[
https://issues.apache.org/jira/browse/SOLR-12713?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16598762#comment-16598762
]
Shawn Heisey commented on SOLR-12713:
-------------------------------------
bq. I do need this feature to improve indexing throughput in some cases
Putting different indexes on different disks (or, even better, different
servers) can improve indexing speed if all the indexes are receiving data at
the same time, but if you want to REALLY make your indexing fast, you need to
use a multi-threaded or multi-process indexing system that indexes multiple
batches of data to the same index in parallel.
> make "solr.data.dir" points to multiple base path, and collection's replicas
> may distribute in them
> ---------------------------------------------------------------------------------------------------
>
> Key: SOLR-12713
> URL: https://issues.apache.org/jira/browse/SOLR-12713
> Project: Solr
> Issue Type: New Feature
> Security Level: Public(Default Security Level. Issues are Public)
> Components: Server
> Reporter: weizhenyuan
> Priority: Major
>
> As discussion in user mail-list，It's of worthy of making "solr.data.dir"
> points to multiple base paths, which collection's replicas may distribute in.
> Currently， solr only could points to a single common dataDir，to maximize
> indexing throughput in same cases，solr maybe introduce a new feature to
> support multiple dataDir paths.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
