[
https://issues.apache.org/jira/browse/SOLR-2592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13526522#comment-13526522
]
Yonik Seeley commented on SOLR-2592:
------------------------------------
bq. That being said the implementation is going in a different direction that
will move the custom hashing configuration to be persisted within Zookeeper to
allow a client to be aware the custom hashing on the collection without having
to parse the solrconfig. I have not yet had the chance to review the work Yonik
has committed so far.
Right - I've made good progress on that front in trunk, and we'll figure out
how to get it ported back to 4x. I'm in the process of adding more tests right
now.
The separator: I went with "!" by default since it doesn't require URL
encoding, but is less common than underscore. It also reminded me of the bang
paths of old-style UUCP email addresses (like bigco!user).
A composite id router is enabled by default since the bangs are optional.
For querying, I used the param that Michael started with, "shard.keys".
So you can do shard.keys=bigco!,littleco! at query time.
(the bangs matter! leaving it out will simply query the shard containing a
simple document id, while putting it in will query a range of documents all
covered by that domain).
I've added distributed short-circuiting as well... if you only need to query a
single shard, and you send the query directly to a replica of that shard that
is active, the distributed search phase will be skipped and we'll drop directly
to a local search.
> Custom Hashing
> --------------
>
> Key: SOLR-2592
> URL: https://issues.apache.org/jira/browse/SOLR-2592
> Project: Solr
> Issue Type: New Feature
> Components: SolrCloud
> Affects Versions: 4.0-ALPHA
> Reporter: Noble Paul
> Fix For: 4.1
>
> Attachments: dbq_fix.patch, pluggable_sharding.patch,
> pluggable_sharding_V2.patch, SOLR-2592.patch, SOLR-2592_progress.patch,
> SOLR-2592_query_try1.patch, SOLR-2592_r1373086.patch,
> SOLR-2592_r1384367.patch, SOLR-2592_rev_2.patch,
> SOLR_2592_solr_4_0_0_BETA_ShardPartitioner.patch
>
>
> If the data in a cloud can be partitioned on some criteria (say range, hash,
> attribute value etc) It will be easy to narrow down the search to a smaller
> subset of shards and in effect can achieve more efficient search.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
