July 05, 2013
CHEMNITZ and BAYREUTH, Germany, July 5 -- The new high‐performance computer at the University of Bayreuth was inaugurated during an elaborate ceremony. The Bavarian State Minister of Science, Research and Art, Dr. Wolfgang Heubisch, and the Lord Mayor of the City of Bayreuth, Brigitte Merk‐Erbe, officially activated the system with the name "btrzx3" for its users. Offering a performance of 97.6 TeraFlop/s, the HPC cluster is currently numbered among the world's highest‐ performance computers, reaching place #486 in the renowned TOP500 list. The supercomputer was developed and installed by the German manufacturer MEGWARE.
Following on from 2005 and 2009, the Bavarian centre of higher education once again commissioned the supercomputer specialist MEGWARE with the order to deliver cutting‐edge, high‐performance technology. The high‐performance computer btrzx3 consists of a total of 12 server racks. It integrates 416 nodes, each consisting of two AMD Opteron 6348 2.8 GHz processors, and four 4‐way servers. The computer nodes have 64GB or 128GB of RAM and are connected via a fast QDR Infiniband network.
Supercomputers are of crucial importance in research. The new btrzx3 is mainly intended for highly parallel computing tasks. The computing tasks that demand a high single‐core performance can draw on the predecessor system "btrzx5", which MEGWARE also supplied. Scientists in Bayreuth are principally involved in the computation of tasks in the fields of physics, biochemistry and material sciences. The new resources are already in use at the moment to calculate the behaviour of plasma, as found on the sun's surface, in their dependence on magnetic fields. Roughly 8,000 of the over 10,000 processor cores are momentarily occupied with these calculations. The btrzx3 is in operation 24/7 in order to cater to the high demand for this kind of computing power.
The HPC cluster is housed in the newly erected laboratory and internship building in the University of Bayreuth, soon to play host to several faculties, including biochemistry, geosciences and mechatronics. The sheer size of the system and its continuous operation generate extremely high temperatures – and this waste heat is used to supply the entire building.
"Publication of the latest ranking of the world's highest‐performance supercomputers demonstrates that for the first time, four MEGWARE systems are found in the TOP500 list and the Chemnitz‐based company is clearly the number 1 among German supercomputing manufacturers", says Jana Mitschke, Marketing Manager at MEGWARE. In addition to Bayreuth, further MEGWARE TOP500 clusters are located at the universities in Oslo, Mainz and Vienna.
MEGWARE Computer Vertrieb und Service GmbH is among of Germany and Europe's leading manufacturers of high performance computing and IT technology. As a full service provider, the company operates in the fields of consulting, development, sales and service, and has realised over 650 HPC projects on behalf of companies, universities and research institutions. MEGWARE develops tailored and innovative solutions, streamlined to the needs of the clients. Please find further information under www.megware.com.
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?