A comprehensive algorithmic analysis of finite state dependent queueing models and exponentially distributed workstations is formulated and presented. The material handling system is modeled with finite state dependent queueing network M/G/c/c models and the individual workstations are modeled with exponentially distributed single and multi-server M/M/c queueing models. The coupling of these qu...

Shared memory, one of the most popular models for programming parallel platforms, is becoming ubiquitous both in low-end workstations and high-end servers. With the advent of low-latency networking hardware, clusters of workstations strive to offer the same processing power as high-end servers for a fraction of the cost. In such environments, shared memory has been limited to page-based systems...

This paper demonstrates that blocking due to nite buuering capacity of the system workstations can be a destabilizing factor for distributed scheduling policies applied in multi-class queueing networks, in addition to the currently identiied sources of instability. The issue is important since, in practice, all systems modeled by such networks operate under buuering capacity constraints. * The ...

In this paper, we present a task scheduling heuristic, based on a parallel genetic algorithm (PGA). The algorithm schedules parallel programs, represented as directed acyclic graphs (DAGs), onto multi-processor systems with dynamic interconnection networks (DIN). It combines PGA with list scheduling approach and takes into account inter-processor communication cost, link contention in the DIN a...

A multi-dimensional finite element method (GWADAPT) which utilizes local mesh adaptation is used to calculate groundwater flow and subsurface transport of contaminant in saturated and unsaturated conditions. The model uses simple modifications to increase computational speed and reduce storage, allowing the model to run on PCs and workstations. GWADAPT can be accessed from the web and downloade...

For the HERA-B experiment at HERA multi-processor systems (farms) for the last levels of the trigger and data acquisition system are planned. Based on implementation studies a concept for a prototype system was derived which will be set up in 1997. It will contain already the main components of the nal system. The hardware decision for the processor nodes will be mainly innuenced by the price/p...

poster presentations Posters will be manned by their authors. Thursday 31 march 13:45 – 14:45

This paper describes the management of heterogeneity in Stardust, an environment for parallel programming above networks of heterogeneous machines, which can include distributed memory multi-computers and networks of workstations. Applications using Stardust can communicate both through message passing and distributed shared memory. Stardust is currently implemented on an heterogeneous system i...

Clustering is a widely used architecture to implement scalable, highly-available systems. The ORPG clustering technology is based on a loosely-coupled multi-tasking software architecture. It allows the ORPG system to run on multiple nodes (UNIX hosts, e.g. workstations and PCs) and thus, makes ORPG scalable. When the processing resource requirement increases beyond the current RPG hardware limi...