Parallel Programming Talk #80 - Professor Uzi Vishkin discusses Using Simple Abstraction to Guide the Reinvention of Computing for Parallelism

Announcing Intel Concurrent Collections for Haskell 0.1. Read the blog by Ryan Newton (Intel). Listen/watch Ryan talk with with Paul Steinberg on the Teach Parallel show.

UPCRC Illinois Summer School to be held July 19-23, 2010 at the University of Illinois at Urbana-Champaign. Prerequisites for the summer school include solid programming experience (C, C++, C# or Java languages) and a demonstrated interest in applying multicore programming to academic or professional pursuits.

If you have questions you'd like to see up discuss, ideas for show topics or just want to send fan mail.... Send Email to parallelprogrammingtalk@intel.com

Uzi Vishkin, Professor of ECE University of Maryland Institute for Advanced Computer Studies

Professor Uzi Vishkin started his work on parallel computing in 1979 as a PhD student at the Tech- nion, Israel. His initial focus was on parallel algo- rithms and parallel algorithmic thinking. He co- authored several articles that helped build a theory of parallel algorithms – also known as PRAM algo- rithms. This work led to his invention of the PRAM- On-Chip desktop supercomputer framework that scales beyond 1000 processors on a chip. He ws inducted as an ACM Fellow for playing “a lead- ing role in forming and shaping what thinking in parallel has come to mean in the fundamental theory of Computer Science.” Vishkin is currently a professor of electrical and computer engineering at the University of Maryland Institute for Advanced Computer Studies (UMIACS). He has also worked for IBM T.J. Watson, New York University, and was the chair of computer science at Tel Aviv University.

Abstract

The sudden shift from single-processor computer systems to many-processor parallel ones requires reinventing much of Computer Science (CS): how to actually build and program the new parallel sys- tems. CS urgently requires convergence to a robust parallel general-purpose platform that provides good performance and is easy enough to program by at least all CS majors. Unfortunately, lesser ease-of- programming objectives have eluded de- cades of parallel computing research. The idea of starting with an established easy parallel program- ming model and build an architecture for it has been treated as radical by vendors. This talk advo- cates a more radical idea. Start with a minimalist stepping-stone: a simple abstraction that encapsu- lates the desired interface between programmers and system builders.

Coming up next week on Parallel Programming Talk.

Watch Parallel Programming Talk every Tuesday at 8:00AM PT.

And remember, let's be thread safe out there.