MSc students will have access to an impressive range of world-class parallel supercomputers and HPC technologies.
Part of EPCC’s remit has always been to provide access to leading-edge supercomputing resources for grand-challenge research projects throughout the UK.
The main HPC resource used for teaching is a dedicated shared memory machine called Ness. Students will also have access to HECToR, the UK National Service, one of the largest academic supercomputers in Europe. We also have access to machines with a wide variety of different architectures including a parallel FPGA machine (incorporating 64 FPGAs and 32 Xeon processors), a top-of-the-range multicore consumer laptop, a 2048 processor Bluegene system (which was the first of its kind to be installed in Europe back in 2005) and a new multi-processor GPGPU machine for evaluating the use of graphics processors for general HPC.
All of EPCC’s courses are taught in EPCC’s dedicated training room. Use of this room allows us to easily intermix lectures with practical sessions. The room contains 26 SunRay terminals with smartcard access that allow the user to quickly log in and move their session between machines.
Training Room
Training Facilities at EPCC, some pictures of the training room.
Below, you can find more information on some of EPCC’s larger machines which students often have access to during their projects such as HECToR, Ness,and BlueGene.
HECToR
HECToR (High End Computing Terascale Resources) is the latest national high-performance computing service for the UK academic community.
It gives UK scientists the means to undertake complex computational simulations across a range of scientific disciplines including climatology, earth sciences, chemistry, materials, fluid dynamics, atomic and molecular physics, plasma physics and nanoscience. The service began in 2008 and is expected to operate for six years, with an initial theoretical peak capability of 60 Tflop/s. The system has subsequently been upgraded and now contains 5564 quad core compute nodes (22656 cores) giving a theoretical peak performance of 208 Tflop/s. The machine also contains 112 Cray X2 vector processors.
The computer hardware was provided by Cray and it is accommodated and managed by UoE HPCX Ltd at the University of Edinburgh’s Advanced Computing Facility. NAG Ltd provides computational science and engineering support for users of the system.
http://www.hector.ac.uk
Ness
Ness is a parallel machine that is used to support local projects such as the MSc in HPC, our European visitor scheme and general research activities. Ness is supplied by Sun, and based on AMD Opteron processors running Linux.
Ness has a shared-memory architecture which allows users the option of running large threaded jobs (eg OpenMP) as well as message-passing jobs. The system has two back-end X4600 SMP nodes, both containing 16 processor-cores; there is 2GB of memory per core.
Ness has basically the same combination of processor technology, operating system and compiler suite as the new HECToR national supercomputer and is thus a useful system for us easily to investigate portability and performance of HECToR applications (if not scalability).
Documentation and further information