Dept of Energy CSGF magazine 2007-8 -- DOE Lab Research

Lawrence Berkeley National Laboratory

Power Play

(Page 3 of 3)

“We don’t discriminate against students,” Simon says; they’re often the people doing the nitty-gritty coding for their major professors’ research. DOE CSGF fellows who get access to NERSC can work with some powerful computers:

Seaborg, an IBM RS/6000 SP with 6,656 processors (6,080 of which are available to run scientific computing applications). The system has a peak performance of 10 teraflops, or 10 trillion operations per second.
An IBM p575 POWER5 system with 976 processors named Bassi. Bassi’s peak performance is 7.4 teraflops and it has 100 terabytes of disk storage.
A 712-processor Opteron Linux cluster named Jacquard, with a peak speed of 3.1 teraflops.
PDSF, a 700-processor Linux cluster dedicated to high-energy physics research.
DaVinci, a 32-processor SGI Altix shared memory cluster devoted to visualization, data analysis, and long-running interactive work.

NERSC also has two data storage systems:

The NERSC Global File System (NGF), with about 150 terabytes (trillion bytes) of user-accessible storage. NGF allows users to create and access a single file from any of the lab’s high-performance computing systems.
The High-Performance Storage System (HPSS), with a theoretical capacity of 22 petabytes (quadrillion bytes) for long-term archival data storage.

NERSC’s new Cray XT4, named Franklin
Click image for larger version and more information

In the fall of 2007, NERSC also will bring its latest computer system on line. Dubbed Franklin, the Cray XT4 will have 19,344 compute CPUs, at least two gigabytes of memory per CPU, and a sustained performance of 16 teraflops, as opposed to a theoretical peak performance of 100 teraflops. With future upgrades, Franklin could have a theoretical peak of 1 petaflops — one quadrillion calculations per second.

Franklin will increase the number of NERSC computer cycles available for research by a factor of 16, Simon says — and every one of them is needed. “Researchers often ask for many more processor hours than we can actually accommodate. In the last couple of years we had requests that were more than six or seven times what we had available,” he adds. Most researchers got only a fraction of the processor hours they wanted. With Franklin, “We expect we will have, for once, enough cycles to keep everybody happy” — but not for long. Demand for computing time is constantly growing.

That’s why, as soon Franklin is stabilized and running, NERSC will begin preparing for the next system, called NERSC-6 for now. NERSC-6 is likely to start life as a 1-petaflops-capable machine. It’s also likely to have even more processing cores on a single chip, a change that poses challenges for the future NERSC director.

“There’s always enough work to do,” Simon says. “There’s exciting stuff to do as long as computers grow and become more powerful. We never stand still.”