Parallel 2012 » Agenda »

Can we make sequential software rare?

Parallel hardware is ubiquitous. Parallel software is not. Software developers need to abandon their sequential ways and embrace parallel programming. Traditionally, we've tried to address this issue at the language level creating countless interesting but little used parallel languages. This approach has not worked in the past, and there is no indication it will work in now. In this keynote, I will propose a different approach based on software architectures, design pattern languages, and sophisticated software transformation tools. By moving the parallel language into the background and focusing on how software is designed, we can transform the software industry and hasten the day when sequential software is rare.

Referent

Tim Mattson

is a parallel programmer (Ph.D. Chemistry, UCSC, 1985). Tim has been with Intel since 1993 where he has worked with brilliant people on great projects including: (1) the first TFLOP computer (ASCI Red), (2) the OpenMP API for shared memory programming, (3) the OpenCL programming language for heterogeneous platforms, (4) Intel's first TFLOP chip (the 80 core Terascale chip), and (5) Intel’s 48 core, Single Chip Cloud Computer research processor. Tim has published extensively including the books Patterns for Parallel Programming (with Beverly Sanders and Berna Massingill, Addison Wesley, 2004) "An Introduction to Concurrency in Programming Languages" (with Matthew J. Sottile and Craig E Rasmussen, CRC Press, 2009), and the "OpenCL Programming Guide" (with Aaftab Munshi, Ben Gaster, James Fung, and Dan Ginsburg, Addison Wesley, summer 2011).