[
https://issues.apache.org/jira/browse/BEAM-9752?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17548512#comment-17548512
]
Danny McCormick commented on BEAM-9752:
---------------------------------------
This issue has been migrated to https://github.com/apache/beam/issues/20144
> Too many shards in GCS
> ----------------------
>
> Key: BEAM-9752
> URL: https://issues.apache.org/jira/browse/BEAM-9752
> Project: Beam
> Issue Type: Improvement
> Components: sdk-py-core, sdk-py-harness
> Reporter: Ankur Goenka
> Priority: P3
>
> We have observed case where the data was spread very thinly over
> automatically computed number of shards.
> This caused wait for the buffers to fill before sending the data over to gcs
> causing upload timeout as we did not upload any data for while waiting.
> However, by setting an explicit number of shards (1000 in my case) solved
> this problem potentially because all the shards had enough data to fill the
> buffer write avoiding timeout.
>
> We can improve the sharding logic so that we don't create too many shards.
> Alternatively, we can improve connection handling so that the connection does
> not timeout.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
