The first tangible step in the recent surge of parallelism on our campus came five years ago last month, when three students (it’s all undergraduates at St.
This year Caltech hosted the Humanity+ conference. I had been given a press pass to attend, but for a variety of reasons, plus perhaps a bit of stay at home fever on my part I chose to send my avator to the conference instead, no, actually I streamed it instead. I sat transfixed in front of my computer for most of the first day and for as as much as I had time to the second day which was unfortunately just a few hours. I am only going to touch upon some of the many ideas that were discussed, and hope I do justice to them. Please forgive me if I misquote anyone.
- Visvesvaraya Technology University and Intel Multicore Curriculum Revision
- Coming Soon-- Live streaming of Supercomputing 2010 in New Orleans
- Teaching Parallelism Through Gaming-- University of Southern California
- 3 Tips to Parallelize Your Code-- How to Avoid being Pelted by Grapefruit-sized Hailstones
- Missed the Academic Community at IDF?
Here are some things we learned:
- Parallelism can be used to describe both hardware and software... and it can be easy to confuse which one you're talking about. Clear abstractions and definitions are needed - students don't respond well to ambiguity.
Here's a quick report off my initial reactions after spending a couple of hours getting oriented to the Manycore Testing Lab (MTL) through "VIP access", from my perspective as a CS prof at a small college.
We hear from industry that they're looking for developers with experience in parallel programming. We see a demand for the processing power and speed that can only come from multiple cores working in cooperation, but university curricula, especially in the US, hasn't kept speed with the demand… Parallelism is still seen as something as reserved for HPC, PhD's, and the most devoted of gear heads.