[
https://issues.apache.org/jira/browse/CASSANDRA-5515?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13726900#comment-13726900
]
Tyler Hobbs commented on CASSANDRA-5515:
----------------------------------------
With that in mind, I'm leaning towards using Metric's
[Meters|http://metrics.codahale.com/manual/core/#meters], which have pretty low
overhead and would be be easy to dump and restore. The only flaw is that they
are hard-coded with 1m, 5m, and 15m windows; for compaction decisions, I feel
like it would be nice to have a multi-hour window, and the 1 and 5m are
probably too short to be useful for CASSANDRA-5519 (if we do something similar
to my suggestion on that ticket). We could of course subclass and hardcode
with windows of our choosing.
> Track sstable coldness
> ----------------------
>
> Key: CASSANDRA-5515
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5515
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Tyler Hobbs
> Fix For: 2.0.1
>
> Attachments: 0001-Track-row-read-counts-in-SSTR.patch
>
>
> Keeping a count of reads per-sstable would allow STCS to automatically ignore
> cold data rather than recompacting it constantly with hot data, dramatically
> reducing compaction load for typical time series applications and others with
> time-correlated access patterns. We would not need a separate age-tiered
> compaction strategy.
> (This will really be useful in conjunction with CASSANDRA-5514.)
--
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
