[
https://issues.apache.org/jira/browse/SOLR-11271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mikhail Khludnev updated SOLR-11271:
------------------------------------
Attachment: SOLR-11271.patch
* extracting {{AbstractCloudStream}}, which is extended by
{{CloudSolrStream}}.and {{FacetStream}}
* it just does facet values merging in {{FacetStream._read()}} instead of
{{FacetModule}}
* now it pass {{StreamExpressionTest.testFacetSandboxStream()}}
* {{FacetStream}} doesn't do actual streaming, yet. It pulls all tuples from
{{json.facet}}
* I have no idea how to implement all possible aggregations, but it doesn't
sounds promising. Is it possible to use one of {{json.facet}} classes inside of
StreamingExpressions?
* just a note: I suppose even if facet merging is implemented in
StreamingExpression, it hardly makes possible to merge 100M facet values from
30 shards, since I tried it with plain Java.
> merge shard facets with stream expressions
> -------------------------------------------
>
> Key: SOLR-11271
> URL: https://issues.apache.org/jira/browse/SOLR-11271
> Project: Solr
> Issue Type: New Feature
> Security Level: Public(Default Security Level. Issues are Public)
> Components: streaming expressions
> Reporter: Mikhail Khludnev
> Attachments: SOLR-11271.patch
>
>
> It's necessary for heavy facets.
> It should stream per shard facets and merge facet values with stream
> expressions.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
