[SPARK-15369] Investigate selectively using Jython for parts of PySpark

• Type: Improvement
• Status: Closed
• Priority: Minor
• Resolution: Won't Fix
• Component/s: PySpark
• Assignee: Unassigned
• Reporter: Holden Karau
• Created: 17/May/16 19:46
• Updated: 30/Nov/16 03:51
• Resolved: 13/Oct/16 20:17

Description

Transferring data from the JVM to the Python executor can be a substantial bottleneck. While Jython is not suitable for all UDFs or map functions, it may be suitable for some simple ones. We should investigate the option of using Jython to accelerate these small functions.

Issue Links

• links to: [Github] Pull Request #13571 (holdenk)