Fujitsu (News - Alert) has announced the deployment of a new supercomputer, dubbed the Oakleaf-FX. The new system is booting up at the Supercomputing Division of the University of Tokyo's Information Technology Center (SCD/ITC).
Fujitsu says the supercomputer, based on their PRIMEHPC FX10 supercomputer, is one of the few petaflops-class systems in the world, capable of performing over 1 quintillion floating-point operations per second. The theoretical peak performance of the system is 1.13 petaflops.
“Combining high performance, scalability, and reliability with superior energy efficiency, PRIMEHPC FX10 further enhances Fujitsu's technology used in the K computer, which has achieved the world's top-ranked performance,” a statement by the company said.
The supercomputer encompasses 4,800 nodes over 50 racks. For management of the system, 74 PRIMERGY servers are used. The system will use FEFS to implement a reliable, distributed file system, spread out over 234 ETERNUS file storage devices, giving 3.2 petabytes of total disk space.
For software, the system runs the the Technical Computing Suite, which includes C, C++ and FORTRAN programming languages.
The supercomputer is intended for use on scientific projects requiring a lot of number-crunching, including astrophysics, earth science, materials science, energy research, bioscience, and others. It will also be used to train both undergraduates and graduate students in the in and outs of High Performance Computing (HPC).
“Through Oakleaf-FX, SCD/ITC will pursue a variety of advanced research and development initiatives led by researchers inside and outside of the university, as well as promote the training of the next generation of leaders in the field of computational science,” the Fujistu statement said.
Priority for use of the computer among SCD/ITC’s approximately 1,500 researches both in and out of the university will be given to large projects.
“With the new system,” the company said, “Fujitsu will also play a role in SCD/ITC's activities, while leveraging the company's extensive product lineup, robust service framework, and over thirty years of HPC support know-how to provide comprehensive support in all areas of R&D and analysis work.”