[
https://issues.apache.org/jira/browse/HBASE-4583?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13486703#comment-13486703
]
Lars Hofhansl commented on HBASE-4583:
--------------------------------------
[~jdcryans]
bq. The way we currently upsert in the MemStore is saving us from flushing
hundreds of GBs of crap than then needs to be compacted.
That is actually not entirely true. With HBASE-4241 it would flush the store
more frequently, but during flush older versions of KVs are ignored (i.e. not
flushed at all, and thus they do not need to be compacted later). That's the
reason why the "radical" patch is not 100x slower, but just 15% (with VERSIONS
set to 1).
Please have a look at the "less radical" patch.
Because of HBASE-4241 I still prefer the "radical" version (trunk-v3). With
many increments for a single row the search in the memstore for upsert might
even become counter productive.
Anyway... I'm fine with committing either of the two patches here. Both should
be correct in terms of MVCC behavior.
> Integrate RWCC with Append and Increment operations
> ---------------------------------------------------
>
> Key: HBASE-4583
> URL: https://issues.apache.org/jira/browse/HBASE-4583
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Fix For: 0.94.3, 0.96.0
>
> Attachments: 4583-trunk-less-radical.txt, 4583-trunk-radical.txt,
> 4583-trunk-radical_v2.txt, 4583-trunk-v3.txt, 4583.txt, 4583-v2.txt,
> 4583-v3.txt, 4583-v4.txt
>
>
> Currently Increment and Append operations do not work with RWCC and hence a
> client could see the results of multiple such operation mixed in the same
> Get/Scan.
> The semantics might be a bit more interesting here as upsert adds and removes
> to and from the memstore.
--
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
