the operation in isolation is more than 10x, probably
a lot more.

It seems like we should be able to do much better than that. Might be worth doing some profiling
to understand where the cycles are going. This is dominating query runtime so much you could
probably see what's happening with disable_codegen=true and perf top if you run the query
in a loop.


-- 
To view, visit http://gerrit.cloudera.org:8080/7438
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I37ad6232d7953bd75c18dc86e665b2b501a1ebe1
Gerrit-PatchSet: 4
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Taras Bobrovytsky <tbobrovytsky@cloudera.com>
Gerrit-Reviewer: Dan Hecht <dhecht@cloudera.com>
Gerrit-Reviewer: Michael Ho <kwho@cloudera.com>
Gerrit-Reviewer: Taras Bobrovytsky <tbobrovytsky@cloudera.com>
Gerrit-Reviewer: Tim Armstrong <tarmstrong@cloudera.com>
Gerrit-Reviewer: Zach Amsden <zamsden@cloudera.com>
Gerrit-HasComments: Yes

Mime
View raw message