[
https://issues.apache.org/jira/browse/CASSANDRA-1930?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jonathan Ellis updated CASSANDRA-1930:
--------------------------------------
Affects Version/s: (was: 0.7.0)
(was: 0.7.0 rc 3)
(was: 0.7.0 rc 2)
(was: 0.6.9)
(was: 0.6.8)
(was: 0.7.0 rc 1)
(was: 0.6.7)
(was: 0.6.6)
(was: 0.7 beta 2)
(was: 0.6.5)
(was: 0.7 beta 3)
(was: 0.7.1)
(was: 0.6.4)
(was: 0.6.3)
(was: 0.6.2)
(was: 0.6.1)
(was: 0.8)
(was: 0.7 beta 1)
(was: 0.6)
(was: 0.5)
(was: 0.4)
(was: 0.3)
(Let's leave "Affects version=unknown" to mean "affects all versions," and
otherwise affects version to be the minimum version affected. The
exhaustive-list-of-versions approach is obsoleted whenever a new version is
created, e.g. 0.6.10 or 0.7.2.)
> db.Table flusherLock write lock fairness policy is sub-optimal
> --------------------------------------------------------------
>
> Key: CASSANDRA-1930
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1930
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Kelvin Kakugawa
> Assignee: Kelvin Kakugawa
> Fix For: 0.8
>
> Attachments: CASSANDRA-1930.0001.patch
>
>
> h4. scenario:
> 1) high write throughput cluster
> 2) external services adding material cpu contention
> h4. symptoms:
> The row mutation stage falls *very* behind.
> h4. cause:
> ReentrantReadWriteLock's fair policy causes write lock acquisition / release
> to require a material amount of CPU time.
> h4. summary:
> When there are other services contending for the CPU, the RRW lock's fair
> policy causes write lock acquisition / release to take enough time to
> eventually put threads waiting for read lock acquisition very behind. We
> repro'd this scenario by reducing the scope of the write lock to: 1 boolean
> check, 1 boolean assignment, and 1 db.Memtable instantiation (itself, just: 2
> variable assignments) w/ the same performance. Modifying the fairness policy
> to be the default policy allowed the row mutation stage to keep up.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
