Abstract
The efficient execution of irregular data parallel applications, on dynamically shared computing clusters, requires novel approaches to manage the runtime load distribution. Stich environments have an unpredictable dynamic behaviour both due to the application requirements and to the available system's resources. This uncertainty was the main motivation to propose and evaluate an application level scheduler, where decisions are efficiently taken with improved accurate predictions on the environment's current and near future state, based on available incomplete and aged measured data. Bayesian decision networks are used as the scheduler's decision making mechanism: its effectiveness to manage the load distribution of a parallel ray tracer is assessed and compared with alternative strategies. The evaluation results, with complex scenes on a 7 shared nodes cluster with dynamically variable workloads, show considerable performance improvements over blind strategies, and stress the benefits over a sensor based deterministic approach of identical complexity.

Abstract
The efficient execution of irregular parallel applications on shared distributed systems requires novel approaches to scheduling, since both the application requirements and the system resources exhibit an unpredictable behavior. This paper proposes Bayesian decision networks as the paradigm to handle the uncertainty a scheduler has about the environment's current and future states. Experiments performed with a parallel ray tracer show promising performance improvements over a deterministic approach of identical complexity. These improvements grow as the level of system sharing and the application's workload irregularity increase, suggesting that the effectiveness of decision network based schedulers grows with the complexity of the environment being managed.

Abstract
This article presents an evaluation study of point-to-point and collective communication performance on a parallel processing system, a 16 node Parsytec PowerXplorer, using three different communication environments: PARIX, PVM and MPI.

Abstract

Abstract
Physically based global illumination rendering at interactive frame rates would enable users to navigate within complex virtual environments, such as archaeological models. These algorithms, however, are computationally too demanding to allow interactive navigation on current PCs. A technique based on image subsampling and spatiotemporal coherence among successive frames is exploited, while resorting to progressive refinement whenever there is available computing power. A physically based ray tracer (Radiance) is used to compute reflected radiance at the model's triangles vertices. Progressive refinement is achieved increasing the sampling frequency by subdividing certain triangles and requesting shading information for the resulting vertices. This paper proposes and evaluates different criteria for selecting which triangles to subdivide. A random criterium and two criteria based on Normalized Luminance Differences are evaluated: one operating on image space, the other on object space. Results, obtained with a model of an old roman town, show that the object space criterium is able to locate and represent visual discontinuities, such as shadows, and does so requiring less triangle subdivisions than the other two.

Abstract
The computational requirements of full global illumination rendering are such that it is still not possible to achieve high-fidelity graphics of very complex scenes in a reasonable time on a single computer. By identifying which computations are more relevant to the desired quality of the solution, selective rendering can significantly reduce rendering times. In this paper we present a novel component-based selective rendering system in which the quality of every image, and indeed every pixel, can be controlled by means of a component regular expression (crex). The crex provides a flexible mechanism for controlling which components are rendered and in which order. It can be used as a strategy for directing the light transport within a scene and also in a progressive rendering framework. Furthermore, the crex can be combined with visual perception techniques to reduce rendering computation times without compromising the perceived visual quality. By means of a psychophysical experiment we demonstrate how the crex can be successfully used in such a perceptual rendering framework. In addition, we show how the crex's flexibility enables it to be incorporated in a predictive framework for time-constrained rendering. Copyright © 2005 by the Association for Computing Machinery, Inc.