SGI to Power 1.9 Petaflop Supercomputer at University of Tokyo

Print Friendly, PDF & Email

SGI_logoThe University of Tokyo has chosen SGI to perform advanced data analysis and simulation within its Information Technology Center. The center is one of Japan’s major research and educational institutions for building, applying, and utilizing large computer systems. The new SGI system will begin operation July 1, 2016.

The SGI integrated supercomputer system for data analysis and simulation will support the needs of scientists in new fields such as genome analysis and deep learning in addition to scientists in traditional areas of computational science,” said Professor Hiroshi Nakamura, director of Information Technology Center, the University of Tokyo. “The new system will further ongoing research and contribute to the development of new academic fields that combine data analysis and computational science.”

The University of Tokyo’s Information Technology Center (ITC/UTokyo) serves as a hub for the Joint Usage/Research Center for Interdisciplinary Large-scale Information Infrastructures (JHPCN), a consortium of supercomputer centers across eight leading universities. ITC/UTokyo provides more than 2,000 users inside and outside the university with access to supercomputing systems in scientific fields including engineering, earth sciences, space physics and material science. With increasing demands on its existing supercomputers for traditional scientific computing and big data analysis, ITC/UTokyo decided to invest in a more powerful SGI integrated supercomputer system for data analysis and simulation research projects in the areas such as medical science, biology, bioinformatics and image processing. In addition, the new system will function as a test environment for a future supercomputer system which will further research in machine learning, artificial intelligence and other complex areas of study.

The SGI air-cooled system consists of 540 general computing and accelerated computing nodes. The storage system contains a parallel file system with active storage of 5.04 petabytes and a fast file cache system of 209 terabytes. The system utilizes the high-speed interconnection technology, InfiniBand EDR 4x, to connect each node at 100 Gbps or faster.

Both types of computing nodes are equipped with the latest Intel Xeon E5-2695 v4 processors. In addition, each accelerated computing node is equipped with two NVIDIA Tesla P100 GPU accelerators. By efficiently combining nodes equipped with the latest CPUs and those equipped with the latest GPUs, a compact system of 540 nodes achieves high performance of up to approximately 1.9 petaflops (PFLOPS).

We are proud that the University of Tokyo’s Information Technology Center selected the SGI supercomputer for data analysis and simulation,” said Gabriel Broner, vice president and general manager of HPC, SGI. “Working together with the university, we will provide top scientists in Japan with a large-scale information infrastructure based on the most advanced technologies in the industry.”

Sign up for our insideHPC Newsletter