scispace - formally typeset
Search or ask a question

Showing papers in "Future Generation Computer Systems in 1995"



Journal ArticleDOI
TL;DR: In this article, the authors compare update and invalidation-based protocols for their ability to reduce or hide memory access latencies and for their ease of implementation under relaxed memory consistency models.

52 citations


Journal ArticleDOI
TL;DR: The GAMMA model is extended to its probabilistic version — called P-GAMMA model to realise evolutionary computations — namely Probabilistic, classifier, bucket-brigade learning and genetic algorithms.

37 citations


Journal ArticleDOI
Yi Pan1
TL;DR: This paper presents a parallel algorithm to find the smallest (or largest) kth element in a set of N totally ordered (but not sorted) data items and shows that this algorithm can be generalized to process an oversized order statistics problem efficiently.

9 citations


Journal ArticleDOI
TL;DR: The implementation on a Connection Machine (CM2) of a Monte Carlo method for rarefied gas flow simulation is presented; the physical model and the algorithm are described and the performances obtained are comparable to those obtained on a CRAY XMP.

3 citations


Journal ArticleDOI
TL;DR: A message passing multiprocessor model for computation based on functional languages has been suggested that follows the applicative (eager) order of reduction, giving it an edge in exploiting parallelism over those following normal order.

2 citations


Journal ArticleDOI
TL;DR: Using a large subset of the BECAUSE benchmarks, the performances of the POMPC compiler are compared with results obtained with native CM-2 compilers: C∗, PARIS and slicewise versions of CM Fortran.

2 citations


Journal ArticleDOI
TL;DR: The benchmarking by BERTIN (F) of the SUPERNODE SN1000 parallel architecture from PARSYS within the framework of the BECAUSE Project is presented and principles of parallelisation which were used are presented.

1 citations


Journal ArticleDOI
TL;DR: It is shown that the basic floating point performance of the iPSC/2 is no longer competitive with current workstations such as the IBM RS6000/320 and on the other hand, a single node of i PSC/860 is found to give comparable performance to such workst stations.

1 citations


Journal ArticleDOI
TL;DR: This paper describes in this paper a strategy for solving the compressible Navier-Stokes system using unstructured meshes on parallel distributed memory machines.

1 citations


Journal ArticleDOI
TL;DR: The FGCS work on legal reasoning is reviewed with the aim of assessing its contribution to the field of automated legal reasoning, and the wider aims of the FGCS project, which may be summarised as the development of new technologies intended to open up important new classes of applications.

Journal ArticleDOI
TL;DR: Preliminary results for unstructured CFD computations on the KSR-1 parallel processor feature a virtual shared memory environment and high performance results are attainable only when using a message passing paradigm.

Journal ArticleDOI
TL;DR: A first attempt to parallelise the monospecies, inviscid compressible option of a CFD application software package called CALIFE is presented and initial obtained results using the Domain Decomposition Method without overlapping are presented.

Journal ArticleDOI
TL;DR: A model for the parallel execution of subset-equational languages, a class of functional languages that allow constrained nondeterminism, is described, which is suitable for implementation on both conventional and unconventional machines.

Journal ArticleDOI
TL;DR: Methods that could be used to implement a parallel preconditioned CGS solver on the iPSC/860 are reviewed and some performance estimates made using the results of certain BBS tests on this machine.