[MAPREDUCE-4755] Rewrite MapOutputBuffer to use direct buffers & allow parallel sort+collect

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Not A Problem
• Affects Version/s: 3.0.0-alpha1
• Labels: optimization sort

Environment

Ubuntu 12.10 x86_64 (Bulldozer 8-core)

• Assignee: Gopal Vijayaraghavan
• Reporter: Gopal Vijayaraghavan
• Created: 27/Oct/12 08:37
• Updated: 12/May/16 18:22
• Resolved: 06/Oct/14 19:40

Description

The MapOutputBuffer has been written with a very severe constraint on the amount of memory it can consume. This results in code that has to page-in & page-out (i.e spill) data as it passes through the map buffers.

With the advent of the java.nio package, there is a fast and portable MMap alternative to handling your own buffers. This exists outside the GC space of Java and yet provides decently fast memory access to all the data.

The suggestion is that using mmap() direct buffers can be faster when a spill is involved and simpler than the current spill logic when given enough address space & uses the buffer caches to deliver best effort I/O.

Issue Links

• relates to: HAMA-559 Add a spilling message queue
• relates to: MAPREDUCE-3235 Improve CPU cache behavior in map side sort