You have until January 15, 2009 to apply for allocations of high-performance computer time and storage resources that are available through the TeraGrid for the allocation period of April 1, 2009, through March 31, 2010.
Awards are no longer separated into medium and large-sized allocations (formerly MRAC and LRAC). Awards of any size can now be submitted to any of the four quarterly allocations review meetings via POPS, the system for TeraGrid allocation requests. Start-up and education staff allocation requests are handled separately. For more information, see: https://pops-submit.teragrid.org/.
New and upgraded systems are now available. NCSA’s Lincoln, new in October, 2008, delivers peak performance of 47.5 teraflops and is designed to push the envelope in the use of heterogeneous processors for scientific computing. Also new in October, Spur, TACC’s HPC visualization cluster, enables users to access the high performance, parallel files systems on Ranger, NSF’s first “Path to Petascale” system. Close coupling with Ranger enables Spur users to visualize terascale data sets without having to transfer files from Ranger to a separate visualization resource for post-processing. Spur nodes are integrated into the InfiniBand network fabric of Ranger, leveraging a peak performance of 579.4 teraflops. In February, 2009, NICS’ Kraken will be upgraded to an XT5 system with a peak performance of more than 600 teraflops. As the most powerful system on the NSF TeraGrid, Kraken will deliver in excess of 700 million CPU hours per year and will include 16,000 compute sockets, 100 trillion bytes (100 terabytes) of memory, and 2,300 trillion bytes (2.3 petabytes) of disk volume.
For a complete catalog of TeraGrid resources, please visit: http://teragrid.org/userinfo/hardware/resources.php