Advancing UK Research to Solve the Mysteries of Our Universe

The University of Cambridge and Durham University announce DiRAC: new supercomputers powered by the latest generation of Dell EMC PowerEdge servers.

When it comes to research, more computing power can lead to breakthrough discoveries – faster. This is the case with DiRAC, whose researchers are unlocking insights into our universe with powerful high performance computing clusters, built with the latest generation of Dell EMC PowerEdge servers.

DiRAC (Distributed Research utilising Advanced Computing) is a distributed computing facility with deployments across the UK, including at the University of Cambridge and Durham University, researching the universe’s biggest questions – such as “what is dark matter?” and “what is the universe made of?” Equipped with powerful HPC systems designed to meet the many high performance computing needs of the scientific community, DiRAC is helping researchers make new discoveries to answer these questions and more.

High-resolution simulation of a Moon-forming impact colored by the temperature. High-resolution simulation of a Moon-forming impact colored by the temperature.
(Provided by the Institute for Computational Cosmology at Durham University)

Driving advanced research and computing at the University of Cambridge

Among the first to benefit from the latest Dell EMC PowerEdge servers, the University of Cambridge will expand the Cambridge Service for Data Driven Discovery (CSD3) system, with more than 400 PowerEdge C6520 servers with the recently announced 3rd Gen Intel Xeon Scalable processors.

CSD3 will also deploy more than 80 PowerEdge XE8545 servers. Configured with 3rd Generation AMD EPYC™ processors and NVIDIA® A100 Tensor Core GPUs with NVLink™, these dual socket servers are a powerhouse for AI and advanced computing workloads – an ideal match for Cambridge’s data-driven research.

CSD3 is further enabled by NVIDIA HDR InfiniBand networking and NVIDIA HDR InfiniBand-connected BlueField-2 DPUs to offload infrastructure management such as security policies and storage frameworks from the host while providing acceleration and isolation for workloads to maximize input/output performance. Networking performance is further accelerated by NVIDIA HDR InfiniBand’s In-Network Computing engines.

Together with the existing PowerEdge servers, the CSD3 system is projected to deliver four petaflops of application performance—or four quadrillion floating point operations per second—to advance research across astrophysics, nuclear fusion power generation and lifesaving clinical medicine applications. 

Accelerating cosmology research at Durham University

First launched as a prototype in October 2020, at Durham University’s Institute for Computational Cosmology (ICC), the COSMA8 supercomputer is set to fully deploy in October 2021, with more than 90 PowerEdge C6525 servers.

Equipped with both 2nd and 3rd Generation AMD EPYC processors, Direct Liquid Cooling and NVIDIA® HDR InfiniBand® networking, the COSMA8 system will deliver the computational power and efficiency needed for the University’s data-intensive research into dark energy, black holes and how the universe was formed. With this rich mix of technologies, COSMA8 will deliver the high performance computing and storage resources researchers need to run large-scale simulations and workloads.

COSMA8 won’t stop there, as Durham University plans to expand the system to more than 600 compute nodes over the next year and support the ExCALIBUR program, addressing the challenges of exascale computing to accelerate time to discovery and innovation.

Durham University’s ICC was also recently named the newest Dell Technologies HPC & AI Center of Excellence, joining the ranks of world-class HPC facilities testing new technologies and sharing best practices across the industry.

Powering the future of research

The new supercomputers developed in collaboration with the University of Cambridge and Durham University will expand the incredibly valuable research capabilities for DiRAC’s scientific community. With the most powerful PowerEdge servers yet, researchers across the UK will be able to unlock the full value of their data to solve the world’s greatest mysteries and biggest challenges.

To learn more about how Dell Technologies HPC Solutions with Dell EMC PowerEdge servers can help customers innovate, adapt and grow, please read the Innovation Engine point of view paper.

Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries.

Pete Manca

About the Author: Pete Manca

Pete Manca is an accomplished senior executive with a history of creating, marketing, and selling high quality solutions to mission-critical enterprise customers. In his current role as Senior Vice President and General Manager of APEX, Pete is developing solutions to enable Dell Technologies’ portfolio to be sold as-a-Service. Prior roles within Dell included GM of the Converged and Hyper-Converged business unit, OEM Solutions, Tech Alliances and Integrated Solutions. Prior to Dell, Pete was CEO of Egenera, a Converged Infrastructure and Cloud Computing technology company in the Boston area.