Github user sujithjay commented on the issue:
https://github.com/apache/spark/pull/20821
Hi @weiwee , I have a few questions about this contribution:
1. What are the use cases for which this partitioning strategy is more
suitable, compared to the existing strategies?
2. You mention a 23 % drop in vertex replication compared to
EdgePartition2D. What is your source on this?
Could you please update the above information in the JIRA ticket itself,
and maybe initiate a discussion on the dev mailing-list regarding the utility
of this partitioning strategy. Thank you.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
