The PVM system, a software framework for heterogeneous concurrent computing in networked environments, has evolved in the past several years into a viable technology for distributed and parallel processing in a variety of disciplines. PVM supports a straightforward but functionally complete message passing model, and is capable of harnessing the combined resources of typically heterogeneous networked computing platforms to deliver high levels of performance and functionality. In this paper, we describe the architecture of PVM system, and discuss its computing model, the programming interface it supports, auxiliary facilities for process groups and MPP support, and some of the internal implementation techniques employed. Performance issues, dealing primarily with communication overheads, are analyzed, and recent findings as well as experimental enhancements to are presented. In order to demonstrate the viability of PVM for large scale scientific supercomputing, the paper incl...

/ j i t- " o) i' N9_-36932

The NAS parallel benchmarks are a set of applications that embody the key characteristics of typical processing in computational aerodynamics. Five of these, the kernel benchmarks, have been implemented on the PVM system, a software system for network-based concurrent computing, with a view to determining the efficacy of networked environments for high-performance computational aerodynamics applications. We present results of porting and executing the NPB kernels in three different cluster environments using low- to medium-powered workstations on Ethernet and two types of FDDI networks. Our results indicate that mediocre to good performance could be obtained despite the communications intensive nature of the applications. In most cases, we were able to achieve performance levels within an order of magnitude of a Cray Y/MP-1 on 8-workstation clusters via optimizations to the PVM infrastructure alone, i.e. with little or no algorithmic modifications. However, our results also indicate th...

In this paper, six portable parallel programming environments are compared. For each environment, communication bandwidths are reported for simple 2 node and 4 node benchmarks. Reproducibility was a top priority, so these tests were run on an isolated ethernet network of identical SPARCstation 1 workstations. Earlier reports of this work omitted opinions reached during the benchmarking about the effectiveness of these environments. These opinions are included in this paper since they are based on a unique experience; the experience of running two programs on six different environments. 1 Introduction Parallel computing -- ranging from clusters of workstations to tightly coupled distributed memory machines -- is the dominant form of supercomputing. Software development for these computers, however, is difficult and seriously lags behind developments in parallel hardware. To address this situation, numerous groups have designed programming models that span all MIMD multiprocessor system...

This report describes the coordination language CoLa which is especially aimed at providing a universal platform for the development of applications in the field of distributed artificial intelligence in a massively parallel environment. CoLa is intended to be an add on to existing programming languages, enriching the host programming model with advanced coordination features to master massive parallelism. The features provided by CoLa include a high level identification abstraction for purpose of communication of processes and an extended semantics on messages. The CoLa programming model allows to define complex communication topologies logically, therefore facilitating the development of portable and scalable applications. Keywords: Coordination language, massively parallel systems, distributed artificial intelligence, synchronization, communication. Contents 1 Introduction 2 2 Fundamental Principles 2 2.1 Locality : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : ...