[
https://issues.apache.org/jira/browse/SOLR-8591?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Joel Bernstein updated SOLR-8591:
---------------------------------
Description:
Now that we have synchronous streaming and continuous streaming (DaemonStream),
it makes sense to add *batch streaming*.
Code will be added to the /stream handler so that when it sees the BatchStream
it will send the stream to an executor to be run.
Sample syntax:
{code}
batch(parallel(update(rollup(select()))))
{code}
The pseudo code above runs a parallel rollup in batch mode and sends the output
to a SolrCloud collection.
was:
Now that we have synchronous streaming and continuous streaming (DaemonStream),
it makes sense to add batch streaming.
Code will be added to the /stream handler so that when it sees the BatchStream
it will send the stream to an executor to be run.
Sample syntax:
{code}
batch(parallel(update(rollup(select()))))
{code}
The pseudo code above runs a parallel rollup in batch mode and sends the output
to a SolrCloud collection.
> Add BatchStream to the Streaming API
> ------------------------------------
>
> Key: SOLR-8591
> URL: https://issues.apache.org/jira/browse/SOLR-8591
> Project: Solr
> Issue Type: New Feature
> Reporter: Joel Bernstein
>
> Now that we have synchronous streaming and continuous streaming
> (DaemonStream), it makes sense to add *batch streaming*.
> Code will be added to the /stream handler so that when it sees the
> BatchStream it will send the stream to an executor to be run.
> Sample syntax:
> {code}
> batch(parallel(update(rollup(select()))))
> {code}
> The pseudo code above runs a parallel rollup in batch mode and sends the
> output to a SolrCloud collection.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
