A study of the implementation patterns among massively threaded applications for many-core GPUs reveals that each of the seven most commonly used algorithm and data optimization techniques can enhance the performance of applicable kernels by 2 to 10× in current processors while also improving future scalability. The featured Web extra is a video interview with author John Stratton, who describes how implementation patterns can improve future scalability. (http://youtu.be/fgn9LJbInMw)

@article{stratton:2012:optimization_patterns,
author = {Stratton, John A. and Rodrigues, Christopher and Sung, I-Jui and Chang, Li-Wen and Anssari, Nasser and Liu, Geng and Hwu, Wen-mei W. and Obeid, Nady},
title = {Algorithm and Data Optimization Techniques for Scaling to Massively Threaded Systems},
journal ={Computer},
volume = {45},
number = {8},
year = {2012},
pages = {26-32},
doi = {http://doi.ieeecomputersociety.org/10.1109/MC.2012.194},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
}