August 02, 2012
Technion, the Israel Institute of Technology, has just installed a new high performance computing cluster, an SGI-manufactured system known as “Tamnun.” While the system is far less potent than elite supercomputers like Sequoia or SuperMUC, it holds the title of Israel’s fastest civilian cluster. The Times of Israel covered the new system in an article earlier this week.
Technion was founded in Haifa, Israel back in 1912 and officially began operations in 1924. Today, its campus encompasses just over half a square mile and has a population of 12,856 students. More than 70 percent of Israel’s founders and managers of high-tech industries are alumni of the institute. Technion also claims that its graduates lead 80 percent of Israeli NASDAQ companies.
Despite those credentials, there’s not a lot of supercomputing horsepower at the institute, nor in Israel as a whole. The country currently claims just three systems on the TOP500 list. The fastest machine, an IBM BladeCenter, delivers 77.7 Linpack teraflops, which placed it at number 309 on the latest rankings. While performance numbers were not revealed for the new Tamnum cluster, it’s safe to assume it delivers less than 20 teraflops, which would keep it well below TOP500 territory.
Tamnun consists of 88 compute nodes, each with a pair of 2.4 GHz, six-core Xeons and 96 GB of memory. Four servers have been equipped with NVIDIA Tesla M2090 GPUs. Storage consists of 28 nodes equipped with 500 GB drives, 52 nodes with 1 TB SATA drives, and 4 nodes with 1200 GB SAS drives. Everything is stitched together via InfiniBand, but the management network communicates over a couple of GigE switches.
This time-lapse video below shows Tamnum’s delivery and installation at the facility.
Tamnum was funded by the Russell Berrie Nanotechnology Institute and the Minerva Foundation and their users will get first crack at the machine. Typical workloads will include research in industry, engineering and medicine. The system will also give Technion the ability to participate in advanced projects at the Partnership for Advanced Computing in Europe (PRACE), which usually requires a minimum of 100,000 core hours. Israel is relatively new to PRACE, having joined the organization in January 2012.
Full story at The Times of Israel
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?