Parallel I/O Architecture of The First ASCI TFLOPS Machine
نویسنده
چکیده
Intel’s Teraflops(TFLOPS) machine is the first ASCI machine that DOE is acquiring. This computer will be 10 times more powerful than the fastest machine today, and will be used primarily to simulate nuclear testing and to ensure the safety and effectiveness of the nation's nuclear weapons stockpile. This machine will contain over 9000 Intel’s Pentium® Pro processors and will provide a peak cpu performance of 1.8 teraflops. To balance the processing power and to avoid I/O bottleneck, the machine is capable of sustained disk transfer rate of 1 GB/second. This paper presents the parallel I/O overview and design of the Intel’s TFLOPS supercomputer, and discusses the architecture and implementation of Cougar I/O subsystem and its interface with Intel’s Parallel File System.
منابع مشابه
ASCI Red - Experiences and Lessons Learned with aMassively Parallel
The Accelerated Strategic Computing Initiative (ASCI) is focused upon advancing three-dimensional, full-physics calculations to the point where \full-system" simulation may be applied to virtual testing. The ASCI program involves Sandia, Los Alamos and Lawrence Livermore National Laboratories. At Sandia National Laboratories, ASCI applications include large deformation transient dynamics, shock...
متن کاملAn Overview of the Intel TFLOPS Supercomputer
Computer simulations needed by the U.S. Department of Energy (DOE) greatly exceed the capacity of the world’s most powerful supercomputers. To satisfy this need, the DOE created the Accelerated Strategic Computing Initiative (ASCI). This program accelerates the development of new scalable supercomputers and will lead to a supercomputer early in the next century that can run at a rate of 100 tri...
متن کاملI/O for TFLOPS Supercomputers
Scalable parallel computers with TFLOPS (Trillion FLoating Point Operations Per Second) performance levels are now under construction. While we believe TFLOPS processor technology is sound, we believe the software and I/O systems surrounding them need improvement. This paper describes our view of a proper system that we built for the nCUBE parallel computer and which is now commercially availab...
متن کاملResource Scaling Effects on MPP Performance: The STAP Benchmark Implications
ÐPresently, massively parallel processors (MPPs) are available only in a few commercial models. A sequence of three ASCI Teraflops MPPs has appeared before the new millenium. This paper evaluates six MPP systems through STAP benchmark experiments. The STAP is a radar signal processing benchmark which exploits regularly structured SPMD data parallelism. We reveal the resource scaling effects on ...
متن کاملPerformance and Scalability Analysis of Teraflop-Scale Parallel Architectures Using Multidimensional Wavefront Applications
We develop a model for the parallel performance of algorithms that consist of concurrent, twodimensional wavefronts implemented in a message passing environment. The model, based on a LogGP machine parametrization, combines the separate contributions of computation and communication wavefronts. We validate the model on three important supercomputer systems, on up to 500 processors. We use data ...
متن کامل