High Performance Computing at Newcastle University
Articles,  Blog

High Performance Computing at Newcastle University


Rocket is the University-wide High Performance Computing Service. It’s open to everybody and it’s free at the point of use. There are three pillars to research. Experiment. Theory and Modelling. HPC is essential for modelling. It’s a kind of computing enactment of many hands make light work. It’s a bit like having an office full of people. If you try and do something on one single computer it might take days and days. You have people who farm things out to different people. If you can take that same thing and break it down into little sub tasks and dish it out to a load of computers, then you might be able to do it in a couple of hours. The Rocket has over 5,000 cores, it’s split into 122 separate nodes which are all connected with a fast InfiniBand connection, and all of those share 500 terabytes of storage. The University has funded the system to the tune of £2 million. That’s being used to pay for a system and also for staff to make the system useful to people. Well the useful thing about High Performance Computing is the fact that we can do very big sums, very complicated calculations, relatively quickly. I need computers to run my simulations. With the HPC support I can get, I can make my research much more internationally leading and internationally competitive. We have a data explosion in every single aspect of science and society and we need the appropriate computational resources to be able to process this data. An HPC culture within Newcastle University would mean interdisciplinary work. People haven’t had the option in the past in the University of using HPC, so it’s making them aware that it’s there. It’s putting them in contact with each other, it’s raising the level of knowledge across the University, and it’s just making it part of everyday research. I’m a social scientist but I also have an awareness of computing and HPC and what that means in terms of Bitcoin When I’ve shared with my colleagues my research they then get a better understanding of HPC and what it can lead to. So it’s a University controlled system. It’s run by an HPC steering group, and we can provide the service much cheaper. For a lot of the big super computers that are national or regional, people have to spend a lot of time writing proposals to get access to those facilities So it’s a lot of work for for us as academics. If we have a nice, big, powerful HPC service it’s going to reduce massively queuing times. It’s much easier and nicer for us if we can just have a local machine we can use. I think this recognition of how important and how central computational research is, to the University and also nationally and internationally is the most important thing.

Leave a Reply

Your email address will not be published. Required fields are marked *