University of Cambridge claims 'fastest' supercomputer for telescope project

The University of Cambridge (UoC) has expanded its partnership with Dell to help build what it claims is the UK’s "fastest" academic supercomputer.

Share

The University of Cambridge (UoC) has expanded its partnership with Dell to help build what it claims is the UK’s "fastest" academic supercomputer.

The high-performance computing (HPC) cluster has been developed to provide energy efficient support to construct the world’s largest radio telescope, the Square Kilometer Array (SKA).

The SKA project is the result of a multinational collaboration to build a telescope housing the world’s largest streaming data processor. With a projected completion date of 2018, the telescope will allow scientists to survey the sky "more than ten thousand times faster than ever before" and "map a billion galaxies out to the edge of the universe", says the SKA project.

UoC was selected, in partnership with the Science and Technology Facilities Council (STFC), to drive the system development. It therefore needed a fast, energy-efficient supercomputer to handle the research required for its new role.

UoC selected Dell, NVIDIA and Mellanox to provide a specially designed HPC cluster. Paul Calleja, director of HPC service at University of Cambridge, said: “Dell’s Fresh Air Cooling was an ideal fit for our needs. We needed to create the most efficient supercomputer in the world in order to back the prestige of the SKA project, and we couldn’t do that with energy draining technology. Dell’s server solutions let us increase compute power while decreasing our energy expenditure.”

UoC has deployed 128 Dell PowerEdge T620 servers to produce a high computational performance of 240TF (teraflops).

The University of Edinburgh’s Parallel Computing Centre (EPCC) currently claims to house the UK's most powerful supercomputer, using technology from Cray.

"Recommended For You"

OCF to deliver universities’ HPC power to businesses on demand IBM: We need scientists to take part in 'world's biggest Big Data challenge'