[
https://issues.apache.org/jira/browse/HBASE-8665?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13671091#comment-13671091
]
Sergey Shelukhin commented on HBASE-8665:
-----------------------------------------
W.r.t. property of the store:
1) Current priority is derived entirely from store state. In fact it only
depends on number of files and blocking limit, so there's no such thing as
important store other than blocked/about-to-block store.
2) It seems to be the obvious cause of this particular issue, see above. When
selection is added to queue, its priority is low. As store fills up it's "real"
priority changes, but we have no way to influence it. Bumping its priority when
we queue another one is just a hack around that imho; if it even helps. If
blocked store comes into a queue full of non-blocked stores why should we bump
them? And how much.
3) Additionally if we pre-select in advance we simply end up with inefficient
compactions, for example in this case we could have compacted 6 small files in
one go, but would have ended up compacting 3 and 3 if it spent less time in the
queue.
> bad compaction priority behavior in queue can cause store to be blocked
> -----------------------------------------------------------------------
>
> Key: HBASE-8665
> URL: https://issues.apache.org/jira/browse/HBASE-8665
> Project: HBase
> Issue Type: Bug
> Reporter: Sergey Shelukhin
> Assignee: Sergey Shelukhin
>
> Note that this can be solved by bumping up the number of compaction threads
> but still it seems like this priority "inversion" should be dealt with.
> There's a store with 1 big file and 3 flushes (1 2 3 4) sitting around and
> minding its own business when it decides to compact. Compaction (2 3 4) is
> created and put in queue, it's low priority, so it doesn't get out of the
> queue for some time - other stores are compacting. Meanwhile more files are
> flushed and at (1 2 3 4 5 6 7) it decides to compact (5 6 7). This compaction
> now has higher priority than the first one. After that if the load is high it
> enters vicious cycle of compacting and compacting files as they arrive, with
> store being blocked on and off, with the (2 3 4) compaction staying in queue
> for up to ~20 minutes (that I've seen).
> I wonder why we do thing thing where we queue compaction and compact
> separately. Perhaps we should take snapshot of all store priorities, then do
> select in order and execute the first compaction we find. This will need
> starvation safeguard too but should probably be better.
> Btw, exploring compaction policy may be more prone to this, as it can select
> files from the middle, not just beginning, which, given the treatment of
> already selected files that was not changed from the old ratio-based one (all
> files with lower seqNums than the ones selected are also ineligible for
> further selection), will make more files ineligible (e.g. imagine with 10
> blocking files, with 8 present (1-8), (6 7 8) being selected and getting
> stuck). Today I see the case that would also apply to old policy, but
> yesterday I saw file distribution something like this: 4,5g, 2,1g, 295,9m,
> 113,3m, 68,0m, 67,8m, 1,1g, 295,1m, 100,4m, unfortunately w/o enough logs to
> figure out how it resulted.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
