GitHub user jianqiao opened a pull request:
https://github.com/apache/incubator-quickstep/pull/239
Add ThreadPrivateCompactKeyHashTable for aggregation.
This PR implements a new specialized aggregation hash table that is
preferable for two-phase (i.e. aggregate locally and then merge) aggregation
with small-cardinality group-by keys. To use this hash table, it also requires
that the group-by keys have fixed-length types with total byte size no greater
than 8 (so that each composite key can be packed into a 64-bit QWORD).
This PR together with #237 can improve the performance TPC-H Q01 from
~11.5s to ~5.3s, with scale factor 100 on a cloud lab machine.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/jianqiao/incubator-quickstep
thread-private-compact-ht
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/incubator-quickstep/pull/239.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #239
----
commit 99af4690ae4462e7b6843bc92c4cb3ef31a1d75a
Author: Jianqiao Zhu <jianq...@cs.wisc.edu>
Date: 2017-04-22T04:23:13Z
Add ThreadPrivateCompactKeyHashTable as a fast path data structure for
aggregation.
----
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
