[
https://issues.apache.org/jira/browse/CASSANDRA-7019?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15140520#comment-15140520
]
Branimir Lambov commented on CASSANDRA-7019:
--------------------------------------------
Thanks for the updates.
I haven't looked at dropping overwritten data, or cell level tombstones yet --
the idea still is to avoid deserializing (though we currently don't). I intend
to look at that after I have a non-deserializing option to compare with.
bq. should we always enable this for the single-sstable tombstone compactions?
It shouldn't hurt. I'll make the change.
bq. we should probably run this for a while on a large data set
Any ideas how to try that? Stress does not appear to support deletions. I'll
try adding that now.
> Improve tombstone compactions
> -----------------------------
>
> Key: CASSANDRA-7019
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7019
> Project: Cassandra
> Issue Type: Improvement
> Components: Compaction
> Reporter: Marcus Eriksson
> Assignee: Branimir Lambov
> Labels: compaction
> Fix For: 3.x
>
>
> When there are no other compactions to do, we trigger a single-sstable
> compaction if there is more than X% droppable tombstones in the sstable.
> In this ticket we should try to include overlapping sstables in those
> compactions to be able to actually drop the tombstones. Might only be doable
> with LCS (with STCS we would probably end up including all sstables)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
