Phoenix++: Modular MapReduce for Shared-Memory Systems
Ashwin K Kayyoor - University of Maryland, College Park
Abstract
Authors: Justin Talbot, Richard M. Yoo, Kristos Kozyrakis (Stanford University). |
Venue: Proceedings of the second international workshop on MapReduce and its applications, (MAPREDUCE) 2011.
|
Abstract:
This paper describes our rewrite of Phoenix, a MapReduce framework for shared-memory CMPs and SMPs. Despite successfully demonstrating the applicability of a MapReduce-style pipeline to shared-memory machines, Phoenix has a number of limitations; its uniform intermediate storage of key-value pairs, ineffi cient combiner implementation, and poor task overhead amortization fail to e fficiently support a wide range of MapReduce applications, encouraging users to manually circumvent the framework. We describe an alternative implementation, Phoenix++, that provides a modular, flexible pipeline that can be easily adapted by the user to the characteristics of a particular workload. Compared to
Phoenix, this new approach achieves a 4.7-fold performance improvement and increased scalability, while allowing users to write simple, strict MapReduce code.
|
This talk is organized by Abdul Quamar