The MEMORY CHANNEL network is a dedicated cluster interconnect that provides virtual shared memory among nodes by means of internodal address space mapping. The interconnect implements direct user-level messaging and guarantees strict message ordering under all conditions, including transmission errors. These characteristics allow industry-standard communication interfaces and parallel programming paradigms to achieve much higher efficiency than on conventional networks. This paper presents an overview of the MEMORY CHANNEL network architecture and describes DIGITAL‘s crossbarbased implementation of the second-generation MEMORY CHANNEL network, MEMORY CHANNEL 2. This network provides bisection bandwidths of 1,000 to 2,000 megabytes per second and a sustained process-to-process bandwidth of 88 megabytes per second. One-way, processto-process message latency is less than 2.2 microseconds.

this paper, we present an overview ofPSE version 1.0 and explain why it was designed and selected for use with HPF programs. We then discuss cluster definition and management, describe the PSE application model, and discussPSE's message-passing communication options, including an optimized transport for message passing. We conclude with our performance results.

New features of Fortran are changing the way in which scientists are writing and maintaining large analytic codes. Further, a number of these new features make it easier for compilers to generate highly optimized architecture-specific codes. Among the most exciting kinds of architecture-specific optimizations are those having to do with parallelism. This paper describes Fortran 90 and the standardized language extensions for both shared-memory and distributed-memory parallelism. In particular, three case studies are examined, showing how the distributed-memory extensions (High Performance Fortran) are used both for data parallel algorithms and for single-program– multiple-data algorithms.

Abstract not found