[
https://issues.apache.org/jira/browse/SOLR-5146?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14628874#comment-14628874
]
Shawn Heisey commented on SOLR-5146:
------------------------------------
I suspect that the coding challenges associated with getting SolrCloud working
with transient cores are, in a relative sense, not all that hard for someone
who has a good grasp on Solr internals. It might take some time and
experimentation to work through the issues, but I would not expect it to be
extremely challenging.
I think the really nasty problems are operational and exist because SolrCloud
is geared towards a setup where the entire cluster is fully online and very
responsive at all times. The process of loading a single core of moderate size
is typically pretty fast, only taking a few seconds, but that can slow down
substantially if the server is very busy, which it would be if it is frequently
unloading and loading large transient cores. In that situation, response time
can quickly grow, which can cause zkClientTimeout problems, sometimes leading
to a performance death spiral of index recoveries and exceeded timeouts. The
solution to SolrCloud death-spiral problems is usually more resources (memory,
CPU, or possibly additional servers), but that defeats one of the most
important reasons that transient cores exist -- keeping hardware costs down.
> Figure out what it would take for lazily-loaded cores to play nice with
> SolrCloud
> ---------------------------------------------------------------------------------
>
> Key: SOLR-5146
> URL: https://issues.apache.org/jira/browse/SOLR-5146
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Affects Versions: 4.5, Trunk
> Reporter: Erick Erickson
> Assignee: Shalin Shekhar Mangar
> Fix For: Trunk
>
>
> The whole lazy-load core thing was implemented with non-SolrCloud use-cases
> in mind. There are several user-list threads that ask about using lazy cores
> with SolrCloud, especially in multi-tenant use-cases.
> This is a marker JIRA to investigate what it would take to make lazy-load
> cores play nice with SolrCloud. It's especially interesting how this all
> works with shards, replicas, leader election, recovery, etc.
> NOTE: This is pretty much totally unexplored territory. It may be that a few
> trivial modifications are all that's needed. OTOH, It may be that we'd have
> to rip apart SolrCloud to handle this case. Until someone dives into the
> code, we don't know.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
