cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Benedict (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-7282) Faster Memtable map
Date Thu, 05 Jul 2018 21:00:00 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-7282?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16534146#comment-16534146
] 

Benedict commented on CASSANDRA-7282:
-------------------------------------

Thanks.  I'm surprised, and will dig into this a little more myself.

However, we have to be careful with these microbenchmarks.  The whole state fits in L1 cache
here, but there's a good chance this will not be hot in L1 in live operation, given how poor
our general cache usage is; it's certainly less likely with the tree approach (given it is
much larger in memory).  So if the scores are similar (which they are), I would still prefer
the array variant.  Still, it should be outright faster, so I'll take a look and see if I
can figure out why you're seeing this behaviour.

> Faster Memtable map
> -------------------
>
>                 Key: CASSANDRA-7282
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-7282
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Benedict
>            Assignee: Michael Burman
>            Priority: Major
>              Labels: performance
>             Fix For: 4.x
>
>         Attachments: jasobrown-sample-run.txt, profile.yaml, reads.svg, run1.svg, writes.svg
>
>
> Currently we maintain a ConcurrentSkipLastMap of DecoratedKey -> Partition in our
memtables. Maintaining this is an O(lg(n)) operation; since the vast majority of users use
a hash partitioner, it occurs to me we could maintain a hybrid ordered list / hash map. The
list would impose the normal order on the collection, but a hash index would live alongside
as part of the same data structure, simply mapping into the list and permitting O(1) lookups
and inserts.
> I've chosen to implement this initial version as a linked-list node per item, but we
can optimise this in future by storing fatter nodes that permit a cache-line's worth of hashes
to be checked at once,  further reducing the constant factor costs for lookups.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org


Mime
View raw message