Skip to Main Content
With the increasing cost of providing computing and storage facilities for research capability, it has become important that universities make increasingly efficient use of their existing resources. The strategy developed at the University of Bristol has involved the design and construction of a campus grid. This has enabled better utilization of spare capacity on available systems connected to the grid and also the separation, for the first time, of those applications that require parallel compute resources (i.e. closely coupled systems) and those that require serial resources. The overall implementation of this strategy was driven by two prime considerations. Firstly, the contributing cluster systems are, in the main, very heavily used and connecting them onto the grid must be done in a non-invasive way to ensure currently executing applications don't have to be interrupted or any other cluster operations compromised. Secondly, the grid has been considered by many scientific users to be over hyped. This implied that we needed to get a system up and running quickly however 'simple' it was to prove the concept was viable. It was hoped that involving real users at an early stage of the grid's construction would lead to them evangelising about the system and thus, encourage more users to use the system.
Date of Conference: 24-27 July 2005