Computation/HPC

Research Computation Services


Much research and some student project-work requires computational resources that cannot be satisfied by individual workstations.  Our High Performance Computing (HPC) facilities provide  facilities that can provide much greater computational performance, run-times, memory and backed-up data storage than is practical on normal workstations. These services are available either by conventional remote logins or in many cases by e-science Grid interfaces such as Globus.

For detailed information follow the links to the following major services:

  • Iridis - a Linux Cluster for Researchers.
    A large commodity computational cluster with over 1000 processor-cores, running Linux, and based on AMD Opteron servers. This facility is intended for researchers who need to run parallel jobs or large numbers of sequential jobs. We regret that we do not have the resources to be able to grant access to undergraduates or MSc students
  • Lyceum - a Linux teaching cluster
    A smaller but very capable Linux cluster intended mainly for undergraduate and MSc project work. It has a powerful head node for interactive work and 16 beefy compute nodes for longer batch jobs. 
  • Windows Condor Service - a pool of Windows PCs
    The Condor service service allows users to submit compute-intensive jobs from their own machine to a networked pool of Windows workstations.

The following Services are due to be decomissioned during 2009 and are not recommended for new users.

Access to Iridis is by application web form.

Individual MSc & undergraduate students who require access to the  Linux Teaching Cluster, Lyceum, will need to get their course tutor to email Serviceline. Alternatively course tutors can email Serviceline with a list of students who require access to the cluster.