[
https://issues.apache.org/jira/browse/CASSANDRA-6621?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14029432#comment-14029432
]
sankalp kohli commented on CASSANDRA-6621:
------------------------------------------
"Would it be simpler to just look at all sstables on the new node after we have
bootstrapped (but before we start compacting) and try to "optimally" distribute
them in levels? Feels like we could do a better job in this case"
Yes this will be better if we can afford to pause the compaction on organic
writes. I was trying to give an approach in which we don't pause compaction.
The thing to keep in mind is that for large nodes, it will take several hours
to bootstrap and compaction won't be running on the new organic data. This
might be a problem for cases where people insert into C* in batches.
> STCS fallback is not optimal when bootstrapping
> -----------------------------------------------
>
> Key: CASSANDRA-6621
> URL: https://issues.apache.org/jira/browse/CASSANDRA-6621
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Bartłomiej Romański
> Assignee: Marcus Eriksson
> Priority: Minor
> Labels: compaction
>
> The initial discussion started in (closed) CASSANDRA-5371. I've rewritten my
> last comment here...
> After streaming (e.g. during boostrap) Cassandra places all sstables at L0.
> At the end of the process we end up with huge number of sstables at the
> lowest level.
> Currently, Cassandra falls back to STCS until the number of sstables at L0
> reaches the reasonable level (32 or something).
> I'm not sure if falling back to STCS is the best way to handle this
> particular situation. I've read the comment in the code and I'm aware why it
> is a good thing to do if we have to many sstables at L0 as a result of too
> many random inserts. We have a lot of sstables, each of them covers the whole
> ring, there's simply no better option.
> However, after the bootstrap situation looks a bit different. The loaded
> sstables already have very small ranges! We just have to tidy up a bit and
> everything should be OK. STCS ignores that completely and after a while we
> have a bit less sstables but each of them covers the whole ring instead of
> just a small part. I believe that in that case letting LCS do the job is a
> better option that allowing STCS mix everything up before.
> Is there a way to disable STCS fallback? I'd like to test that scenario in
> practice during our next bootstrap...
> Does Cassandra really have to put streamed sstables at L0? The only thing we
> have to assure is that sstables at any given level do not overlap. If we
> stream different regions from different nodes how can we get any overlaps?
--
This message was sent by Atlassian JIRA
(v6.2#6252)
