Operating Systems virtualize the execution environment of a task by managing the available resources of the system. On traditional architectures a task executes on a static environment, i.e. every instant of a task will be carried upon a fixed set of ...

The HIGHER RADIX HYPERCUBE (HRH) is investigated in this paper as an interconnection network for multiprocessors as well as a virtual network. The HRH is based on a radix higher than two, the radix for the hypercube. We have compared several topological ...

Code generation for existing parallel computers involves partitioning the work into schedulable units, assigning the resulting computational units to specific processors, and generating the code for interprocessor communication and synchronization. The ...

Two sieve algorithms, a pipeline and a parallel version are presented for a distributed-memory multiprocessor. Traditionally, the sieve algorithm is solved using a pipeline approach. This algorithm has a high communication to computation ratio and as a ...

This paper presents the results of macrotasking and microtasking a Navier-Stokes algorithm on the Cray-2. The algorithm is a compact difference scheme for the solution of the incompressible, two-dimensional, time dependent Navier-Stokes equations. Two ...

Tightly-coupled parallel computer systems are available from several vendors. The mechanism for process scheduling employed by virtually all of these systems is preemptive by job priority. This scheme for process scheduling consumes significant ...

Two dimensional interconnection schemes have some inherent advantages because of their linear area and constant wire-lengths. The nearest-neighbor mesh is such a topology that has enjoyed a widespread acceptance. We investigate a family of bus-based ...

A parallel pipelined data flow coprocessor has been developed for the 68000 based Commodore Amiga workstation. The coprocessor, based on Nippon Electric Corporation's μPD7281 Image Pipelined Processor (ImPP), was designed as an algorithm processor for ...

A non-symmetric algorithm is presented that achieves mutual exclusion in a decentralized computer network by exchanging messages and a synchronization queue. The algorithm is unique in the sense that the number of messages required to achieve mutual ...

Process Originated Migration in a distributed system comprised of both Sun and MicroVax workstations is described. An introductory section characterizes distributed systems, distinguishes them from network systems, and summarizes existing work in ...

"Teams of Robots" (TOR) is a promising research topic of computer science that is currently attracting attention from research computer scientists. A TOR explores how to organize two or more robots into one team, to enable them to cooperate with each ...

Many critical embedded systems which have very high reliability requirements operate in real time. Considering the significance of their applications, design of highly reliable software is a very important research area. In the development of reliable ...

The neural network (or 'brain-style') approach to computing is useful for developing systems to perform tasks humans traditionally do well and that computers do not. Neural network systems are particularly suited for pattern matching and categorization ...

A new class of "concurrent automata" is constructed. These automata are extensions of pushdown automata and Petri nets. They lie properly between ordinary Petri nets and Turing machines. In the September 1977 issue of Computing survey James Peterson ...

We consider the notions of deadlocks, traps, and liveness in Petri nets from a graph theoretic viewpoint. We characterize deadlocks and traps for a special class of Petri nets, namely complete Petri nets, and then determine necessary and sufficient ...

MOTIVATION. Parallel processing units will operate faster if they have private memories rather than a common memory, because common memories have conflict problems. There is a problem of making multiple individual private memories functionally conform ...

The purpose of this program is to visualize the operation of a token ring. The software has two parts; the first part explains the topology of the token ring network (frame format, multiple bridges and gate), and the second part depicts data ...