Story image

Fujitsu building massive supercomputer for the University of Tokyo

18 Feb 2019

Fujitsu today announced that it has received an order for a large-scale ‘massively parallel’ supercomputer system for the Information Technology Center of The University of Tokyo.

According to the tech specialist, the new system - named ‘Oakbridge-CX’ - will achieve a theoretical peak performance of 6.6 petaflops by clustering 1,368 next-generation Fujitsu Server PRIMERGY x86 servers that feature Intel Xeon Scalable Processors.

To put that in perspective, Petaflops is short for peta floating point operations per second. Peta is an SI prefix indicating one quadrillion, or 10 to the power of 15, indicating the performance of one quadrillion floating point operations per second.

Fujitsu obviously has much expertise and technologies based in this space, as it has extensive experience of building and operating supercomputer systems for customers both in and outside of Japan. In light of this, the company is already in the process of building the supercomputer and hopes to have it operational by July this year.

The Computer Center of The University of Tokyo was established in 1965, specialising in shared use of large-scale computing by researchers in Japan. The Information Technology Center has been using supercomputers to advance cutting-edge research in science and engineering.

The Center has also played a key role in the Joint Usage/Research Center for Interdisciplinary Large-scale Information Infrastructures (JHPCN) and has contributed to R&D in a wide range of fields, including the natural sciences, engineering, and the social sciences by providing computational resources to High Performance Computing Infrastructure (HPCI).

According to The University of Tokyo, the new system is also expected to be used for field trials designed to create data utilisation platforms with a vision of achieving Society 5.0 - a human-centric society that delivers both economic development and resolutions to societal issues through systems that fuse cyberspace (virtual spaces) and physical space (the real world) at a high level.

Fujitsu is confident that by developing and offering high-performance supercomputers, the company will continue to contribute to the advancement and use of such technologies as computer science, simulations, data utilisation, and AI.

Bluzelle launches data delivery network to futureproof the edge
“Currently applications are limited to data caching technologies that require complex configuration and management of 10+ year old technology constrained to a few data centers."
DDN completes Nexenta acquisition
DDN holds a suite of products, solutions, and services that aim to enable AI and multi-cloud.
Trend Micro introduces cloud and container workload security offering
Container security capabilities added to Trend Micro Deep Security have elevated protection across the DevOps lifecycle and runtime stack.
Veeam joins the ranks of $1bil-revenue software companies
It’s also marked a milestone of 350,000 customers and outlined how it will begin the next stage of its growth.
Veeam enables secondary storage solutions with technology partner program
Veeam has worked with its strategic technology alliance partners to provide flexible deployment options for customers that have continually led to tighter levels of integration.
Veeam Availability Orchestrator update aims to democratise DR
The ability to automatically test, document and reliably recover entire sites, as well as individual workloads from backups in a completely orchestrated way lowers the total cost of ownership (TCO) of DR.
Why flash should be considered the storage king
Not only is flash storage being used for recovery, it has found a role in R&D environments and in the cloud with big players including AWS, Azure and Google opting for block flash storage options.
NVIDIA's data center business slumps 10% in one year
The company recently released its Q1 financial results for fiscal 2020, which puts the company’s revenue at US$2.22 billion – a slight raise from $2.21 billion in the previous quarter.