[
https://issues.apache.org/jira/browse/SOLR-6816?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14384519#comment-14384519
]
Timothy Potter commented on SOLR-6816:
--------------------------------------
Seems like the leader is doing a lookup for the existing doc in
DistributedUpdateProcessor#versionAdd:
{code}
boolean updated = getUpdatedDocument(cmd, versionOnUpdate);
{code}
Anyway, it seemed like I was treading on dangerous ground going down that path.
Part of this re-ordering / mixing deletes / updates, etc. is why I liked the
bulkAdd parameter ... I want all this version checking safety when I need it,
but if I'm pushing in 100's of thousands of docs per second (e.g. logs), I
don't want any of that slowing me down unnecessarily. But I'll hold off on that
until I've measured the improvement of initializing the version buckets
correctly. Thanks for your continued support on this!
> Review SolrCloud Indexing Performance.
> --------------------------------------
>
> Key: SOLR-6816
> URL: https://issues.apache.org/jira/browse/SOLR-6816
> Project: Solr
> Issue Type: Task
> Components: SolrCloud
> Reporter: Mark Miller
> Priority: Critical
> Attachments: SolrBench.pdf
>
>
> We have never really focused on indexing performance, just correctness and
> low hanging fruit. We need to vet the performance and try to address any
> holes.
> Note: A common report is that adding any replication is very slow.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
