[
https://issues.apache.org/jira/browse/HADOOP-7076?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Niels Basjes updated HADOOP-7076:
---------------------------------
Attachment: HADOOP-7076-2011-01-26.patch
I got some valuable review comments from Chris Douglas which I've processed in
this new version.
Summary of the changes:
- The only existing file that is changed is the unit test class TestCodec
(fixed bugs + reused unit tests).
- No other changes to any existing classes or interfaces (so only new files).
- Refactored the Split Seams test to allow any compressed text file as test
input.
- Tested with synthetic (part of unit test) and several of my own real logfiles.
- Tested with fixed line lengths (as small as 1 byte) and variable (random)
line lengths
- Tested with both a normal and a file that is really a concatenation of a lot
of very small gzip files.
- Tested with both the built in and the native gzip implementation.
> Splittable Gzip
> ---------------
>
> Key: HADOOP-7076
> URL: https://issues.apache.org/jira/browse/HADOOP-7076
> Project: Hadoop Common
> Issue Type: New Feature
> Components: io
> Reporter: Niels Basjes
> Assignee: Niels Basjes
> Attachments: HADOOP-7076-2011-01-26.patch, HADOOP-7076.patch
>
>
> Files compressed with the gzip codec are not splittable due to the nature of
> the codec.
> This limits the options you have scaling out when reading large gzipped input
> files.
> Given the fact that gunzipping a 1GiB file usually takes only 2 minutes I
> figured that for some use cases wasting some resources may result in a
> shorter job time under certain conditions.
> So reading the entire input file from the start for each split (wasting
> resources!!) may lead to additional scalability.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
