Beijing, will have one, most probably under the patronage of Chinese Academy of Science, where Lenovo is an incumbent with a large 1000-CPU Itanium Quadrics system right now. This will probably come on line first, since it is the central node of China National Grid.
The financial capital of Shanghai has a the go-ahead for another Petaflop, at its supercomputer centre. The incumbent there is (far less known) Dawning. Each of these is going ahead with 100+ TFLOP ‘pilot’ systems this year.
Galactic computing, set up by the well-known Steve Chen of ex-Cray fame, uses its base in the ever-prosperous city of Shenzhen, now the richest city of Guangdong province together with the capital Guangzhou.
Espected major nodes of China’s supercomputer grid. Shenzhen is not on the plan but are funding their own supercomputer
The second stage of ChinaGrid project is from 2007 to 2010, covering 30 to 40 key universities in China. The focus will extend from computational grid applications to information service grid (e-information), including applications for a distance learning grid, digital Olympic grid, etc. The third stage will be from 2011 to 2015, extending the coverage of the ChinaGrid project to all the 100 key universities. The focus of the third stage grid application will be even more diverse, including instrument sharing (e-instrument).
The EU has the phase II of the EGEE Grid. It is built on the EU Research Network GÉANT and exploits Grid expertise generated by many EU, national and international Grid projects to date.
Japan has the National Research Grid Initiative NAREGI.
The USA has Teragrid. The TeraGrid 2005-2010: In August 2005, NSF’s newly created Office of Cyberinfrastructure extended support for the TeraGrid with a $150 million set of awards for operation, user support and enhancement of the TeraGrid facility.
A list of completed grids is here
A Chinese company red Neuron has announced the completion of the Tensor MPU2016 High Performance Computing technology demonstration and development platform. It is a key part in reducing the cost of a gigaflop/sec to $250-750 by 2009.
The Tensor MPU2016, with 16 processor cards containing Freescale 8641D SoC (system on chip) processors and Xilinx Virtex-4FX FPGAs, is an ideal platform for companies which may be working on the development of high performance solutions for the embedded systems market. MPU or, Master Processing Unit, is a novel approach that provides high density and reliability while preserving CPU and interconnect flexibility. Initial performance tests resulted in achieving a HP Linpack benchmark score of 32 Gigaflops for a single chassis configuration, which triples the performance demonstrated by the prototype Tensor MPU1016 system produced by Red Neurons in the first quarter of 2007.
LANL’s plans to scale the IBM “Roadrunner” to petascale Linpack performance in a couple of years. Argonne National Laboratory has a 280-teraflop Blue Gene/L and intends to advance to a peak petaflop in a few years. So in 2010, China should have three petaflop supercomputers, Japan and the USA could each have two.
U.S. Department of Energy’s National Nuclear Security Administration (NNSA) selected IBM to design and build the world’s first supercomputer to harness the immense power of the Cell Broadband Engine (Cell B.E.) processor aiming to produce a machine capable of a sustained speed of up to 1,000 trillion calculations per second, or one petaflop.
India is also developing a supercomputer that can reach one petaflop. The project is under the leadership of Dr. Karmarkar who invented the Karmarkar’s algorithm. The Tata group of companies are funding the project. CDAC is also building a supercomputer that can reach one petaflop by 2010