Illinois professor co-authors groundbreaking, hands-on approach for teaching parallel programming concepts

1/28/2010

Prof. Hwu and David Kirk (NVIDIA) give a hands-on approach to basic concepts of parallel programming and GPU architectures.

Written by

Illinois computer science affiliate professor Wen-mei Hwu (ECE) and David Kirk of NVIDIA teamed up in 2007 to teach one of the nation’s first university courses on programming massively parallel processors. Graduate students of all disciplines and from around the globe line up to enroll in their summer school intensives.

Now this dynamic teaching collaboration has once again produced another rich resource for students interested in high-performance parallel computing. Considered the first text of its kind, Programming Massively Parallel Processors: A Hands-on Approach teaches students the basic concepts of parallel programming and GPU architecture and prepares them to work in an industry that has moved to multicore processors.

  • Students learn to effectively program massively parallel processors using real-world case studies and actual software development tools (CUDA and OpenCL).
  • Students develop computational thinking techniques that enable them to think about problems that are amenable to high-performance parallel computing.
  • Students learn how to achieve BOTH high-performance and high-reliability using the CUDA programming model as well as OpenCL.

Harvard University professor, Hanspeter Pfister claims “This book is the most comprehensive and authoritative introduction to GPU computing yet,” and predicts Programming Massively Parallel Processors: A Hands-on Approach will “be the standard reference for years to come.”

According to Hwu, their book was written to provide students of all disciplines – not just computer science and computer engineering students – the ability to think parallel and to be able to use these techniques in their own work.

Wen-mei Hwu also serves as PI for the world’s first NVIDIA CUDA Center of Excellence at the University of Illinois, and co-director of the Intel-Microsoft funded Universal Parallel Computing Research Center (UPCRC Illinois).

David Kirk is an NVIDIA Fellow and former Chief Scientist. He led the development of graphics technology for today’s most popular consumer entertainment platforms. His contributions to graphic design and graphics technology are numerous – over 50 patents and patent applications and more than 50 published articles.

For more information about the text, visit the official website.

About Parallel@Illinois
Parallel@Illinois has been pioneering and promoting parallel computing research and education for over four decades. Current efforts include the Blue Waters petascale supercomputer, a global Cloud Computing Testbed, a CUDA Center of Excellence, a Center for Extreme Scale Computation, the Joint Laboratory for Petascale Computing, an OpenSPARC Center of Excellence, and the Universal Parallel Computing Research Center.
 


Share this story

This story was published January 28, 2010.