Evaluating MapReduce for Multi-core and Multiprocessor Systems

@article{citeulike:3023017, abstract = {This paper evaluates the suitability of the MapReduce model for multi-core and multi-processor systems. MapReduce was created by Google for application development on data-centers with thousands of servers. It allows programmers to write functional-style code that is automaticatlly parallelized and scheduled in a distributed system. We describe Phoenix, an implementation of MapReduce for shared-memory systems that includes a programming API and an efficient runtime system. The Phoenix run-time automatically manages thread creation, dynamic task scheduling, data partitioning, and fault tolerance across processor nodes. We study Phoenix with multi-core and symmetric multiprocessor systems and evaluate its performance potential and error recovery features. We also compare MapReduce code to code written in lower-level APIs such as P-threads. Overall, we establish that, given a careful implementation, MapReduce is a promising model for scalable performance on shared-memory systems with simple parallel code.}, address = {Los Alamitos, CA, USA}, author = {Ranger, Colby and Raghuraman, Ramanan and Penmetsa, Arun and Bradski, Gary and Kozyrakis, Christos }, citeulike-article-id = {3023017}, doi = {http://dx.doi.org/10.1109/HPCA.2007.346181}, journal = {hpca}, keywords = {map-reduce, multi-core}, pages = {13--24}, posted-at = {2008-07-20 17:21:47}, priority = {0}, publisher = {IEEE Computer Society}, title = {Evaluating MapReduce for Multi-core and Multiprocessor Systems}, url = {http://dx.doi.org/10.1109/HPCA.2007.346181}, volume = {0}, year = {2007} }

TY - JOUR ID - citeulike:3023017 L3 - citeulike-article-id:3023017 TI - Evaluating MapReduce for Multi-core and Multiprocessor Systems JF - hpca VL - 0 SP - 13 EP - 24 AD - Los Alamitos, CA, USA PB - IEEE Computer Society N2 - This paper evaluates the suitability of the MapReduce model for multi-core and multi-processor systems. MapReduce was created by Google for application development on data-centers with thousands of servers. It allows programmers to write functional-style code that is automaticatlly parallelized and scheduled in a distributed system. We describe Phoenix, an implementation of MapReduce for shared-memory systems that includes a programming API and an efficient runtime system. The Phoenix run-time automatically manages thread creation, dynamic task scheduling, data partitioning, and fault tolerance across processor nodes. We study Phoenix with multi-core and symmetric multiprocessor systems and evaluate its performance potential and error recovery features. We also compare MapReduce code to code written in lower-level APIs such as P-threads. Overall, we establish that, given a careful implementation, MapReduce is a promising model for scalable performance on shared-memory systems with simple parallel code. KW - map-reduce KW - multi-core AU - Ranger, Colby AU - Raghuraman, Ramanan AU - Penmetsa, Arun AU - Bradski, Gary AU - Kozyrakis, Christos PY - 2007/// UR - http://dx.doi.org/10.1109/HPCA.2007.346181 ER -