Jasmin supercomputer completes £4m Viglen upgrade to serve more projects

Jasmin, the environmental scientific data analysis and simulation facility, has completed a £4 million supercomputer upgrade through IT services firm Viglen.

Share

Jasmin, the environmental scientific data analysis and simulation facility, has completed a £4 million supercomputer upgrade through IT services firm Viglen.

Jasmin, run by the Science and Technologies Facilities Council (STFC) on behalf of NERC (Natural Environment Research Council), will now be able to deliver petascale processing as a result of the upgrade.

Established in 2012, Jasmin serves the environmental science community, particularly when it comes to manipulating data in connection with earth observation and creating climate and weather models. During its first year of operation Jasmin is said to have reached optimum capacity, so the upgrade will bolster its role as a system supporting varied science communities.

Existing users include the Climate and Environmental Monitoring from Space (CEMS) organisation, and the national centres of Earth Observation and Atmospheric Science.

The new 4000-core compute capability features 200 servers manufactured by Viglen, which configured a complex, multi-vendor solution using a 10GB Mellanox network, and an additional 7 Petabytes of Panasas disk capacity.

The system at STFC’s Rutherford Appleton Laboratory in Oxfordshire is said to have one of the largest data storage systems in the world. Jasmin's expansion was funded by the National Environmental Research Council (NERC), to supplement its high performance storage capability, enabling it to continue providing cutting-edge data analysis resources to UK academics and partners.

“Following its successful launch Jasmin was quickly operating at optimum capacity and we had a dire need for additional power and storage,” said Jasmin infrastructure manager Jonathan Churchill. “Viglen's integrated compute, network and storage solution is allowing us to add much improved cloud capability.”

He added: “We now have low latency network interface levels for message passing interface (MPI) programmes, which means we can run bigger parallel jobs and more of them.”

Jasmin says that while many of its projects fall within the areas of climate science and earth observation, with the extra capacity it can now additionally support wider communities and smaller projects, based on a range of cloud services. New projects include environmental genomics, hydrology and earthquake science.

"Recommended For You"

Government backs 8 top technologies to fuel growth University of Cambridge claims 'fastest' supercomputer for telescope project