INTERCONNECTS

TACC Upgrades System with Faster Quad-Core AMD Opteron Processors: The Texas Advanced Computing Center (TACC) has announced that “Ranger,” its world-class Sun supercomputer, has been upgraded with new Quad-Core AMD Opteron processors. TACC now offers the national open science community a computing resource with a peak performance rating of 579.4 teraflops. “Ranger was the first major supercomputing system to use AMD’s native quad-core technology, proving its performance on large-scale science,” said Jay Boisseau, TACC director. “We’re very excited to work with AMD and Sun to upgrade Ranger’s performance with even faster AMD processors, providing the national open science community with unprecedented computing power.” All 15,744 Quad-Core AMD Opteron processors (62,976 cores) operating at 2.0 GHz have been replaced with 2.3 GHz processors, effectively adding 75.4 teraflops to the system's peak performance rating. “TACC’s recent upgrade to Ranger is a significant milestone for its open science researchers and showcases the significant advantage of the supercomputer’s design,” said Patrick Patla, general manager, AMD Server/Workstation Division. “The ability to simply and seamlessly upgrade the processor is a key benefit of AMD’s Direct Connect Architecture. In conjunction with Sun’s innovative Constellation blade design, TACC has been able to offer even more performance on a system that was already one of the world’s highest performing x86 supercomputers.” From a research standpoint, the performance improvement is anywhere from 5 to 15 percent depending on the application. With the increased number of floating point operations per second the machine can produce, users gain additional performance per CPU hour and the ability to solve bigger science problems more efficiently. “This processor upgrade gives Ranger more peak performance than other supercomputer for open science research,” Boisseau said. “In addition, Ranger has more memory [123 terabytes] than any other supercomputer in the world. Together, these capabilties make it a uniquely powerful computational resource for discovery.” Ranger went into production last Feb. 4 as the first of the new National Science Foundation (NSF) “Path to Petascale” systems. Ranger’s deployment marked the beginning of the Petascale Era in high-performance computing (HPC) where systems are approaching a thousand trillion floating point operations per second and manage a thousand trillion bytes of data. Ranger is the largest HPC computing resource on the NSF TeraGrid, a nationwide network of academic advanced computing centers that provides scientists and researchers access to large-scale computing power and resources. Ranger will provide more than 500 million processor hours of computing time to the science community each year the system is in production, roughly equivalent to running a typical dual-core desktop non-stop for 100,000 years. Ranger is also the largest system for academic research based on the Sun Constellation System, and is one of the largest open science computing systems in the world. 'Open science' refers to research that is published and the results are made publicly available for the betterment of society. Any researcher at a U.S. institution can submit a proposal to request an allocation on Ranger. The request must describe the research, justify the need for such a powerful system to achieve new scientific discoveries and demonstrate the proposer's team has the expertise to utilize the resource effectively.

• 90 percent of the system is dedicated to the NSF TeraGrid.
• 5 percent of the system is allocable to Texas higher education institutions.
• 5 percent of the system is allocable to industry through TACC’s Science and Technology Affiliates for Research (STAR) Program.

To submit a proposal to request an allocation, please visit the NSF TeraGrid Web site. Researchers at Texas higher education institutions, please contact Chris Hempel, hempel@tacc.utexas.edu. Industry contacts interested in the STAR program, please contact Melyssa Fratkin, mfratkin@tacc.utexas.edu.