Make the feature "Parallel Readdir" production ready

[poornimag]

No description provided.

[ShyamsundarR]

What does this involve? some further details would help in understand what part of the feature is still pending, or are there bugs to be fixed?

Further, we can drop 3.11 from the issue subject, as that ties a feature to a release, we can track and update that using the milestones instead.

[poornimag]

No defaults are changed. In 3.10 release notes, this feature was mentioned as Experimental. This issue is for restating the 3.10 release note content, but without experimental tag.

The 3.10 release not content:
Currently the directory listing gets slower as the number of bricks/nodes increases in a volume, though the file/directory numbers remain unchanged. With this feature, the performance of directory listing is made mostly independent of the number of nodes/bricks in the volume. Thus scale doesn’t exponentially reduce the directory listing performance. (On a 2, 5, 10, 25 brick setup we saw ~5, 100, 400, 450% improvement consecutively)

To enable this feature:
#gluster volume set performance.readdir-ahead on
#gluster volume set performance.parallel-readdir on

To disable this feature:
#gluster volume set performance.parallel-readdir off
If there are more than 50 bricks in the volume it is good to increase the cache size to be more than 10Mb (default value):
#gluster volume set performance.rda-cache-limit

[gluster-ant]

A patch https://review.gluster.org/17390 has been posted that references this issue.
Commit message: doc: Updated release notes for several features/changes in 3.11.0