September 07, 2011
WARWICK, England, Sept. 7 -- A new server and storage cluster based at University of Warwick's Centre for Scientific Computing (CSC) will support the simulation requirements of literally hundreds of new and existing research projects. It will enable researchers to create finer scale and more realistic simulation models, create more test scenarios and get better, more accurate predictions into areas such as Magneto Hydro Dynamics, computational fluid dynamics and the simulation of air turbulence. The processing power of the new server and storage cluster will reach peak performance of 35.75 teraflops and is 3.5 times more powerful than the University's existing 3-year old cluster. It will include 120TB's of shared storage which will enable researchers to store data more easily and securely.
Live and operational since 1st July 2011, the server and storage cluster already supports researchers from the Chemistry, Physics, Engineering, Mathematics, Statistics, Computer Science and Systems Biology Departments. Research using the server cluster includes:
The University is planning to expand access to the cluster to additional departments including: Manufacturing, Life Sciences and the Business School. Researchers can access the machine on a fair usage policy using software freely available on their own workstations and desktop PCs.
The server and storage cluster's design, integration and configuration was supplied by data processing, data management and data storage provider, OCF plc. It is part funded by the Engineering and Physical Sciences Research Council (EPSRC) and from research grants and activity in Warwick.
The server and storage cluster is built using IBM hardware including the IBM System x iDataPlex server. The server and storage cluster uses 274 IBM System x iDataPlex dx360 M3 servers each with two Intel Xeon X5650 2.66 GHz 6 Core Processors (a total core count of 3288). There are an additional six IBM iDataPlex servers with 12 Nvidia GPU processors, which will enable the University to test the benefits of using some applications on the processor. The storage solution consists of two IBM System Storage DS3512 Arrays presented to the cluster over Infiniband by two IBM System x3650 M3 Servers using IBM’s General Parallel File System.
"We started high performance computing around 10 years ago to support and progress our research," says Mark Rodger, director of the Centre for Scientific Computing (CSC), University of Warwick. "Our new server cluster will allow us to continue support for our researchers and ensure the University remains at the forefront of research with some of the latest technology available, including IBM iDataPlex servers, GPU processors and 120TB of central, shared storage."
The University of Warwick is also working with OCF on plans to re-sell any excess capacity on the server and storage cluster to UK small and medium businesses using OCF's enCORE Infrastructure on Demand service. This will enable the University to earn revenue towards the development of a new server and storage cluster.
Source: University of Warwick's Centre for Scientific Computing
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?