The title of this Seattle Weekly article is so good that I'm using it for this post. In fact, the newspaper revisits the different contracts that Cray Inc. has signed with the U.S. Department of Defense (DoD) in 2008. It says that with this $30 million commitment, Cray Inc. is helping the DoD to 'test its newest bulletproof vests, gauge the accuracy of modern missiles, and even forecast the weather on battlefields.' These contracts, which were unveiled in February 2008, concern the delivery of four XT5 massively parallel, blade-style supercomputers. One of these supercomputers is already having an upgrade: the Army Research Laboratory Major Shared Resource Center (ARL MSRC) has increased its computing capability from 100 to 200 teraflops.
You can see above a picture of one of these Cray XT5 supercomputers. (Credit: Cray XT5 product page on Cray Inc. website) Of course, you know that a Cray XT5 system named Jaguar installed at Oak Ridge National Laboratory (ORNL) is the world's fastest supercomputer for science with a peak performance of 1.64 petaflops. But did you know that ORNL has already start to install a second petaflops machine named Kraken? (Source: Frank Munger, Knoxville News Sentinel, November 15, 2008)
Now, let's go back to the Seattle Weekly article more for the tone than for exclusive revelations. The author, Rick Anderson asked Cray "what tasks, exactly, its computers are capable of performing for the Department of Defense today. Thanks but no thanks, said Cray. 'I wouldn't say the work is hush-hush,' Cray spokesperson Nick Davis said. 'But we prefer not to comment.' Davis referred us to some company press releases, which, for the average PC user, require deciphering. The notices speak of 'advanced parallel global addressing programming models,' used by 'critical defense applications in solid and structural mechanics that require extremely low latency communications.'"
Here is a link a link to a February 19, 2008 Cray Inc. press release which illustrates Anderson's view. "The Cray XT5 system is the world's most scalable Linux-based supercomputer combining unprecedented sustained application performance with exceptional manageability, lower cost of ownership and broad application and tools support to large supercomputer class machines. The Cray XT5 supercomputer delivers superior sustained application performance from a single cabinet to massive scale while offering lower power consumption, industry-leading high density packaging, innovative cooling technologies and a fully upgradeable path from Cray XT3(TM) and Cray XT4(TM) systems." Is this information or disinformation?
Now, let's read an article by Timothy Prickett Morgan for IT Jungle about these Cray contracts with Uncle Sam (February 26, 2008). "Cray said last week that the U.S. Department of Defense had awarded the company a multi-year contract, including hardware, software, and services, worth $30 million for the installation of four XT5 massively parallel, blade-style supercomputers. The XT5 and related XT5h supers are the fourth-generation kickers to the 'Red Storm' Opteron-Linux cluster that Cray created in 2002 for the U.S. Department of Energy, which manages America's nuclear weapons stockpile. The DoD is buying five supers, and four of them will be Cray XT5 machines. (See Cray Revamps Supercomputers with XT5 Designs for more details on these machines.) The DoD is in the midst of modernizing its supercomputing capability, and the Cray XT5s will be plunked down at its Army Research Laboratory, its Naval Oceanographic Office, and its Arctic Region Supercomputing Center."
Finally, let's take a peek at what the Army says in ARL MSRC Doubles its Sustained Performance. "In late 2008, the Army Research Laboratory Major Shared Resource Center (ARL MSRC) will increase its computing capability from 100 to 200 TFLOPS. This factor of two increase in capability will make the ARL MSRC one of the most capable computing centers in the Department of Defense (DoD)." Here is a short description of this new supercomputer. "the most powerful of the new systems is a Cray XT5 with 10,400 AMD Opteron processor cores (aka Barcelona). The system will have the largest memory capacity in the HPCMP at 41.6 TB, along with approximately 500 TB of fiber channel attached disk with a bandwidth of 33 GB/s. This system is scheduled for delivery on 15 September 2008 with full acceptance no later than December 11th 2008."
Sources: Rick Anderson, Seattle Weekly, December 17, 2008; and various websites
You'll find related stories by following the links below.