June 22, 2011
MOSCOW, June 22 — T-Platforms, an international developer of supercomputers and supplier of a full range of solutions and services for high-performance computing, announced the completion of its project to modernize Russia’s most powerful supercomputer, “Lomonosov”. As a result, the performance of the computer complex, installed at Lomonosov Moscow State University, has reached the level of 1.3 PFLOPS, unsurpassed in Russia, which positioned it to achieve 13th place in the latest edition of the Top500 most powerful supercomputers in the world.
The Lomonosov supercomputer was created in 2009 to conduct fundamental research in aerospace, nuclear, biomedical, oil and gas and many other branches of science. This computer complex, which had a capacity of 420 Tflops when it was created, has come to embody all the potential of Russia’s supercomputer industry. A host of scientific projects have been carried out with the Lomonosov complex as their basis, by order of some of the largest state and commercial corporations in the country. The ever-growing number of tasks being set, and their increasing complexity, meant that the supercomputer’s performance had to be improved. A decision was taken by the management of Moscow University computer center, in conjunction with T-Platforms’ leading specialists and developers, to undertake a two-stage project to modernize the computer complex.
The first stage, which was completed at the end of 2010, saw the supercomputer’s peak performance being upgraded to 510 Tflops. During the second stage of the project, cutting-edge TB2-TL hybrid blade systems with NVIDIA Tesla X2070 accelerators were added. In the course of modernizing the computer complex, the volume of its data bank was increased by 100 TB. Thus, the initial performance of the Lomonosov supercomputer has now been more than tripled, and a reliable technological basis has been created in order to solve the global scientific problems of the foreseeable future.
“No large scientific projects, whether in the study of global climate change, nanotechnologies, geological surveying, or the development of new materials and the analysis of their interaction, can do without supercomputers today. The complexity of these tasks is growing day by day, which requires us to boost the performance of computer complexes,” says Vsevolod Opanasenko, T-Platforms’ CEO. “In the project to modernize the Lomonosov supercomputer, we relied on T-Platforms’ state-of-the-art technological developments, which enabled us to raise the peak complex capacity to a record-breaking 1.3 Pflops. It is no wonder that such impressive results did not go unnoticed by the Top500 authors, and in its latest edition the Moscow University supercomputer came a well-deserved 13th among the world’s most powerful supercomputers.”
“The scientists at our university have seen for themselves the rich potential of our supercomputers when carrying out practical scientific research,” remarked academician Viktor Sadovnichy, President of Lomonosov MSU. “Today we can be proud of the new materials and technologies developed with the help of modeling on the supercomputer, and in the future, the next-generation high-performance complexes will serve as a basis for the majority of fundamental research. Increasing the performance of the Lomonosov computer complex will enable us to carry out this work at an absolutely new level which could not have been attained in the past.”
“Many key studies in our university are undertaken on the basis of high-performance computer complexes. Moreover, we are actively cooperating with the leading international supercomputer centers,” said Vladimir Voevodin, Deputy Scientific Director of Lomonosov Moscow State University scientific and research computer center. “However, after a while the existing facilities became insufficient for carrying out a number of scientific projects, so a decision was taken to modernize the Lomonosov supercomputer complex. In the course of this project, T-Platforms added computing modules with accelerators to the complex, thus noticeably increasing the efficiency of massively parallel computing. In this way, we have created technological reserves for our scientific activity, and the global supercomputer community has once again been treated to a demonstration of what Russia can do.”
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?