[CRUNCH-3] Replicated ("map-side") joins

• Type: New Feature
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 0.3.0
• Component/s: MapReduce Patterns
• Assignee: Gabriel Reid
• Reporter: Josh Wills
• Created: 05/Jul/12 22:10
• Updated: 17/Sep/12 06:41
• Resolved: 07/Jul/12 05:59

Description

Replicated joins are a common way to improve performance when joining a large dataset with a small one. The smaller dataset is loaded into memory in the mapper/reducer tasks, and is then joined with the larger dataset as the large one is processed by the MapReduce job itself.