GitHub user xuchuanyin opened a pull request:
https://github.com/apache/carbondata/pull/1953
[CARBONDATA-2091][DataLoad] Support specifying sort column bounds in data
loading
Enhance data loading performance by specifying sort column bounds
1. Add row range number during convert-process-step
2. Dispatch rows to each sorter by range number
3. Sort/Write process step can be done concurrently in each range
Tests added and docs updated
After implementing this feature, the data load performance has gained about
25% enhancement (80MB/s/Node -> 102MB/s/Node) in my scenario with only 1 bounds
provided.
Be sure to do all of the following checklist to help us incorporate
your contribution quickly and easily:
- [x] Any interfaces changed?
`Only internal used interfaces are changed`
- [x] Any backward compatibility impacted?
`No`
- [x] Document update required?
`Yes, added the usage of this feature to documents`
- [x] Testing done
Please provide details on
- Whether new unit test cases have been added or why no new tests
are required?
`Yes`
- How it is tested? Please attach test report.
`Tested in 3-node cluster and local machine`
- Is it a performance related change? Please attach the performance
test report.
`Yes. After implementing this feature, the data load performance has gained
about 25% enhancement (80MB/s/Node -> 102MB/s/Node) in my scenario with only 1
bounds provided. `
- Any additional information to help reviewers in testing this
change.
`I refactored the bucket related feature and treated the range and
bucket as the similar logic`
- [x] For large changes, please consider breaking it into sub-tasks under
an umbrella JIRA.
`Not related`
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/xuchuanyin/carbondata
0208_support_specifying_sort_column_bounds
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/carbondata/pull/1953.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 #1953
----
commit 11463dd22db17f2e1858e0a1f3ebfeb07e3ec0e9
Author: xuchuanyin <xuchuanyin@...>
Date: 2018-02-08T08:30:09Z
Support specifying sort column bounds in data loading
Enhance data loading performance by specifying sort column bounds
1. Add row range number during convert-process-step
2. Dispatch rows to each sorter by range number
3. Sort/Write process step can be done concurrently in each range
Tests added and docs updated
----
---
