[
https://issues.apache.org/jira/browse/HBASE-1177?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12715346#action_12715346
]
Jim Kellerman commented on HBASE-1177:
--------------------------------------
Based on the graph
https://issues.apache.org/jira/secure/attachment/12409623/zoom+of+columns+vs+round-trip+blowup.jpg
the blow-up in round-trip time happens between 8 and 23 columns (approximately
- different runs will have different end points), but the amount of time spent
in getRow remains pretty constant through this range.
For larger numbers of columns (as depicted in
https://issues.apache.org/jira/secure/attachment/12409622/getRow+%2B+round-trip+vs+%23+columns.jpg
), getRow and round-trip time seem to scale pretty linearly.
This is pretty strange, but at least it doesn't seem related to server-side I/O.
> Delay when client is located on the same node as the regionserver
> -----------------------------------------------------------------
>
> Key: HBASE-1177
> URL: https://issues.apache.org/jira/browse/HBASE-1177
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.19.0
> Environment: Linux 2.6.25 x86_64
> Reporter: Jonathan Gray
> Assignee: Jim Kellerman
> Priority: Blocker
> Fix For: 0.20.0
>
> Attachments: Contribution of getClosest to getRow time.jpg,
> Contribution of next to getRow time.jpg, Contribution of seekTo to getClosest
> time.jpg, getRow + round-trip vs # columns.jpg, getRow times.jpg,
> ReadDelayTest.java, screenshot-1.jpg, screenshot-2.jpg, screenshot-3.jpg,
> screenshot-4.jpg, zoom of columns vs round-trip blowup.jpg
>
>
> During testing of HBASE-80, we uncovered a strange 40ms delay for random
> reads. We ran a series of tests and found that it only happens when the
> client is on the same node as the RS and for a certain range of payloads (not
> specifically related to number of columns or size of them, only total
> payload). It appears to be precisely 40ms every time.
> Unsure if this is particular to our architecture, but it does happen on all
> nodes we've tried. Issue completely goes away with very large payloads or
> moving the client.
> Will post a test program tomorrow if anyone can test on a different
> architecture.
> Making a blocker for 0.20. Since this happens when you have an MR task
> running local to the RS, and this is what we try to do, might also consider
> making this a blocker for 0.19.1.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
