July 22, 2013
The University of Aberdeen has gone live with Maxwell, a new HPC cluster from Dell that will provide researchers with a faster, more centralized, and easier to manage HPC resource than the university previously had. The system is slated to begin a new era of cross-department collaboration and scientific discovery at the 518-year-old Scottish university.
Maxwell, which is named after the pioneering Scottish physicist James Clerk Maxwell, was commissioned to replace the aging and disparate HPC clusters, which were near their end of life. Maxwell will allow researchers to process data much faster than they could previously, and to focus on science, as opposed to spending their time administering the older HPC resources.
Maxwell was built from a combination of Dell products, including PowerEdge C6220 servers, PowerVault MD3200 and MD1200 storage, and a PowerConnect 6248 switch. The system, which is housed in the Edward Wright Data Centre and was implemented with the help of Dell and Alces Software, has a total of 608 Intel Xeon cores and 6.3 TB of memory, spread across standard memory, high memory, and very high memory nodes.
According to the university's IT Services group, each of Maxwell's 20 standard nodes is composed of two, 8-core Intel Xeon E5-2660 Sandy Bridge processors and 64 GB of total memory. Each of its 16 high memory nodes is composed of two, 8-core Intel Xeon E5-2660 Sandy Bridge processors and 256 GB of total memory. The single very high memory node has four 8-core E7 Xeon 8-core processors, and is equipped with 1 TB of memory. The storage system has a capacity of 55 TB and uses the Lustre parallel file system.
Maxwell will be used across the university's four principal interdisciplinary focus themes, including Energy, The North, Environment & Food Security, and Pathways to Healthy Living, according to a Dell UK press release. Research into these areas--and cross pollination between disciplines--will be easier now that the university has one standardized HPC system that is controlled and managed by IT professionals, as opposed to the previous model, in which departments owned and managed their own HPC resources.
"Our main aim of the project is to free researchers from the burden of system management and provide a computational service backed by a dedicated team of IT professionals," says Dr. Brian Robertson, head of infrastructure management at the University of Aberdeen. "Researchers will have more time to carry out their research rather than dealing with more mundane administrative issues. The new cluster will also open up potential for research projects across departments and encourage more collaboration to spark new and exciting opportunities."
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?