THE YEAR IN PERSPECTIVE
Supercomputing continues to evolve at a rapid rate, and events at NERSC in 2003 demonstrate this quite clearly. For a few months we could claim to run the largest unclassified system in the U.S. But as we all know, these superlatives don’t last. What does last are the scientific accomplishments obtained using large systems. It is with great satisfaction that we write this introduction to a report that describes some truly astounding computational results that were obtained with the help of NERSC in 2003.
In July 2003 Dr. Raymond Orbach, Director of the DOE Office of Science, while testifying about high-end computing before the U.S. Congress, explicitly mentioned NERSC’s role in the discovery that the expansion of the Universe is accelerating, as well as the recent U.S. decision to rejoin the international collaboration to build the ITER fusion reactor. “What changed,” he said, “were simulations that showed that the new ITER design will in fact be capable of achieving and sustaining burning plasma.” From accomplishments such as these, it is clear that we really have entered the new era where computational science is an equal partner with theory and experiment in making scientific progress.
|
|
At NERSC the biggest change in 2003 was the upgrade of our current Seaborg platform from 5 to 10 Tflop/s peak performance. This created one of the largest systems in the U.S. dedicated to basic computational science. It now gives NERSC users around-the-clock access to an unprecedented 6,556 processors, coupled with one of the largest memory systems anywhere. DOE-supported computational scientists continue to have access to one of best possible resources to further the DOE mission in the basic sciences.
The upgrade to 10 Tflop/s went very smoothly, and the new system came online one month earlier than planned. The first applications test results on the upgraded Seaborg were very impressive: not only could users quickly scale several applications to 4,096 processors, but they were also able to achieve very high sustained performance rates, more than 60% of peak in a few cases. The speed of the upgrade process and the new applications results are a testimony to the expert staff at NERSC. We have again demonstrated our capability to quickly field and use a new system productively.
The expanded computational power of Seaborg made it possible to initiate a number of new projects. In the first half of 2003, NERSC started the scalability program, which gave users the opportunity to evaluate the scaling of their applications. For many users, Seaborg became the first platform on which they could explore calculations using several thousand processors.
New opportunities for computational science breakthroughs will hopefully arise in 2004 through the new “Innovative and Novel Computational Impact on Theory and Experiment” (INCITE) program at NERSC. The goal of the program is to support a small number of computationally intensive, large-scale research projects that can make high-impact scientific advances through the use of a substantial allocation of computer time and data storage at NERSC. We just completed the selection of the first three INCITE projects, and we hope to announce breakthrough results from these projects in next year’s annual report.
|
|
One of the signposts of change in 2002 was the launch of the Earth Simulator system in Japan, which reenergized the high-end computing community in the U.S., resulting in a number of new activities in 2003 that made a profound impact on computational science and high performance computing. Both the High End Computing Revitalization Task Force (HECRTF) report and the Science-Based Case for Large-Scale Simulation (SCaLeS) report demonstrated that the high-end computing community in the U.S. is well positioned for future developments and growth in supercomputing. The SCaLeS report gives us a glimpse of what simulations are possible at sustained speeds in the range of tens of teraflop/s, and the HECRTF report lays out the critical research issues that we need to address in order to reach petascale computing performance. This is exciting news, because we are thinking big again in supercomputing. We are thrilled that the DOE computational science community, in particular our NERSC users, are ready to take the lead in moving to the next level of computational science.
One small step towards this bright future will happen at NERSC in the next few months. We will have the opportunity to acquire a new computing system that will take over some of the existing user workload from Seaborg, making even more time available on Seaborg for large-scale capability computing. This will be just an intermediate step towards our next big acquisition with NERSC 5. We are looking forward to another year of both scientific and computing accomplishments at NERSC. As always, this progress would not be possible without the NERSC staff, who continue to tirelessly dedicate their time, skill, and effort to make NERSC the best scientific computing resource in the world. Our special thanks to all of you.
Horst D. Simon
NERSC Center Division Director
William T. C. Kramer
Division Deputy and NERSC Center General Manager