"Superconductor" => Big Data and GPUs, Amber Update
- Feb 24, 2014 · 6:15 PM
Come for Table discussions, Member Self-Intro, What's New, Application Showcase, and Advanced Application Development Techniques! Exchange ideas, meet experts, share code... all HPC & GPU, all practical, all cutting-edge.
Agenda:
General Discussion:
6:15-6:50pm: What’s new and first-time attendee intros
Main Program:
7:00-7:50pm: Superconductor: A Language for Scaling Charts with Design and GPUs (Leo Meyerovich, Ph.D., Founder of Graphistry)
8:00-8:30pm: Extreme Acceleration of Molecular Dynamics: An Update (Scott Le Grand, Ph.D., Amazon)
Superconductor: A Language for Scaling Charts with Design and GPUs
Visualization is a weak link for big data tools: shoving 1MM rows into standard charts breaks their visual design and kills interactivity. In our mission to scale charts, we built the Superconductor language. It automatically compiles declarative visualizations into GPU code (WebCL+WebGL). This talk will explore how we’re redesigning and optimizing core charts for time series and graph data. Superconductor achieves several magnitudes of speedup over traditional JavaScript, which we then leverage to build interactive visualizations with high data visibility and information density.
The Superconductor compiler can be downloaded at http://www.sc-lang.com and we’re launching http://www.graphistry.com to provide accelerated charts built on top of it.
Bio:
Leo Meyerovich is the founder of Graphistry, which provides interactive visualizations of 2-3 magnitudes more data in web browsers by exploiting multicore and GPU hardware. Previously, he researched parallel computing and language design at UC Berkeley. Past projects include the Superconductor language for big data visualization, the first parallel web browser, the first functional reactive web language Flapjax, and the Margrave and ConScript languages for secure web programming. His research has been incorporated into commercial web browsers and frameworks and received awards at OOPSLA 2013, PLDI 2012, and OOPSLA 2009. If you need to show a lot of time series or graph data in the browser, he's your guy.
Extreme Acceleration of Molecular Dynamics: An Update
Bio:
Scott Le Grand is Principal Engineer at Amazon developing life science services for the Elastic Compute Cloud (EC2). He is named a CUDA Fellow in 2011 and has ported AMBER to NVIDIA Kepler GPUs, achieving a 50+% performance boost. 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. In 2011, 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. 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.
Location:
Open space;
Carnegie Mellon Silicon Valley;
NASA Research Park Bldg 23;
Mountain View, CA 94043;
Directions to Carnegie Mellon Silicon Valley;
Google Map 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
37 went
More
Our Sponsors
-
NVIDIA
Ongoing; Host for 2015 Meetups
People in this
Meetup are also in:
-
Silicon Valley Deep Learning Group
4,244 Deep Learners
-
Silicon Valley Cloud Computing
9,230 Cloudsters
-
Silicon Valley Automotive Open Source
2,360 Members
-
Silicon Valley Big Data Science
7,863 Makers
-
H2O AI & Deep Learning
2,545 Makers
-
IBM Watson and Cognitive Computing
2,782 Members
Will the slides be posted? I was unfortunate to miss this event...