The super computing market has undergone huge changes over the past four years and is now a technology open to a broad base of researchers. This is relevant to all of us because it has enabled complicated research involving modeling or analysis to be completed in a fraction of the time and money, and in ways that were not possible before. Advances in many areas such as physics, chemistry, economics and medial research rely on this technology.
Last week I caught up with Dr. Paul Calleja, Director HPC Services, from Cambridge University. Paul and his team manage a large world-class research cluster to provide central compute power to hundreds of users spread across fifteen departments. At the time it was launched in 2006, the Dell HPC solution was the fastest academic supercomputer in the country and according to Top500, ranked in the top 50 of the world’s most powerful supercomputers.
As public sector and commercial sector money is increasingly under scrutiny, broadening the affordability and removing the complexity from High Performance Cluster technology is a key area Dell is helping customers with. I asked Paul a few questions about how the Dell HPC solution is run at Cambridge including his financial model, service and performance goals.
We went on to talk about how similar services could be used in the commercial sector as well as key metrics used to judge the fitness of an HPC system as user demand grows.