Github user viirya commented on the issue:
https://github.com/apache/spark/pull/16633
@scwf No. A simple example: if there are 5 local limit which produce 1, 2,
1, 1, 1 rows when limit is 10. If you shuffle to 5 partitions, the
distributions for each local limit look like:
1: (1, 0, 0, 0, 0)
2: (1, 1, 0, 0, 0)
3: (1, 0, 0, 0, 0)
4: (1, 0, 0, 0, 0)
5: (1, 0, 0, 0, 0)
So the final rows in 5 partitions are (5, 1, 0, 0, 0) which is not
uniformly distributed.
You don't know how many rows each local limit can get. So how do you know
how many partitions and how many rows to retrieve for each partitions?
---
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
