[
https://issues.apache.org/jira/browse/LUCENE-5604?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13967851#comment-13967851
]
Yonik Seeley commented on LUCENE-5604:
--------------------------------------
The JVM recognizes pairs of shifts that amount to a rotate and replaces them
with an intrinsic.
bq. Initial patch, Lucene tests pass, but solrj doesn't yet compile....
Right - SolrJ does not have lucene dependencies. Solr also depends on the
*exact* hash, so it can't be tweaked (for example if a variant turns out to be
better for lucene indexing). Perhaps Lucene should just make a copy of the one
it needs (the byte[] version).
> Should we switch BytesRefHash to MurmurHash3?
> ---------------------------------------------
>
> Key: LUCENE-5604
> URL: https://issues.apache.org/jira/browse/LUCENE-5604
> Project: Lucene - Core
> Issue Type: Improvement
> Components: core/index
> Reporter: Michael McCandless
> Assignee: Michael McCandless
> Fix For: 4.9, 5.0
>
> Attachments: BytesRefHash.perturb.patch, LUCENE-5604.patch
>
>
> MurmurHash3 has better hashing distribution than the current hash function we
> use for BytesRefHash which is a simple multiplicative function with 31
> multiplier (same as Java's String.hashCode, but applied to bytes not chars).
> Maybe we should switch ...
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
