Computer Scientists
Don't use supercomputers
Chemists, physicists and increasingly biologists use supercomputers. If Cray is concentrating on a competition for CompScis they might be missing a bit.
There are some teens and kids in their twenties who skip out on university courses for a trip abroad with their friends, where they spend days and nights chugging down (caffeine) brew, losing sleep and wrestling with, er, clusters. But this isn't necessarily a bad thing... In fact, it could mean that they'll one day work at …
Actually, I tend to agree with you YAAC ...
My story confirms exactly what you wrote ...
I happen to hold a post-graduate diploma in super-computing, from the 90s, back at a time where parallel architectures (NUMA) were ramping-up (Cray T3D, T3E), and everyone thought vector-based systems (Cray 2, XMP, YMP etc ...) would die, along with SIMD systems (TMC was still in this business at the time).
After allowing me to work in the field for 5 years, I quickly realised super-computing was a dead end, at least in France, for the following reasons:
1- you have, in this country, only 4 potential employers for such competencies: CEA (Commisariat à l'Energie Atomique), mostly for its DAM (Division des Applications Militaires) branch, in charge of weapons, the weather forecast service, the CNRS (national research) service, and a certain government agency, which is doing that kind of thing, without too much buzz.
2- when you are working for one of the above, and you are leaving for whatever reason, you soon realise the others are only recruiting from certain sources only (read, friends)
End of the day, I changed entirely jobs, super-computing is far behind me, and I never have regretted this move. I'm paid ways over what my former colleagues are doing, just because I moved to something more widely usefull.
Super-computing is great, but unfortunately doesn't pay any bill ...
You guys both make some good points, but these Student Cluster Competitions aren't just about putting together computers and racing them. These kids have to run real scienfitic apps, running the gamut from weather forecasting to molecular dynamics to financial modeling. In order to do this, they have to investigate and understand the apps and the science behind them. Many of these students are majoring in scientific fields, which, as you point out, require some fairly deep knowledge of computer science.
"You guys both make some good points, but these Student Cluster Competitions aren't just about putting together computers and racing them. These kids have to run real scientific apps, running the gamut from weather forecasting to molecular dynamics to financial modeling. In order to do this, they have to investigate and understand the apps and the science behind them. Many of these students are majoring in scientific fields, which, as you point out, require some fairly deep knowledge of computer science."
Yes, totally agree. This was exactly my job, as described in post 2 above.
You need computer science proficiency _and_ proficiency for the science you're working for (met, nuclear dynamics, etc ...).
My point is, this kind of profile was and still is a niche market. You're not making any money, and you have no lever to ever get any good salary, again, in France, due to so few potential employers.
I'm sure other countries can be different. For example, I'm sure people working for NSA are making good money with super computing jobs, since the agency has quite a nice budget ...