July 15, 2013
The University of Birmingham has taken the wraps off the second generation of its BlueBEAR supercomputer. The Linux-based cluster will be used by researchers at the UK university, as well as a wider group of computer scientists and particle physicists.
BlueBEAR II is an IBM iDataPlex system composed of about 850 Intel Sandy Bridge CPUs and a smattering of NVIDIA Quadro5000 GPUs. The server, which replaces the first generation of the server that went online in 2007, runs the Scientific Linux OS, and has a theoretical top performance of 15 teraflops.
The university is making several services available on BlueBEAR, including the general HPC services, the large memory service, the GPGPU service, and the MidPlus regional service, which provides HPC resources on other systems in the country for work that can't be accommodated on the BEAR.
According to an InformationWeek article, BlueBEAR will also offer sophisticated visualization capabilities that incorporate "active stereo display and motion tracking" and collaborative conferencing and visualization services that will be helpful to large research groups based in different countries.
The system is part of the overall Birmingham Environment for Academic Research (BEAR) project, which is a set of complementary and inter-linked services designed to meet the research needs of the institution. The system is run by the university, the supercomputer's supplier IT services firm OCF, and other specialist partners, including IBM, Adaptive Computing, Mechdyne, and IOCOM.
Academic researchers at the University of Birmingham's archaeology, economics, science, and engineering departments are lining up to get onto the new system. The chemistry department, for example, is planning to use BlueBEAR to perform nanoscience research. Among the projects planned are creating more cost-effective and environmentally friendly catalysts for fuel cells, according to InformationWeek.
The university will also make a portion of BlueBEAR available to GridPP, a collaboration of particle physicists and computer scientists from the UK and the CERN facility based near Geneva, Switzerland.
While BlueBEAR II has been online for just a few months, there are already upgrades in the works for the Linux-based system, including a "dedicated render farm" that will be a mixture of CPU and GPU nodes that is will be used for off-line rendering.
To widen the potential user base for BlueBEAR beyond researchers who are familiar with using a Linux-based HPC service, the university is also planning to make a Windows service available, according to the InformationWeek article.
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?