The US Department of Defense is planning to deploy one of the fastest supercomputing clusters in the world to boost the US Navy's ability to perform global scale modelling and simulation.
Based at the Naval Oceanographic Office Major Shared Resource Centre (NAVOCEANO MSRC), the 3,000-CPU, Unix-based high-performance computing (HPC) cluster is designed to triple the centre's effective computing power.
The NAVOCEANO MSRC supports of some of the world's most computationally intensive research and development projects. These include new military aircraft, ship and vehicle design, improved missile and projectile design, and advanced research in global high-resolution meteorology and oceanography.
The supercomputer consists of IBM eServer p655 systems connected with IBM's clustering technology.
Each of the new HPC systems is based on Power4+ microprocessors and runs the AIX Unix operating system environment.
The largest of the new systems is expected to run at a peak speed of 20 trillion mathematical operations per second. Typical operational availability of the systems is expected to be over 99 per cent.
"We are particularly pleased to acquire a nearly 3,000 processor system - the largest single system that we have ever fielded," said Cray Henry, director of the Department of Defense HPC modernisation programme, in a statement.
"The tremendous size of this system will allow us to explore, as never before, the limits of scalability for our key applications and our ability to harness the massive power of ultra-scale HPC systems."
Engineer calculates that Chengdu's plan to replace streetlights with artificial moonlight would cost $100bn
Dark matter holds the Universe together - and gravitational waves could help identify it
Addison Lee is working on autonomous taxis for commuting and pleasure
IBM and Technical University of Munich team demonstrate how Shor's algorithm, which can't be cracked by conventional computers, can be solved quickly with quantum computing