[ 
https://issues.apache.org/jira/browse/SOLR-2592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13507945#comment-13507945
 ] 

Yonik Seeley commented on SOLR-2592:
------------------------------------

Latest issue I think I'm running into is manual shard assignment... when a core 
comes up and says "hey, I'm foo-shard of bar-collection", and that collection 
is dynamically created on the fly since it doesn't exist yet.  I think we have 
to assume "custom sharding" in that case... assigning hash ranges doesn't make 
sense.  But our current indexing code happily splits up all of the shards at 
the time (without recording any hash range because numShards was never passed) 
and indexes to them based on that (which is a no-no if new shards can be 
manually added at any time).

I guess in situations like this, we should create the collection with an 
"implicit" document router.  The shard the document belongs to is defined by 
the shard it's sent to (it's custom sharding with the shard coming from the URL 
essentially).

Before I diverge too much from trunk, I think I may try to first clean up what 
I have and get that committed back (the introduction of collection properties 
and associated cleanups).
                
> 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
>         Attachments: dbq_fix.patch, pluggable_sharding.patch, 
> pluggable_sharding_V2.patch, SOLR-2592.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: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to