GitHub user xuchuanyin opened a pull request:
https://github.com/apache/carbondata/pull/1177
[CARBONDATA-1281] Support multiple temp dirs for writing files while
loading data
# Modifications
This feature mainly focus on avoiding disk hot-spot in single massive data
loading, changes are made in two parts:
1. randomly choose a yarn local folder to write sort temp file in
sort-process;
2.randomly choose a yarn local folder to write carbondata file in
write-process.
# Usage
To enable this feature, user should enable
`carbon.using.multi.temp.dir=true` and `carbon.use.local.dir=true`.
# Performance
In my case, this feature improves the loading performance from 35M/s/node
to 70+M/s/node
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/xuchuanyin/carbondata
local_multiple_temp_dirs_4_load
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/carbondata/pull/1177.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 #1177
----
commit ee5e00cdae5c64fcfad7d7cb1b1127dc5bb79bb4
Author: xuchuanyin <[email protected]>
Date: 2017-07-17T13:44:49Z
Support multiple temp dirs for writing files while loading data
randomly choose a dir to write sort temp files
randomly choose a dir to write carbondata files
----
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
