VISUALIZATION

Built by ClusterVision Using Dual-Core Intel Processor-Based Dell Servers and Qlogic InfiniPath: ClusterVision, specialist in Linux supercomputer clusters, today announced the completion of one of the world's fastest supercomputers for the High Performance Computing Facility of the University of Cambridge. Built using Dell's 9th generation servers and storage units and Dual-Core Intel Xeon processors, the supercomputer will provide a central compute and storage facility to all scientists of the University of Cambridge. More than 1152 Dual-Core Intel Xeon 5160 processors provide the University of Cambridge with more than 27 TeraFLOPs (27 x 10^12 calculations per second) of raw computing power. On the TOP500 list of fastest supercomputers in the world, which was released today, the Cambridge installation is the 20th fastest supercomputer in the world and the 7th fastest in Europe. It is also the fastest cluster supercomputer in Europe based on Intel or AMD x86 architecture. Dr Paul Calleja, Director of High Performance Computing at the University: "Hosting one of the fastest supercomputers in the world is a major milestone for the Cambridge High Performance Computing Facility. I have been very pleased with the working partnership between Dell, Intel and ClusterVision which supplied the compute cluster. Dell's hardware has proven to be very robust under the computational load already placed on it for the TOP500 benchmark runs. ClusterVision's role has been key in rapidly turning the Dell-supplied hardware into a usable and manageable cluster, ready for the TOP500 benchmarks. We will soon make the cluster available to the university's academic community as a whole." Dr Alex Ninaber, Technical Director of ClusterVision, comments: "We are very proud to be part of this project building one of the fastest supercomputers in the world for one of the UK's most prestigious universities and we look forward to supporting the University of Cambridge in achieving maximum benefit from the cluster." Dr Ninaber continues: "More and more customers realise that only specialist companies like ClusterVision have the skills, experience and software to successfully manage and complete complex integration projects like these. Our ClusterVisionOS™ allowed us to very quickly deploy the right software across all compute, storage and networking hardware, diagnose problems caused by unstable hardware, and in this case, prepare the cluster to run the demanding TOP500 benchmarks." "Dell is pleased to be teaming up with Cambridge University on the UK's fastest academic supercomputer. ClusterVision played a key part in the installation and integration of the supercomputer, and will be providing first-line support services to Cambridge University. In addition to providing a central compute and storage facility to the University, the centre will also run the Dell HPC Benchmarking Programme, a no-cost service that is designed to help customers scope their technology needs", said Iain Campbell, General Manager Public Sector, Dell UK. "Dell and its partners have worked to tight timescales to build and integrate this machine to provide one of the fastest supercomputers in Europe", continued Campbell. "By using the centre, customers can receive independent advice and find out more about the capability of their equipment to further maximise their IT investment." "The result of Intel's collaboration with the University of Cambridge, ClusterVision and Dell has set a new standard in High Performance Computing. The rapid and successful deployment of the cluster and its level of performance demonstrate the unprecedented power of Dual-Core Intel Xeon processors coupled with the benefits of expert integration," said Dr. Ben Bennett, Worldwide Director of HPC Marketing Programs, Intel. Technical Note Each of the 576 Dell PowerEdge 1950 compute servers in the cluster has two Dual-Core Intel Xeon 5160 processors at 3GHz and 8GB of memory. All compute servers are interconnected by an InfiniBand network from QLogic InfiniPath host adapters and switches from SilverStorm Technologies. 60 TeraBytes of storage is provided by 18 Dell PowerVault MD1000 disk arrays connected to the cluster network over 10 Gigabit Ethernet links. The storage pool is managed by the TerraGrid parallel filesystem. The cluster runs the ClusterVisionOS, ClusterVision's own Linux-based cluster software distribution which provides an integrated software environment for the effective use and management of the system. The cluster comprises nine computational units (CUs), each with its own ClusterVisionOS Sub-Master. Each CU has 65 nodes for computation. The CUs are managed by two failover Sub-Masters. What is unique about the setup is that the CUs can run independently - if one CU fails the other CUs will keep on functioning. Also, if one of the ClusterVisionOS Main Masters fails, the cluster will keep on running. User jobs, however, are not limited to a single CU and can span multiple CUs. The construction using independent CUs allows for scaling up to far larger clusters with tens of thousands servers. Users log into one of the four DNS balanced ClusterVisionOS Login Nodes.