[
https://issues.apache.org/jira/browse/CASSANDRA-7019?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15098054#comment-15098054
]
Branimir Lambov commented on CASSANDRA-7019:
--------------------------------------------
bq. Perhaps we could apply it to the higher (L3+?) levels in leveled compaction?
This makes sense. Better still, we can probably exclude older tables altogether
(as they can't have newer tombstones) which should avoid bringing higher-level
tables on a lower-level compaction.
bq. do we drop overwritten data as well?
No. I was hoping the table iterators could deserialize just a row header, which
after reading the code more closely doesn't appear to be the case. I'll do some
performance evaluations next and experiment with different variations (e.g.
slicing to reach the row within partition instead of iterating as well as
looking within the row).
> 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)
