Showing papers in "Future Generation Computer Systems in 1994"

TL;DR: This work combines the idea of Composition-Tree with ideas developed earlier, to develop an abstract execution model that supports full Prolog semantics while at the same time avoiding redundant inferences when computing solutions to (purely) independent and-parallel goals.

TL;DR: Experimental results for two Navier-Stokes test problems show that the theoretically predicted dependency of the combined numerical and parallel efficiencies of multigrid methods on the number of processors employed is in fact very weak, leading to the conclusion thatMultigrid is an appropriate candidate for solving partial differential equations on massively parallel machines.

TL;DR: A new logic programming framework involving Linda-like primitives is proposed first and focuses on the key concepts and control operators, which subsumes existing concrete proposals and provides a general framework well-suited for their semantic analysis.

TL;DR: The advantages and disadvantages produced by the use of virtual channels are analyzed, proposing a new approach to maximize their efficiency, which uses time-dependent selection functions, associating a threshold to some virtual channels.

TL;DR: Initial findings show that it is possible to obtain good performance without the need for low level parallel programming.

TL;DR: This work describes how it can be integrated in a given OOL in a clean and elegant fashion, using only already existing concepts — i.e. without modifying the OOL syntax and semantics — to exploit the potential power of massively parallel systems.

TL;DR: A new interconnection network, called the open intelligent network, is proposed and added toSYMPATI2 to form SYMPATIX, which allows asynchronous transfers of data between the different processing elements of the new system and allows the efficient interconnection of specific modules.

TL;DR: This paper overviews Paragon, and introduces a simple classification scheme for analysis of Paragon specifications, and a translation strategy for static systems in terms of a digital hardware description language.

TL;DR: By using trace-driven simulation, several cache architectures of tightly coupled multiprocessor mainframes are evaluated and the influence of the coherence protocol on the cache performance is analysed and relative differences of 20 to 30% in the miss ratios are shown.

TL;DR: This special issue, a selection of the conference papers highlights some important aspects on software tools for parallel computing as well as on real applications on parallel systems.

TL;DR: The results of the experiments related to the efficiency of the communication of a T800 transputer based system are presented to determine the basic hardware parameters for the communication capabilities of the system.

TL;DR: The implementation of the Because Benchmark Set programs on the Connection Machine CM-2 is presented and discussed, and the numerical method used (spectral method) is explained, in order to understand the algorithmic solutions that are proposed.

TL;DR: A way of treating axisymmetric problems of electromagnetic scattering based upon a property of circulant matrices is presented and an implementation on a parallel computer which is quite efficient is shown.

TL;DR: The objectives of the BECAUSE Project are recalled in order to replace the BBS in the context of this ESPRIT Project and the applications from which test programs have been extracted are presented and all tests are described in detail.

TL;DR: The architecture of the parallel management kernel for the parallel inference engine PIE64 is described, focusing on how to treat load distribution and scheduling in highly parallel symbolic processing.