I have to say that I LOVE my job, I get to talk to professors every day and learn about the cool ways that they are utilizing Intel’s technology in their research and curriculums. It is so fun to let them know about the cool tools we have available for them. Like V-tune, Thread checker and Parallel Studios.
I also have the fortune to talk to professors who have already started to THINK PARALLEL, and have started to weave parallel programming into their curriculums. A few weeks ago I was talking to Dr. Joel Adams at Calvin College in Grand Rapids, MI. and he was telling me about how he has been thinking parallel.
I want to share some of the links to his courses and work with the students at Calvin College so that you too can start to Think Parallel. In my many hours of conversations I have repeated heard “I know I need to start to do this, but it seems to be such a huge undertaking” It is, the advent of multiple processors fundamentally changes the way that Computer Sciences curriculums need to be taught. This is not a trivial undertaking. This requires us to change the way in which we think.
About a year ago, I took up running because I got tired of hearing my own litany of excuses as to why I could not exercise. In January, I joined Team in Training and ran my first ½ marathon on May 31st. I went a whole 6 days before I signed up for another ½ marathon. I am running 2 ½ marathons in October. All of this effort started with 1 step out my front door. Your journey to integrating parallelism into your curriculum starts the same way. Like my running, you simply need to start to make the changes and just keep putting one foot in front of the other.
OK, here is what Dr. Adams is doing to prepare his students for the parallel world we now live in:
“We are trying to integrate parallelism across our curriculum, introducing all students to it in the second course. Here are links to some of our resources:
The lab in our second course where we first introduce parallelism is here:
Our lab where students do this exercise has quad-core CPUs, so they see decent speedup going from 1 to 2 to 4 threads.
Our senior-level HPC course materials are here:
Students also see parallel issues in our Programming Languages course, and our OS & Networking course, but the two links above are our "bookends". Students in the HPC course (and those doing summer research) use our Beowulf cluster:
You might especially be interested in our "History" page there:
So that should give you some insight into the kinds of things going on here.”
I want to thank Dr. Adams for sharing his work with the Intel Academic Community. If you have started to Think Parallel and want to share your work, please feel free to post some links here. If you need help, contact me and I will be happy to help you get it posted.
Remember, It takes 2 or more to go parallel.