Oxford Chemistry go to next level with Streamline Computing cluster

The University of Oxford, Department of Chemistry is using a powerful new computing cluster to carry out theoretical and computational chemistry calculations.

The High Performance Computing (HPC) cluster was implemented by Streamline Computing and provides over 4TFlops of peak performance across the Dell SC1435 & R905 computing nodes based on AMD quad-core processor technology. To date, just 6 months after commissioning, the cluster has executed nearly 2 million jobs, ranging from those that take just a few 10s of minutes to those that are still executing since the day the machine was made available to users.

Dr Pete Biggs, HPC Manager for the Department of Chemistry said: “The system has worked flawlessly since the day it was commissioned.  It was fully utilized almost from day 1 running commercial, public domain and bespoke code without any problems.  The users are very happy with the performance of the system and the large scale of the machine has allowed calculations that were largely infeasible before.”

The overall Streamline Computing and Dell solution architecture is designed to deliver the maximum performance from commodity hardware supported by a dedicated Streamline Computing technical team in order to ensure high quality optimised management and scheduling of the required chemistry research environment.

These architectural elements include

  • Dedicated cluster management appliances, developed by Streamline Computing’s parent organization, Concurrent Thinking
  • A comprehensive tried-and-tested high performance computing software stack
  • An integrated application code environment to manage the deployment of software licenses  
  • Full integration of the job scheduler and the cluster management appliances for automatic job control

The cluster is being used in many Theoretical Chemistry fields including:

  • Electronic and Optical Processes in Macromolecular Systems
  • Polyamorphism
  • Nanoparticle self assembly
  • Formation of low-dimensional crystal structures
  • Complex potential energy surfaces of clusters
  • Protein self assembly 

“It has been a pleasure to architect, supply and implement this significant HPC cluster to provide the department with the opportunity to enhance their world-class scientific research in Chemistry” said Concurrent Thinking’s CEO, Dr. Michael Rudgyard. “We have once again built upon our strong relationship in this engagement with Dell and on our proven commitment to the high performance computing industry to provide the best platform possible to support the research environment at Oxford University. Our track record of delivering HPC clusters of varying size & scope, including deployment based on our unique combination of Streamline Computing skilled services and our cluster management appliances along with our understanding of the users application base ensures our customers can obtain best performance from commodity hardware components”.