HPC & GPU Supercomputing Group of Silicon Valley

Special topic: Supercomputer in a Desktop: Extreme Acceleration of Molecular Dynamics Using NVIDIA GPUs ( by Scott Le Grand, Ph.D. ) Exchange ideas, meet experts, share code, lectures ... all HPC & GPU, all practical, all cutting-edge.

http://photos1.meetupstatic.com/photos/event/4/b/6/e/event_21799310.jpeg

1. Take Moffett Blvd exit on Hwy 101; on Moffett Blvd, head North.

2. Stop at check point (for NASA Ames Research Park)

• Show Government Issued ID: e.g. Driver's License
• Tell the Guard you are going to Carnegie Mellon Silicon Valley

3. Park at the location labeled as "P" in the map below

• We are in Building 23

Agenda:

General Discussions:

6:15-6:30pm What’s new in HPC & GPU Supercomputing

6:30-6:50pm Member self-intros: 30 seconds for each member

Main Program:

6:50-7:40pm Supercomputer in a Desktop: Extreme Acceleration of Molecular Dynamics Using NVIDIA GPUs ( by Scott Le Grand, Ph.D. )

7:40-7:50pm Discussions

7:50-8:10pm Break Your Multicore Program Repeatedly to Bust Bugs (by Roni Simonian)

8:10-8:20pm Discussions

8:20-8:30pm Book Review: GPU Computing Gems

Supercomputer in a Desktop: Extreme Acceleration of Molecular Dynamics Using NVIDIA GPUs

Biography:
Scott Le Grand is a recent hire at Google. Previously, he was a principal engineer on the CUDA software team at NVIDIA. Scott developed the first molecular modeling system for home computers, Genesis, in 1987, Folderol, the distributed computing project targeted at the protein folding problem in 2000, and BattleSphere, a networkable 3D space shooter for the Atari Jaguar the same year. Surprisingly, all three of these efforts shared a common codebase. More recently, he ported the Folding@Home codebase to CUDA, achieving a 5x speedup over previous efforts, and which currently accounts for ~2.6 petaFLOPs of the project’s computational firepower, and he is in the final stages of porting AMBER to CUDA as well. In a previous life, he picked up a B.S. in biology from Siena College and a Ph.D. in biochemistry from the Pennsylvania State University. In addition, he has written chapters for ShaderX and GPU Gems 3 and co-edited a book on computational methods for protein structure prediction.

Break Your Multicore Program Repeatedly to Bust Bugs

Abstract:
One of the biggest challenges facing developers and testers of concurrent programs is the programs' non-deterministic behavior. Scheduling of threads or processes is affected by a large number of unrelated asynchronous events. An intermittent failure may or may not be captured during a test. Even if such failure is captured, it does not help debugging because in most cases there is no mechanism to reproduce it. In this talk I will demonstrate how these bugs can be found and reproduced in randomized yet controlled deterministic environment. I will introduce Maze, a new development tool which is an implementation of this environment on X86 Linux platform.

Biography:
Roni has over 13 years of software development experience in electronic design automation and computer graphics industries. In 2008 she left EDA to found Ariadne LLC ( http://kloobok.com (http://kloobok.com/)) and develop Maze, a tool for testing and debugging parallel applications. Roni has M.S. in Computer Science from SUNY at Stony Brook and a B.S. in Physics and Applied Math from Moscow Institute of Physics and Technology.

Location:

Room 109/110;
Carnegie Mellon Silicon Valley;
NASA Research Park Bldg 23;
Mountain View, CA 94043;

Directions (http://www.cmu.edu/silicon-valley/about-us/directions.html) to Carnegie Mellon Silicon Valley;

Google Map (http://maps.google.com/maps/ms?gl=us&hl=en&ie=UTF8&msa=0&ll=37.410941,-122.063169&spn=0.019191,0.048923&t=h&z=15&msid=215438781255871976989.00049cacf6f0e5596e5cc) showing parking, check point, and building entrance;

NOTE: You will need a government issued ID (e.g. Driver's License) to enter NASA Research Park