[
https://issues.apache.org/jira/browse/CASSANDRA-5549?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13886121#comment-13886121
]
Benedict edited comment on CASSANDRA-5549 at 1/30/14 1:11 AM:
--------------------------------------------------------------
I've rebased from trunk and fixed the bugs and pushed (-f) to
[merge-5549|https://github.com/belliottsmith/cassandra/tree/merge-5549]. There
were two small issues: in RangeTombstoneList I was setting start[i] = null
before I copied its value; and in FBUtilities.hashToBigInteger() my
"equivalency" rewrite wasn't so equivalent after all, so I reverted it. Whoops.
There are still three new tests failing, but this is because we fixed a bug and
so DefsTest no longer times out; CASSANDRA-6634 addresses these new issues.
was (Author: benedict):
I've rebased from trunk and fixed the bugs and pushed (-f) to
[merge-5549|https://github.com/belliottsmith/cassandra/tree/merge-5549]. There
were two small issues: in RangeTombstoneList I was setting start[i] = null
before I copied its value; and in FBUtilities.hashToBigInteger() my
"equivalency" rewrite wasn't so equivalent after all, so I reverted it. Whoops.
There are still three new tests failing, but this is because we fixed a bug and
so DefsTest no longer times out; CASSANDRA-6636 addresses these new issues.
> Remove Table.switchLock
> -----------------------
>
> Key: CASSANDRA-5549
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5549
> Project: Cassandra
> Issue Type: Bug
> Reporter: Jonathan Ellis
> Assignee: Benedict
> Labels: performance
> Fix For: 2.1
>
> Attachments: 5549-removed-switchlock.png, 5549-sunnyvale.png
>
>
> As discussed in CASSANDRA-5422, Table.switchLock is a bottleneck on the write
> path. ReentrantReadWriteLock is not lightweight, even if there is no
> contention per se between readers and writers of the lock (in Cassandra,
> memtable updates and switches).
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
