Abstract

Cloud Computing has recently come to the fore as one of the most exciting and advanced paradigms in the world of computing. One of the most salient features of Cloud Computing is the ability to dynamically provision services to grow and contract in accordance with consumer demand. The use of virtualisation technologies enables service providers to optimise the use of resources (e.g. compute, storage, bandwidth, etc) whilst minimising operational costs. This paper explores the issues surrounding the optimal placement of data and associated processing algorithms in large scale on demand distributed infrastructures. In addition to critical network considerations such as bandwidth, parallelisation, co-location, etc, considerations about node performance, cost, storage, operating systems, control middleware, processors, and task interdependencies also need to be taken into account. Where data transfers involve very large files network performance considerations will ultimately determine the resource allocation.