[ https://issues.apache.org/jira/browse/CASSANDRA-7032?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14243148#comment-14243148 ]
Jon Haddad commented on CASSANDRA-7032: --------------------------------------- Was the original 256 number chosen in hopes that it would minimize the chance of imbalance? If so, would this patch result in recommending fewer vnodes, say 16? If so, I imagine that would result in less time consuming repair as well as improvements to the spark driver, which, as of the last time I checked, did 1 query per token to achieve data locality. I would assume 16 nodes streaming data to a single one would still achieve the benefits of vnodes, but I'm just picking a number out of the air. > 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, > 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)