Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

University of Birmingham broadens HPC service

Share this on social media:

The University of Birmingham in the UK is enhancing its HPC service by introducing a Windows-based service for Microsoft HPC-aware applications alongside the existing Linux-based service. This move will expand the availability of university’s HPC offerings to a wider variety of research disciplines and is being made possible by the advanced scheduling capabilities of Moab, Adaptive Computing’s patented workload management technology.

The Birmingham Environment for Academic Research (BEAR) is replacing its existing Linux HPC cluster with a new, more powerful and more energy-efficient Linux cluster and introducing a Windows HPC service as part of the requirement to make its services available to more groups within the university. The Windows-based system will allow users to take advantage of a more familiar interface, even potentially enabling HPC access through Microsoft Excel, which will help attract new users.

In order to achieve this, the BEAR group needed the advanced scheduling capabilities provided by the Moab HPC Suite on the Windows HPC service as well as on the Linux service. In addition to offering  policy-based control of workloads and resources, Moab enables users to work with different sets of policies to match the requirements of different customers. Other options offered by Moab, such as project-based accounting and resource allocation, are important for offering a service to a range of users with differing needs that can range from single long-running massively parallel jobs to multiple concurrent parameter-sweep serial jobs.

‘To broaden our user base, we needed a more user-friendly HPC interface, however we found that the native scheduler on the Windows system wasn’t well suited to a multi-user environment,’ said Jonathan Hunt, IT Services at the University of Birmingham. ‘In adopting the new system, we decided to stick with the Moab technology from Adaptive.’