Parallel Buchberger Algorithms on Virtual Shared Memory KSR1
نویسنده
چکیده
We develop parallel versions of Buchbergers Gröbner Basis algorithm for a virtual shared memory KSR1 computer. A coarse grain version does S-polynomial reduction concurrently and respects the same critical pair selection strategy as the sequential algorithm. A fine grain version parallelizes polynomial reduction in a pipeline and can be combined with the parallel S-polynomial reduction. The algorithms are designed for a virtual shared memory architecture and a dynamic memory management with concurrent garbage collection implemented in the MAS computer algebra system. We discuss the achieved speedup figures for up to 24 processors on some standard examples.
منابع مشابه
Performance of Parallel Branch and Bound Algorithms on the KSR1 Multiprocessor
In this paper we consider the parallelization of the branch and bound (BB) algorithm with best-rst search strategy on the KSR1 shared-memory mul-tiprocessor. Two shared-memory parallel BB algorithms are implemented on a 56-processor system. Measurements indicate that the scalability of the two algorithms is limited by the cost of interprocessor communications and by the cost of synchronization....
متن کاملExperiences in Parallelising an Aeronautics Code on the KSR1
Virtual Shared Memory (VSM) has been proposed as the solution to scalable shared memory parallel architectures. This paper reports on parallelising a scientific code from aeronautical engineering to a VSM machine, the KSR1. The code predicts the laminar to turbulent transition point of flow over an aerofoil. The experiences of initial porting and successive optimisation to examine efficiency on...
متن کاملAccelerated Ray Tracing on the KSR 1 Virtual Shared - Memory Parallel Computer
This report describes the results of experiments with accelerated ray tracing on a virtual shared-memory parallel computer. The machine used, the Kendall Square Research KSR1, is described in sufficient detail for its novel features to be appreciated. The report describes a new ray tracer written specifically to capitalise on the KSR1’s features. Models used in the experiments include well-know...
متن کاملIrregular Loop Patterns Compilation on Distributed Shared Memory Multiprocessors
This paper addresses irregular loops compilation on Distributed Memory Parallel Computers (DMPCs) that provide a Shared Virtual Memory. Runtime techniques are introduced to distribute irregular loops so that page movements are reduced. Experimental results for the KSR1 are presented.
متن کاملA Comparison of Shared Virtual Memory and Message Passing Programming Techniques Based on a Finite Element Application
This paper describes the methods used and experiences made with implementing a nite element application on three diierent parallel computers with either message passing or shared virtual memory as the programming model. Designing a parallel nite element application using message-passing requires to nd a data domain decomposition to map data into the local memory of the processors. Since data ac...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1994