Blue Planet to challenge Earth Simulator Computenik

Typical scientific applications are now able to extract only 5-10 percent of the power of US supercomputers. By contrast, the design of the Earth Simulator makes 30-50 percent of its power accessible to the majority of types of scientific applications. Hence the Earth Simulator needs considerable less hardware and considerable of less complex design to reach scientific break throughs than the US machines do.

Lawrence Berkeley and Argonne national laboratories, together with IBM, have responded to the Earth Simulator challenge with a strategic proposal, "Creating Science-Driven Computer Architecture: A New Path to Scientific Leadership". The proposal envisions a new type of development partnership with US computer vendors that goes beyond the mere evaluation of the offerings that those vendors are currently planning for the next decade.

This strategy includes development partnerships with multiple vendors. Teams of scientific applications specialists and computer scientists will work with vendor computer architects to create hardware and software environments that will allow scientists to extract the maximum performance and capability from the hardware.

Three options where explored. They were not only looking at creating the highest performance for scientific application, but were also taking into account cost considerations:

1. At the highest cost per peak teraflop/s, the first option will involve custom components at all levels in an architecture known to be successful in scientific applications, parallel vector processing. The initial stages of this effort have been announced with the evaluation of a beta-test version of the Cray X1 at Oak Ridge National Laboratory.
2. At half this price and with the promise of sustainably high cost-effectiveness, the second option will involve commercial microprocessors in a new architecture that will be programmable in the same way as the first option, ViVA or Virtual Vector Architecture. IBM will partner with Lawrence Berkeley National Laboratory to implement early versions of this architecture and deliver Blue Planet, a 160 Tflop/s mature implementation in the second half of 2005.
3. At the lowest cost per peak Tflop/s, the third option will be based on "system-on-a-chip" architecture that is being explored most visibly in the IBM Blue Gene project. This architecture is arguably the most promising for reaching the petaflop/s goal of this proposal; however, its suitability for general scientific use has not yet been demonstrated. But at half the cost of option 2 and one quarter the cost of option 1, this is path is extremely cost-effective to pursue and provides the best long-term bet currently known to the scientific community. IBM will partner with Argonne National Laboratory to develop new expressions of this architecture and deliver a 180 Tflop/s implementation appropriate for general scientific exploitation in 2005.

Blue Planet, the second option, will provide twice the sustained capability of the Earth Simulator at half the cost. The third option will provide a new architecture family for scientific computing and one that makes a definitive step towards cost-effective petaflop/s computers with high sustained levels of performance the U.S. scientists and vendors hope.

Blue Planet should have a peak of 150 Tflop/s and a sustained performance of 40-50 Tflop/s. It will have 16,384 CPUs in 2,018 nodes to achieve this power.

More information on Blue Planet can be found on: NERSC Blue Planet site. A site compiling information on the US answer to the Earth Simulator is: Ultra Simulator info site.