The ALPS Kernel for Processor Networks
نویسندگان
چکیده
ALPS is an operating system kernel which provides support for developing portable and e cient parallel programs for MIMD machines. It achieves this by supporting a new paradigm for parallel programming, wherein the shared or distributed nature of memory and the topology of the processor interconnection are transparent to the programmer. Using the ALPS approach, a programmer of a processor network can write a portable parallel program which does not depend on the interconnection topology, and which can still be executed e ciently. This paper presents the ALPS kernel primitives and their implementation on a Transputer network and a Hypercube, and demonstrates how topology independence and e cient scheduling can be achieved in a single framework. Key phrases: Operating Systems, Parallel Processing, Processor Networks, Topology Independence, Object-Oriented Programming. The ALPS project is supported by the National Science Foundation under grant CCR-8810388. e er e r r cess r et r s anas andal, ahendra amachandran and rasad ishnu hotla epartment of omputer and Information cience he hio tate niversit , olum us, hio 1 -1 e a : a s s. -s a e.e
منابع مشابه
Communication Kernel for High Speed Networks in the Parallel Environment LANDA-HSN
Due to the development of several new technologies concerning high speed networks, one can get PC clusters for intensive computing at a low price. Parallel programming environments must take into account the heterogeneity of workstation networks into their communication kernels. New technologies must be easily integrated into such environments. The LANDA-HSN project has been designed for an eff...
متن کاملAn approach to Improve Particle Swarm Optimization Algorithm Using CUDA
The time consumption in solving computationally heavy problems has always been a concern for computer programmers. Due to simplicity of its implementation, the PSO (Particle Swarm Optimization) is a suitable meta-heuristic algorithm for solving computationally heavy problems. However, despite the simplicity, the algorithm is inefficient for solving real computationally heavy problems but the pr...
متن کاملAn introduction to kernel-based learning algorithms
This paper provides an introduction to support vector machines, kernel Fisher discriminant analysis, and kernel principal component analysis, as examples for successful kernel-based learning methods. We first give a short background about Vapnik-Chervonenkis theory and kernel feature spaces and then proceed to kernel based learning in supervised and unsupervised scenarios including practical an...
متن کاملA Self-Configuring Distributed Kernel for Satellite Networks
The Space Software Laboratory is developing a self-configuring distributed kernel to be used on future satellite missions. The completion of this system will allow a network of heterogeneous processor nodes to communicate and broadcast in a scalable, self-configuring manner. Node applications software will be transparent to the underlying network architecture, message routing, and number of net...
متن کاملBenchmarking: Performance on ALPS and Formosa Clusters
This paper presents the benchmarking results and performance evaluation of differentclustersbuilt atthe National Center for High-Performance Computingin Taiwan. Performance of processor, memory subsystem andinterconnect is a critical factor in the overall performance of high performance computing platforms. The evaluation compares different system architecture and software platforms. Most super...
متن کامل