[
https://issues.apache.org/jira/browse/CASSANDRA-6474?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13883615#comment-13883615
]
Yuki Morishita commented on CASSANDRA-6474:
-------------------------------------------
I think we don't need so much accuracy for this. We just want to find the set
that contains SSTables of resemblance of, say, more than 0.5.
There is storage efficient improvement called b-bit minwise hashing\[1\]\[2\],
which stores only b bit (b=1, 2, 3,...) for each hash, that we can use in the
case above.
1. http://research.microsoft.com/pubs/120078/wfc0398-lips.pdf
2. http://research.microsoft.com/pubs/152334/cacm_hashing.pdf
> Compaction strategy based on MinHash
> ------------------------------------
>
> Key: CASSANDRA-6474
> URL: https://issues.apache.org/jira/browse/CASSANDRA-6474
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Yuki Morishita
> Assignee: sankalp kohli
> Labels: compaction
> Fix For: 3.0
>
>
> We can consider an SSTable as a set of partition keys, and 'compaction' as
> de-duplication of those partition keys.
> We want to find compaction candidates from SSTables that have as many same
> keys as possible. If we can group similar SSTables based on some measurement,
> we can achieve more efficient compaction.
> One such measurement is [Jaccard
> Distance|http://en.wikipedia.org/wiki/Jaccard_index],
> !http://upload.wikimedia.org/math/1/8/6/186c7f4e83da32e889d606140fae25a0.png!
> which we can estimate using technique called
> [MinHash|http://en.wikipedia.org/wiki/MinHash].
> In Cassandra, we can calculate and store MinHash signature when writing
> SSTable. New compaction strategy uses the signature to find the group of
> similar SSTable for compaction candidates. We can always fall back to STCS
> when such candidates are not exists.
> This is just an idea floating around my head, but before I forget, I dump it
> here. For introduction to this technique, [Chapter 3 of 'Mining of Massive
> Datasets'|http://infolab.stanford.edu/~ullman/mmds/ch3.pdf] is a good start.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
