GitHub user watermen opened a pull request:
https://github.com/apache/incubator-carbondata/pull/719 [WIP][CARBONDATA-844] Avoid to get useless splits
In current implements of CarbonInputFormat.getDataBlocksOfSegment,
1. Get all of the carbondata splits in segments directory.
2. Read the carbonindex and construct the B-tree.
3. Apply filter and get matching splits.
I think we get some useless splits and the operator of getSplits is expensive. So we'd better to do the getSplits after filter:
1. List the segment directory, and filter the path of carbonindex.
2. Read the carbonindex and construct the B-tree.
3. Apply filter and get matching blocks.
4. Get carbondata splits from filtered blocks.
TODOList:
- [ ] How to do something about IUD(isValidBlockBasedOnUpdateDetails)?
- [ ] Fix bug in suite
@jackylk @QiangCai Plz review the code when you have time.
You can merge this pull request into a Git repository by running:
$ git pull
https://github.com/watermen/incubator-carbondata CARBONDATA-844
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/incubator-carbondata/pull/719.patchTo close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #719
----
commit 0de95cb219300c7127c70a3f7e126960ee354e58
Author: Yadong Qi <
[hidden email]>
Date: 2017-04-01T08:42:38Z
Avoid to get useless splits.
----
---
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
[hidden email] or file a JIRA ticket
with INFRA.
---
Locked
