[
https://issues.apache.org/jira/browse/HBASE-11135?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13994786#comment-13994786
]
stack commented on HBASE-11135:
-------------------------------
[~jeffreyz] Sorry, can you restate the below. I do not follow:
bq. Stack Have you played the following? Because HRegion#getNextSequenceId() is
called during a flush with updatesLock.writeLock().
What are you saying with the above? (Pardon me)
bq. If we use two ring buffer, HRegion#getNextSequenceId() will return very
quickly because it won't wait for previous work items such as preWAL, postWAL,
writer.append() etc.
I can try it. Looking at it, it just seemed overly-complicated.
It is my sense that the slow down is the hand-off across the ring buffer, the
passing back off the sequence id to the handler, that is holding it up (I did a
quick measure where I commented out the latch code that is in this patch --
i.e. not having the handler wait on the sequence id -- and I got back the same
numbers we currently have in trunk). If we do double ring buffer, we will
still have to do thread handoff across the first buffer so I foresee us having
the same slow down only w/ a more complex system. What you think?
Does this patch get you what you wanted and unblocks your mvcc + sequence id
unificattion? If so, should we commit it? This patch is getting big. I can
work on perf in a new issue, a smaller one.
> Change region sequenceid generation so happens earlier in the append cycle
> rather than just before added to file
> ----------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-11135
> URL: https://issues.apache.org/jira/browse/HBASE-11135
> Project: HBase
> Issue Type: Sub-task
> Components: wal
> Reporter: stack
> Assignee: stack
> Attachments: 11135.wip.txt, 11135v2.txt, 11135v5.txt, 11135v5.txt
>
>
> Currently we assign the region edit/sequence id just before we put it in the
> WAL. We do it in the single thread that feeds from the ring buffer. Doing
> it at this point, we can ensure order, that the edits will be in the file in
> accordance w/ the ordering of the region sequence id.
> But the point at which region sequence id is assigned an edit is deep down in
> the WAL system and there is a lag between our putting an edit into the WAL
> system and the edit actually getting its edit/sequence id.
> This lag -- "late-binding" -- complicates the unification of mvcc and region
> sequence id, especially around async WAL writes (and, related, for no-WAL
> writes) -- the parent for this issue (For async, how you get the edit id in
> our system when the threads have all gone home -- unless you make them wait?)
> Chatting w/ Jeffrey Zhong yesterday, we came up with a crazypants means of
> getting the region sequence id near-immediately. We'll run two ringbuffers.
> The first will mesh all handler threads and the consumer will generate ids
> (we will have order on other side of this first ring buffer), and then if
> async or no sync, we will just let the threads return ... updating mvcc just
> before we let them go. All other calls will go up on to the second ring
> buffer to be serviced as now (batching, distribution out among the sync'ing
> threads). The first rb will have no friction and should turn at fast rates
> compared to the second. There should not be noticeable slowdown nor do I
> foresee this refactor intefering w/ our multi-WAL plans.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
