Github user holdenk commented on the issue: https://github.com/apache/spark/pull/14579 Sure - so at a bit of a high level and not like exactly on point - copying data out of the Python back to the JVM is kind of slow so if we have multiple python operations we can put together in the same task then we try and do this. Since caching is handled by storing the data inside of the JVM a cached RDD can't be pipelined since we need to copy the result to the JVM. You can see the details in the PipelinedRDD in rdd.py.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org