April 16, 2013
OSTRAVA, April 16 — The Czech Republic has its own National Supercomputing Center. It will operate systems with total computing performance exceeding 1.2 PFLOPs. End of May 2013 the small cluster is going to be in operation, the big cluster should be operational beginning of 2015. The small cluster was delivered by company BULL, and right now a public tender is running for the delivery of the big cluster. Until the new building of the National Supercomputing Center will be finished, the small cluster is placed in a container solution called MoBull. The National Supercomputing Center IT4Innovations was founded at the regional technical university VSB – Technical University Ostrava (VSB-TU Ostrava) in the north-east part of Czech Republic. The center is a member of the key European HPC research infrastructure PRACE (Partnership for Advanced Computing in Europe) and serves as the PRACE contact point for Czech research community.
The first system that is called small cluster consists of 207 computing nodes based on Intel Xeon processors, with a total performance of 73TFLOPs and 13,2 TB of memory. The National Supercomputing Center dedicates small cluster computing resources primarily to scientific computations. Scientists can apply for the computational time through access calls that are announced twice per year. The first access call took place beginning of January 2013. „The applications for computational resources are exceeding our possible capacity“ says Branislav Jansík, the director of IT4I’s supercomputing services. Top scientists from renowned institutions in Czech Republic interested in using the supercomputer sent their applications to the first call. The resources will be used to accelerate research in nanotechnology, biomedicine, drug development, engineering as well as for automotive industry and chemistry. The small cluster will be put into operation after the acceptance tests, which will take place in April and beginning of May. “Acceptance tests will be conducted to verify functionality, performance and stability of the system.“ states the technical director of Czech branch of BULL Tomáš Hlavsa. BULL cooperates on the delivery of the first system of National Supercomputing Center IT4Innovations with Intel. "This new HPC installation marks the beginning of a new phase for Czech HPC community and creates exciting opportunities to develop cutting-edge technology that will create breakthrough innovations in science, education and software." said Pawel Gepner, Intel's EMEA HPC Platform Architect. The name of the small cluster will be selected out of 6200 proposals that were collected during public inquiry.
For the procurement of the main system of the center that is called big cluster, a public tender is currently ongoing. The computing performance is expected to exceed 1 PFLOP and the system will use novel direct hot water cooling technology. The big cluster should be operational beginning of 2015. Besides the two systems (small and big cluster), the centre is planning to purchase also a specialised system that is planned for projects of national importance and its computational power should be around 140 TFLOPs.
The total costs for production, installation and operation of small cluster including the cost of two-years rent of the mobile data center are reaching 3.6 million EUR. The total costs involved in the project of building the National Supercomputing Center IT4Innovations are roughly 70 million EUR. Part of this budget is allocated for the recently started construction of the centre’s building worth 7.5 million EUR, which will be housing the supercomputing technologies as well as the research activities.
About the National Supercomputing Center IT4Innovations
The National Supercomputing Center IT4Innovations is focusing on scientific research in the area on information technologies, with the emphasis to a research and development of supercomputing. From the very beginning another three Czech universities cooperate on the project – University of Ostrava, Silesian University and VUT Brno. Together with them also the Institute of Geonics of Czech Academy of Sciences is contributing to this unique project. Besides to the excellent research the centre runs a HPC infrastructure. In May 2013 so called small cluster with performance of 73 TFLOPs will be put into operation followed by the main computing system (big cluster) with the computational performance exceeding 1 PFLOP at the beginning of 2015. The creation and development of the National Supercomputing Center was supported by the European Social Fund, namely the operational programme Research and Development for Innovations.
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?