By DE Editors
Appro has been awarded a subcontract for Appro 1U-Tetra supercomputers from Lockheed Martin in support of the Department of Defense High Performance Computing Modernization Program (HPCMP). The HPCMP supports DoD objectives to strengthen national prominence by advancing critical technologies and expertise through use of high-performance computing.
As a subcontractor of Lockheed Martin, Appro will provide system integration, project management, support and technical expertise for the installation and operation of the supercomputers. Lockheed, as a prime contractor, will provide overall systems administration, computer operations management, applications user support, and data visualization services supporting five major DoD Supercomputing Resource Centers (DSRCs).
The following are the supercomputing centers where Appro clusters will be deployed through the end of 2010:
- Army Research Laboratory DSRC at Aberdeen Providing Ground, MD,
- US Air Force Research Laboratory DSRC at Wright Patterson AFB, OH,
- US Army Engineer Research and Development Center DSRC in Vicksburg, MS,
- Navy DoD Supercomputing Resource Center at Stennis Space Center, MS,
- Arctic Region Supercomputing Center DSRC in Fairbanks, AK.
Appro 1U-Tetra supercomputers consist of 21 cluster racks powered by eight-core AMD Opteron processors. The system is configured with Appro ACE Management nodes, compute nodes, login nodes, large shared memory nodes, and graphics nodes based on the NVIDIA Tesla GPU. As configured and deployed, the entire system provides 147.5 Tflops aggregate peak performance with 83.4Tflops powered by graphics nodes and a usable memory capacity bandwidth of more than 71.8Terrabytes/s.
All compute and graphic nodes in the system are interconnected by a dual-rail QDR Infiniband interconnect fabric that provides reduced inter-processor latency. The system also offers a bridge directly to an existing global parallel file system through Infiniband fabric allowing high bandwidth parallel access to the storage file system from all of the compute nodes.
The system comes pre-configured with the Appro Cluster Engine (ACE) Management System that supports load balanced fault-tolerant interconnect while maximizing the bandwidth associated with MPI processes, according to the company.
For more information, visit Appro.
Sources: Press materials received from the company and additional information gleaned from the company’s website.