Featured Articles

Intel refreshes CPU roadmap

Intel refreshes CPU roadmap

Intel has revealed an update to its CPU roadmap and some things have changed in 2015 and beyond. Let’s start with the…

More...
Hands on: Nvidia Shield Tablet with Android 5.0

Hands on: Nvidia Shield Tablet with Android 5.0

We broke the news of Nvidia's ambitious gaming tablet plans back in May and now the Shield tablet got a bit…

More...
Nokia N1 Android tablet ships in Q1 2015

Nokia N1 Android tablet ships in Q1 2015

Nokia has announced its first Android tablet and when we say Nokia, we don’t mean Microsoft. The Nokia N1 was designed…

More...
Marvell launches octa-core 64-bit PXA1936

Marvell launches octa-core 64-bit PXA1936

Marvell is better known for its storage controllers, but the company doesn’t want to give up on the smartphone and…

More...
Nvidia GTX 970 SLI tested

Nvidia GTX 970 SLI tested

Nvidia recently released two new graphics cards based on its latest Maxwell GPU architecture, with exceptional performance-per-watt. The Geforce GTX 970…

More...
Frontpage Slideshow | Copyright © 2006-2010 orks, a business unit of Nuevvo Webware Ltd.
Tuesday, 29 June 2010 09:05

Tokyo Institute of Technology to have 2.4-petaflop supercomputer

Written by Nick Farell
y_exclamation

Voltaire spills the beans
Voltaire, which makes scale-out data centre fabrics has spilt the beans on Japan's first 2.4-petaflop supercomputer project. Apparently it was so excited that it got a contract to provide 40 Gb/s InfiniBand switches for the TSUBAME 2.0.

Professor Satoshi Matsuoka of Global Scientific Information and Computing Centre said that the TSUBAME 2.0 will continue to push science forward by providing world-class supercomputer facilities that enable research and development to be completed and utilized more quickly than ever before.

The system is based around Intel Westmere-EP and Nehalem-EX CPUs. It is coupled with more than 4,200 of NVIDIA Tesla 20-series GPUs to provide extreme processing capabilities on each node.

He added that to capitalise on this new level of compute power, the boffins had to design a dual-rail, non-blocking fabric that can support throughput up to 80 Gb/s per node, employing two Voltaire 40 Gb/s InfiniBand connections on each node. This is where Voltaire's high-speed InfiniBand fabric were needed to achieve the highest throughput and lowest latency possible.

The project is being co-ordinated by NEC and the system will begin operation in autumn. The TSUBAME 1.0, will require only 200 square meters of floor space, a reduction by two-thirds, and will exhibit comparable power consumption thanks to numerous green technologies that will be incorporated.

The supercomputer will also be 12 times greater than the current fastest system in Japan, owned by the Japan Atomic Energy Agency. It will have more than 1,400 compute nodes, with 12 Grid Director 4700 40 Gb/s InfiniBand switches, 179 Grid Director 4036 edge switches and 6 Grid Director 4036E switches for high performance bridging to 10 GbE storage.

It will also use more than 3,000 solid-state drives (SSDs) for the world's fastest total data I/O throughput at 2/3 terabytes per second, augmented with DataDirect Networks storage system with over 7 petabytes of capacity and planned 8 petabytes of tape.

The operating system will include both Linux and Microsoft Windows HPC, and will also use cloud and virtual machine technologies in order provide flexible services to its over 2,000 users as well as facilitate advanced machine operation capabilities.

Nick Farell

E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it
blog comments powered by Disqus

 

Facebook activity

Latest Commented Articles

Recent Comments