Overview

Overview: The goal of this workshop is to provide a forum to discuss new and emerging general-purpose purpose programming environments and platforms, as well as evaluate applications that have been able to harness the horsepower provided by these platforms. This year's work is particularly interested on new heterogeneous GPU platforms. The final program can be found HERE. Papers are being sought on many aspects of GPUs, including (but not limited to):

+ GPU applications
+ GPU programming environments
+ GPU architectures
+ Multi-GPU systems

+ GPU compilation
+ GPU power/efficiency
+ GPU benchmarking/measurements
+ Heterogeneous GPU platforms

Keynote: Intel® Xeon Phi™ programmability: the good, the bad and the ugly.

Abstract: The talk will focus on the programming model of the Intel® Xeon Phi™ coprocessor, draw comparisons to programming CPUs on the one hand and to programming GPUs on the other hand. What does programming Xeon Phi have in common with each? Programming the Xeon Phi coprocessor involves multiple aspects, and includes utilizing many cores, explicitly utilizing the vector execution units within the cores, potentially utilizing the HW multi-threading within the cores, writing cache efficient algorithm. Optionally, it may also involve communicating with a host processor.
The talk will explore some of the technology challenges, the broader industry impact and future directions.
Presenter: Robert Geva. Title: Principle Engineer, parallel language architect, Intel.
Download slides

Registration

Please refer to the ASPLOS 2013 website for registration and local information.

Program Comittee

Tor Aamodt , University of British Columbia
Lorena Barba, Boston University
Martin Burtscher, Texas State
John Cavazos, University of Delaware
Albert Cohen, INRIA France
Greg Diamos, NVIDIA
Natalie Enright-Jerger, University of Toronto
Xin Fu, University of Kansas
Michael Gschwind, IBM Research
Michael Heroux, Sandia National Labs
Lee Howes, AMD
Byunghyun Jang, University of Mississippi
David Kaeli, Northeastern University
Paul Kelly, Imperial College
Anton Lokhmotov, ARM
Sonia Lopez Alarcon, RIT
David Luebke, NVIDIA
Nick Moore, Mathworks
Norm Rubin, AMD
Michael Shebanow, Samsung
John Stone, UIUC
Rafael Ubal, Northeastern University
Michael Vai, MIT Lincoln Labs
Sven Verdoolaege, Katholieke Universiteit Leuven
Jeff Vetter, ORNL
Sudhakar Yalamanchili, Georgia Tech
Huiyang Zhou, NC State