About NERSC

The National Energy Research Scientific Computing Center (NERSC) is the flagship scientific computing facility for the Office of Science in the U.S. Department of Energy. As one of the largest facilities in the world devoted to providing computational resources and expertise for basic scientific research, NERSC is a world leader in accelerating scientific discovery through computation. NERSC is located at Berkeley Lab in Berkeley, California.

The more than 3000 computational scientists who use NERSC perform basic scientific research across a wide range of disciplines. These disciplines include climate modelling, research into new materials, simulations of the early universe, analysis of data from high energy physics experiments, investigations of protein structure, and a host of other scientific endeavors. A survey of scientific research performed at NERSC can be found in the NERSC Annual Reports.

NERSC is known as one of the best run scientific computing facilities in the world. While NERSC provides some of the largest computing and storage systems available anywhere, what distinguishes NERSC is its success in creating an environment that makes these resources effective for scientific research. NERSC systems are reliable and secure, and provide a state-of-the-art scientific development environment with the tools needed by the diverse community of NERSC users. NERSC provides intellectual services that allow computational scientists to be more effective -- consultants who are experts in computational science and performance tuning, visualization assistance, training, customized support, and other services.

If you do not already have a NERSC allocation Startup awards are available for new projects that wish to investigate using NERSC resources, or who wish to port or develop new codes. The only requirements are that your project meets the DOE Office of Science Mission and requires high performance computing resources.

NERSC resources include:

• Franklin is a Cray XT4 supercomputer with 9,660 compute nodes. Each node has quad-core AMD processors running at 2.3 GHz. Franklin has 38,640 processor cores available for scientific applications, with 8 GB of memory per node and a total 350 TB of usable disk space.
• Bassi, a 976-processor IBM p575 POWER5 system with 111 compute nodes (8 processors per node) available for scientific computing. Bassi has a peak performance of 7.4 teraflop/s and 100 terabytes of disk storage. The compute nodes are connected to each other with a high-bandwidth, low-latency switching network.
• Jacquard, a 712-processor Opteron cluster running a Linux operating system. Jacquard has a peak performance of 3.1 teraflop/s and 30 TB of disk storage. Jacquard is the first system to deploy Mellanox 12X InfiniBand uplinks in its fat-tree interconnect, reducing network hot spots and improving reliability by dramatically reducing the number of cables required.
• PDSF, a 700-processor Linux cluster with 100 TB of disk storage, used by the high energy physics community for data intensive analysis and simulations. PDSF has been in production longer than any other Linux cluster in the world, undergoing several hardware upgrades since it went online in 1998.
• DaVinci, a 32-processor SGI Altix shared memory cluster, used for visualization, data analysis, and long-running interactive jobs.
• HPSS mass storage. The NERSC HPSS system archives 3.5 petabytes (PB) of data in 61 million files, and has a capacity of 22 PB. HPSS sustains an average transfer rate of more than 100 MB/s, 24 hours per day, with peaks to 450 MB/s, to and from NERSC computational systems and to and from sources outside NERSC, such as scientific experiments.

Access to NERSC from anywhere in the U.S. and the world is available through ESnet. The production NERSC connection to ESnet is 10 gigabits per second (Gbs) ethernet via the Bay Area Metropolitan Area Network which provides dual connectivity to six DOE sites. ESnet has OC-192 (10 Gbs) bandwidth on major backbone links and peering points and OC-12 (622 Mbs) and OC-48 (2.4 Gbs) links over much of the rest of its coverage area.