The machines used for testing and development during this project include the QUB Linux cluster with over 900 cores in a configuration of dual socket nodes (each with a quadcore Xeon E5530 2.40GHz) and up to 24GB RAM available per node with an Infiniband interconnect; two group owned 16 core dual socket, Xeon E5-2690 2.90GHz systems, each with 64GB of RAM and QDR Infiniband; a 16 core dual socket, Xeon E5-2650 2.00GHz machine with 16 GB of RAM and 2 Tesla K20c GPU cards having 4.8 GB RAM each. It is worthwhile to note that on the E5 range of processors, vectorised FP applications may benefit from the now expanded 256 bit registers and fused multiply add and also the new 3 argument instructions provided by Intel's AVX.