NERSC launches Linux Networx supercomputer into production.Linux Networx and the U.S. Department of Energy's (DOE) Office of Science announced today that DOE's National Energy Research Scientific Computing Center (NERSC NERSC - National Energy Research Scientific Computing Center (new name) NERSC - National Energy Research Supercomputer Center (old name)) at Lawrence Berkeley National Laboratory has accepted a 722-processor Linux Networx Evolocity(R) cluster system for full production use by researchers across the nation. Named "Jacquard," the Linux Networx system will provide computational resources to scientists from DOE national laboratories, universities and other research institutions to support a wide range of scientific disciplines including climate modeling, fusion energy, nanotechnology, combustion, astrophysics and life sciences. Established in 1974, NERSC is DOE's flagship facility for unclassified supercomputing. The acceptance test A test performed by the end user to determine if the system is working according to the specifications in the contract. included a 14-day availability test during which a select group of NERSC users were given full access to the Jacquard cluster to thoroughly test the entire system in production operation. Jacquard had a 99 percent availability uptime during the testing while users and scientists ran a variety of codes and jobs on the system. The thorough acceptance testing (programming) acceptance testing - Formal testing conducted to determine whether a system satisfies its acceptance criteria and thus whether the customer should accept the system. by NERSC ensures Jacquard is ready for a production environment for thousands of scientists and researchers across the nation. "NERSC is the leading provider of computing resources for DOE's Office of Science and this new system will provide valuable computational science support for a wide range of users, allowing them to run more detailed simulations with faster turnaround, thereby helping advance scientific discovery," said NERSC General Manager Bill Kramer. The Jacquard system is one of the largest production InfiniBand-based Linux cluster systems and has met rigorous acceptance criteria for performance, reliability and functionality. Jacquard also takes advantage of Mellanox 12X InfiniBand uplinks in its fat-tree interconnect, reducing network hot spots and improving reliability by dramatically reducing the number of cables required. The system has 722 dual AMD Opteron(TM) processors, Model 248, with 640 processors devoted to computation, and the rest used for I/O, interactive work, testing and interconnect management. Jacquard has a peak performance of 3.1 trillion floating point operations per second (teraflop/s). Storage from DataDirect Networks provides 30 terabytes of globally available formatted storage. "By delivering this system to NERSC, we've provided a highly productive computing system to over 2,500 users nationwide," said Robert (Bo) H. Ewald, CEO of Linux Networx. "We are committed to providing NERSC with the most advanced high-performance computing system available and are thrilled that this system will be a key part of major research initiatives taking place throughout the country." Following the tradition at NERSC, the system was named for someone who has had an impact on science and/or computing. In 1801, Joseph-Marie Jacquard invented the Jacquard loom The Jacquard Loom The binary principle embodied in the punched-card operation of the loom was inspiration for the data processing machines to come. (Image courtesy of The Computer Museum History Center, www.computerhistory.org), which was the first programmable machine. The Jacquard loom used punched cards and a control unit that allowed a skilled user to program detailed patterns on the loom. |
|
||||||||||||||||||||

Printer friendly
Cite/link
Email
Feedback
Reader Opinion