Best of Both: Tapping the Leading-edge Computational Capabilities of On-Prem HPC with Cloud Bursting

As a globally focused institution with multiple physical and virtual campuses — Liverpool, London, Suzhou, Singapore and online — the University of Liverpool’s worldwide influence and impact is unrivaled in higher education. But the University isn’t resting on its laurels. Instead, it is on a mission to build on its internationally recognized strengths in research and scholarship to increase collaborative and inter-disciplinary work.

One of the ways in which the University aims to achieve these goals is by arming its researchers and students with leading-edge computational capabilities, allowing them to solve ever-larger and ever-harder problems in less time. That is the case with the University’s groundbreaking high-performance computing environment — a hybrid system that brings together on-premise and public cloud-based resources.

As a Dell EMC case study explains, the HPC environment at the University of Liverpool includes two parallel Linux clusters that provide primary HPC resources on campus. The newer of the two systems is a cluster called Barkla, named after a University of Liverpool graduate and British physicist who won a Nobel Prize for his work in 1917 — exactly 100 years before his namesake supercomputer went into service.

The Barkla HPC cluster incorporates 105 Dell EMC PowerEdge™ servers with Intel® Xeon® Scalable processors, which enable 40 cores at 3.7GHz with 384 GB of memory. These add up to 4,200 cores and 40 TB of memory. All nodes are connected by a 100-Gbps network based on Intel® Omni-Path® Architecture (Intel® OPA).

The Barkla cluster was built by Dell EMC and Alces Flight, a company specializing in HPC software for scientists, engineers and researchers. Based in the UK, Alces Flight designs, builds and supports environments to help users make efficient use of available compute and storage resources.

In addition to the fully managed on-premise Barkla cluster, the solution designed by Dell EMC and Alces Flight provides a connection to Amazon Web Services (AWS) for additional on-demand public cloud-based computational capacity. With this cloud connection, the University comes closer to its goal of enabling researchers and students to run HPC workloads anywhere and at any time.

The combination of on-premise HPC and cloud computing is a big advantage when it comes to the user experience, according to Wil Mayers, technical director of Alces Flight.

“By being able to provide a unified HPC environment that incorporates both Dell EMC on-premise hardware and AWS, we can provide users with a high-quality, consistent experience,” Mayers says in the case study. “Our hope is that this results in further collaborative engagements that push hybrid HPC forward, allowing users to have quick access to some of the best hardware and cloud resources available.”

As for those users, they are all over the map when it comes to academic disciplines. Researchers in health and life sciences make use of HPC systems to tackle the challenges of major health and wellbeing issues in humans, animals and the environment. In the humanities and social sciences, researchers explore how human behavior shapes policy and improves society. And in science and engineering, researchers use the University’s HPC systems to develop innovative solutions in order to tackle many of today’s complex social, economic and environmental challenges.

“This isn’t just research for research sake,” notes Cliff Addison, manager of the advanced research computing team at the University of Liverpool. “Some of this research is going to come up with better factories, more effective aircraft, and better ways to help maintain the health of people and animal herds, in the UK and elsewhere around the world.”

To learn more

For the full story, read the Dell EMC case study “Hybrid HPC.” And to explore Dell EMC PowerEdge servers with Intel Xeon processors in HPC applications, visit Dell EMC solutions for HPC and AI.