Hot Topics Seminar

By: Victor Eijkhout

Texas Advanced Computing Center

Date: Monday, September 22, 2014

Time: 12:00PM-1:00PM, EST

Location: Klaus 1116 E

For more information please contact Dr. Edmond Chow at echow@cc.gatech.edu

~~~~~~~~~~~~~~~~~~~~~~ 

Title:

Integrative Parallel Programming in HPC

Abstract:

With the ongoing proliferation of architecture types (distributed memory clusters, shared memory, co-processors) comes a proliferation of of programming modes (message passing, active messages, loop-based and task-based parallelism, new SIMD variants), complicating the life of a scientific programmer. To extract all possible performance, a code often has to rely on using more than one mode of parallelism, making it hard to maintain, hard to port, and far from `future-proof'.

In this talk I will present a new framework for parallel programming, called the `Integrative Model for Parallelism (IMP)'.  It is based on an abstract model of parallel computing that unifies existing models, yet is detailed enough to give performance comparable to hand-written codes in these models.

I will give a basic overview of the IMP concepts, show a prototype implementation, and indicate future directions of research.

Bio:

Victor is a long-time expert in numerical linear algebra, parallel computing,machine learning.  He holds a Ph.D. in numerical analysis from the university of Nijmegen in the Netherlands.  He has held positions at the University of Illinois Urbana-Champaign, the University of California Los Angeles and the University of Tennessee at Knoxville.  Currently he is a Research Scientist at the Texas Advanced Computing Center of The University of Texas at Austin.  He is the author of the many articles, and the `Templates' book, the textbook, ‘Introduction to High Performance Scientific Computing', and `TeX by Topic'.