GitHub user xuchuanyin opened a pull request:
https://github.com/apache/carbondata/pull/1593
[CARBONDATA-1837] Reusing origin row to reduce memory consumption
Be sure to do all of the following checklist to help us incorporate
your contribution quickly and easily:
- [X] Any interfaces changed?
`ADDED TWO INTERFACE FOR CARBON_ROW_BATCH`
- [X] Any backward compatibility impacted?
`NO`
- [X] Document update required?
`NO`
- [X] Testing done
Please provide details on
- Whether new unit test cases have been added or why no new tests
are required?
`NO. ONLY FOR OPTIMIZATION`
- How it is tested? Please attach test report.
`TESTED IN CLUSTER OF 3 NODES`
- Is it a performance related change? Please attach the performance
test report.
`PERFORMANCE SHOULD BE ENHANCED A BIT, SINCE MEMORY CONSUMPTION IS
REDUCED`
- Any additional information to help reviewers in testing this
change.
`NO`
- [X] For large changes, please consider breaking it into sub-tasks under
an umbrella JIRA.
`NOT RELATED`
NOTES:
===
In data converting process of data loading, Carbondata will convert each
row to another by batch.
Currently, it will create a new batch to store the converted rows, which I
think can be optimized to reuse the old row batch's space, thus will reduce
memory consumption and GC related overhead.
Tests have been done in my local cluster, enhancement gained is not
obvious, but the performance didn't become worse than that of before.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/xuchuanyin/carbondata opt_converter_memory
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/carbondata/pull/1593.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 #1593
----
commit b5b812cfad95e42500c2ed8e1724632dd489e55f
Author: xuchuanyin <[email protected]>
Date: 2017-11-29T03:59:48Z
reuse origin batch to reduce memory consumption
----
---
