GitHub user liancheng opened a pull request:
https://github.com/apache/spark/pull/12153
[SPARK-14369][SQL][WIP][test-hadoop2.0][test-hadoop2.2][test-hadoop2.3]
Locality support for FileScanRDD
## What changes were proposed in this pull request?
This PR adds preliminary locality support for `FileScanRDD` by querying
block locations of all file blocks contained in each `FilePartition`.
Note:
- Rack information is not considered yet.
- S3 is special cased since S3 doesn't provide valid locality information
and its file metadata operation latency can be super high.
## How was this patch tested?
Not finished yet. Trying mocked `FileSystem` to return given list of block
locations. Opening the PR first to ensure it doesn't break other parts and
works for all supported Hadoop versions.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/liancheng/spark spark-14369-locality
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/12153.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 #12153
----
commit 179ff5898cf1aa2acc52514e1703b0534b5b6d7c
Author: Cheng Lian <[email protected]>
Date: 2016-04-04T15:33:18Z
Locality support for FileScanRDD
----
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
