[jira] [Commented] (CASSANDRA-7032) Improve vnode allocation

["Branimir Lambov (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-7032?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231419#comment-14231419 ] 

Branimir Lambov commented on CASSANDRA-7032:
--------------------------------------------

[~benedict], can you point me to some more information on the imbalance that is known to appear and its behaviour with increasing number of nodes? I'd like to get a better understanding of the problem, and how much of it is caused by replica selection rather than token imbalance.

It seems to me that the best approach here is to build in some replication strategy / network topology awareness into the algorithm to be able to account for replica selection. This will complicate the algorithm but, in addition to getting better balance, could also improve the time spent finding replicas (e.g. CASSANDRA-6976).

> Improve vnode allocation
> ------------------------
>
>                 Key: CASSANDRA-7032
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-7032
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Benedict
>            Assignee: Branimir Lambov
>              Labels: performance, vnodes
>             Fix For: 3.0
>
>         Attachments: TestVNodeAllocation.java, TestVNodeAllocation.java
>
>
> It's been known for a little while that random vnode allocation causes hotspots of ownership. It should be possible to improve dramatically on this with deterministic allocation. I have quickly thrown together a simple greedy algorithm that allocates vnodes efficiently, and will repair hotspots in a randomly allocated cluster gradually as more nodes are added, and also ensures that token ranges are fairly evenly spread between nodes (somewhat tunably so). The allocation still permits slight discrepancies in ownership, but it is bound by the inverse of the size of the cluster (as opposed to random allocation, which strangely gets worse as the cluster size increases). I'm sure there is a decent dynamic programming solution to this that would be even better.
> If on joining the ring a new node were to CAS a shared table where a canonical allocation of token ranges lives after running this (or a similar) algorithm, we could then get guaranteed bounds on the ownership distribution in a cluster. This will also help for CASSANDRA-6696.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)