You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@iceberg.apache.org by GitBox <gi...@apache.org> on 2019/10/16 20:37:50 UTC
[GitHub] [incubator-iceberg] aokolnychyi opened a new issue #553: Spark
ReadTask is expensive to serialize
aokolnychyi opened a new issue #553: Spark ReadTask is expensive to serialize
URL: https://github.com/apache/incubator-iceberg/issues/553
In some Spark jobs, we see a substantial scheduler delay. I assume it happens in `TaskSetManager` when Spark serializes Iceberg `ReadTask`. The latter contains a couple of large strings (if you have a lot of columns) and an instance of `FileIO` (which can contain a full Hadoop conf).
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@iceberg.apache.org
For additional commands, e-mail: issues-help@iceberg.apache.org