[
https://issues.apache.org/jira/browse/SOLR-16448?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646910#comment-17646910
]
Shawn Heisey commented on SOLR-16448:
-------------------------------------
[~krisden] Maybe get rid of ALL the white space. I think json is smaller than
just about any format that's not binary, if it is completely minified. Noble's
gzip idea sounds promising as an additional step. The nice thing with the
current format is that the tree in ZK is completely human readable ... so I
think the admin UI needs a way to parse whatever we finally decide to use back
into a pretty human-readable format.
[~noble.paul] Wow. Hard to imagine a cluster that large. The costs for
building and running that cluster must be quite high. Does ZK handle the
binary gzip data OK? I'd be worried about it needing to be base64 encoded or
something.
> State.json serialized with indent size 2, this takes lots of extra bytes for
> large collection
> ----------------------------------------------------------------------------------------------
>
> Key: SOLR-16448
> URL: https://issues.apache.org/jira/browse/SOLR-16448
> Project: Solr
> Issue Type: Improvement
> Security Level: Public(Default Security Level. Issues are Public)
> Components: SolrCloud
> Affects Versions: 9.1
> Reporter: Hitesh Khamesra
> Assignee: Noble Paul
> Priority: Major
>
> For large collection (#shards > 2000), state.json file size becomes more than
> 1 mb. Solr serialized that with indent size 2. By reducing indent size to 0,
> we can reduce size by more than 20%. That will be very efficient.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@solr.apache.org
For additional commands, e-mail: issues-h...@solr.apache.org
