[SPARK-1379] Calling .cache() on a SchemaRDD should do something more efficient than caching the individual row objects.

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Component/s: SQL
• Target Version/s: 1.2.0
• Assignee: Michael Armbrust
• Reporter: Michael Armbrust
• Created: 01/Apr/14 02:30
• Updated: 03/Oct/14 19:35
• Resolved: 03/Oct/14 19:35

Description

Since rows aren't black boxes we could use InMemoryColumnarTableScan. This would significantly reduce GC pressure on the workers.

Issue Links

• is related to: SPARK-3212 Improve the clarity of caching semantics