High-performance computing (HPC), which has remained the exclusive playground for large enterprises and university researchers till recently, still raises lots of questions for engineers and designers. I don’t have all the answers, but I know a few people who do. One of them is Dr. Bill Nitzberg, CTO of PBS Works, Altair Engineering.
For my print article titled “On-Demand Computing with PBS Works” (October 2010), I interviewed Nitzberg’s colleague Robert Walsh, Altair’s director of business development for PBS GridWorks. In the podcast below, Nitzberg picks up where Walsh left off to explain:
- how the weakest link in a HPC cluster might drag down the whole setup;
- how you can remotely tap into Altair’s HPC resources from PBS Works;
- why Altair is currently not charging customers extra for scheduling works on the GPU;
- why estimating run time on HPC applications is such a challenge;
- and more.
About Bill Nitzberg
With over 25 years in the computer industry, spanning commercial software development to high-performance computing research, Dr. Bill Nitzberg is an internationally recognized expert in parallel and distributed computing. He has served on the board of the Open Grid Forum, co-architected NASA’s Information Power Grid, edited the MPI-2 I/O standard, and published numerous papers on distributed shared memory, parallel I/O, PC clustering, job scheduling, and Grid/Cloud computing. In his spare time, Bill tries to reduce his pack weight for his long-distance hiking trips.