Results 1  10
of
10
Behaviour Specification of Parallel Active Objects
 Parallel Computing
, 1998
"... The development of parallel programs is primarily concerned with application speed. This has led to the development of parallel applications in which software engineering aspects play only subordinate roles. In order to increase software quality in parallel applications, we motivate the construction ..."
Abstract

Cited by 6 (3 self)
 Add to MetaCart
The development of parallel programs is primarily concerned with application speed. This has led to the development of parallel applications in which software engineering aspects play only subordinate roles. In order to increase software quality in parallel applications, we motivate the construction of parallel programs by composing active objects which interact by means of an objectoriented coordination model. This paper presents a formalism for specifying the behaviour of parallel active objects and a corresponding notion of behavioural types which can be used for verifying whether certain active objects conform to a specified behaviour. Our approach is based on highlevel Petri nets which enable (besides other benefits) automated analysis, in particular for automated type checking of active objects. We illustrate the usefulness of our approach by presenting reusable active objects for a manager/worker architecture. Their correct interaction is shown by automated checking of behavioural types.
Global optimization of transmitter placement in wireless communication systems
 Proc. High Performance Computing Symposium 2002
, 2002
"... parallel ray tracing, pattern search, transmitter placement. This paper explores the application of a global optimization technique to solve the optimal transmitter placement problem in wireless system design. An efficient pattern search algorithm—DIRECT (DIviding RECTangles) of Jones, Perttunen, an ..."
Abstract

Cited by 6 (4 self)
 Add to MetaCart
parallel ray tracing, pattern search, transmitter placement. This paper explores the application of a global optimization technique to solve the optimal transmitter placement problem in wireless system design. An efficient pattern search algorithm—DIRECT (DIviding RECTangles) of Jones, Perttunen, and Stuckman (1993)—has been connected to a parallel 3D radio propagation ray tracing modeler running on a 200node Beowulf cluster of workstations. The algorithm optimizes, for a given computational investment, the locations of a specified number of transmitters across the feasible region of the design space. The focus of the paper is on the implementations of the DIRECT algorithm and the parallel 3D ray tracing propagation model. Both simulation results and site measurement data are presented in support of the effectiveness of the present work. 1.
Coordination Patterns for Parallel Computing
 Coordination Languages and Models, number 1282 in Lecture Notes in Computer Science
, 1997
"... The aim of this paper is to promote the idea of developing reusable coordination patterns for parallel computing, i.e. customizable components from which parallel applications can be built by software composition. To illustrate the idea, a fundamental manager/worker coordination pattern useful for p ..."
Abstract

Cited by 5 (3 self)
 Add to MetaCart
The aim of this paper is to promote the idea of developing reusable coordination patterns for parallel computing, i.e. customizable components from which parallel applications can be built by software composition. To illustrate the idea, a fundamental manager/worker coordination pattern useful for programming a variety of parallel applications is presented.
MPIPOV: a Parallel Implementation of POVRay Based on MPI
 in: Proc. Euro PVM/MPI’99, Lecture Notes in Computer Science
, 1999
"... . The work presents an MPI parallel implementation of PovRay, a powerful public domain ray tracing engine. The major problem in ray tracing is the large amount of CPU time needed for the elaboration of the image. With this parallel version it is possible to reduce the computation time or to rend ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
. The work presents an MPI parallel implementation of PovRay, a powerful public domain ray tracing engine. The major problem in ray tracing is the large amount of CPU time needed for the elaboration of the image. With this parallel version it is possible to reduce the computation time or to render, with the same elaboration time, more complex or detailed images. The program was tested successfully on ParMa2, a lowcost cluster of personal computers running Linux operating system. The results are compared with those obtained with a commercial multiprocessor machine, a Silicon Graphics Onyx2 parallel processing system based on an Origin CCNUMA architecture. 1 Introduction The purpose of this work is the implementation of a distributed version of the original code of PovRay [1], that is a well known public domain program for ray tracing. The parallelization of this algorithm involves many problems that are typical of the parallel computation. The ray tracing process is very c...
Parallel Incremental Raytracing of Animations on a Network of Workstations
 In Procs. Intl. Conf. on Parallel and Distributed Processing Techniques and Applications (PDPTA`98
, 1998
"... To reduce the computation times required for rendering animations, a new incremental raytracing method that computes only the changed parts of images in an animation sequence is proposed. This method is integrated into a parallel version of the POVRay raytracing package implemented on a network of ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
To reduce the computation times required for rendering animations, a new incremental raytracing method that computes only the changed parts of images in an animation sequence is proposed. This method is integrated into a parallel version of the POVRay raytracing package implemented on a network of workstations using the MPI message passing interface. The parallelization relies on a manager/worker scheme which incorporates dynamic task assignment to achieve load balancing. The results of several experiments indicate that the incremental raytracing method yields a reduction of computation time roughly proportional to the number of changed pixels in a particular animated scene. Almost linear speedups are obtained for the parallel versions running on up to 18 workstations. Keywords: Incremental raytracing, parallelism, network of workstations, MPI 1 Introduction Raytracing [1] is a technique used for producing complex graphics images involving reflections, shadows, transparent objects ...
Data and Computation Modeling for Scientific Problem Solving Environments
, 2002
"... This thesis investigates several issues in data and computation modeling for scientific problem solving environments (PSEs). A PSE is viewed as a software system that provides (i) a library of simulation components, (ii) experiment management, (ii) reasoning about simulations and data, and (iv) prob ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
This thesis investigates several issues in data and computation modeling for scientific problem solving environments (PSEs). A PSE is viewed as a software system that provides (i) a library of simulation components, (ii) experiment management, (ii) reasoning about simulations and data, and (iv) problem solving abstractions. Three specific ideas, in functionalities (ii)(iv), form the contributions of this thesis. These include the EMDAG system for experiment management, the BSML markup language for data interchange, and the use of data mining for conducting nontrivial parameter studies. This work emphasizes data modeling and management, two important aspects that have been largely neglected in modern PSE research. All studies are performed in the context of S 4 W, a sophisticated PSE for wireless system design.
S4W: Globally Optimized Design of Wireless Communication Systems
 In Proceedings of the Next Generation Software Workshop, 16th International Parallel and Distributed Processing Symposium (IPDPS’02
, 2002
"... In this paper, a global optimization technique is applied to solve the optimal transmitter placement problem for indoor wireless systems. An e#cient pattern search algorithmDIRECT (DIviding RECTangles) of Jones, Perttunen, and Stuckman (1993)has been connected to a parallel 3D radio propagatio ..."
Abstract
 Add to MetaCart
In this paper, a global optimization technique is applied to solve the optimal transmitter placement problem for indoor wireless systems. An e#cient pattern search algorithmDIRECT (DIviding RECTangles) of Jones, Perttunen, and Stuckman (1993)has been connected to a parallel 3D radio propagation ray tracing modeler running on a 200node Beowulf cluster of Linux workstations. Surrogate functions for a parallel WCDMA (wideband code division multiple access) simulator were used to estimate the system performance for the global optimization algorithm. Power coverage and BER (bit error rate) are considered as two di#erent criteria for optimizing locations of a specified number of transmitters across the feasible region of the design space. This paper briefly describes the underlying radio propagation and WCDMA simulations and focuses on the design issues of the optimization loop.
Tuning of Algorithms for Independent Task Placement in the . . .
 EUROGRAPHICS SYMPOSIUM ON PARALLEL GRAPHICS AND VISUALIZATION (2004)
, 2004
"... This paper investigates assignment strategies (load balancing algorithms) for process farms which solve the problem of online placement of a constant number of independent tasks with given, but unknown, time complexities onto a homogeneous network of processors with a given latency. Results for the ..."
Abstract
 Add to MetaCart
This paper investigates assignment strategies (load balancing algorithms) for process farms which solve the problem of online placement of a constant number of independent tasks with given, but unknown, time complexities onto a homogeneous network of processors with a given latency. Results for the chunking and factoring assignment strategies are summarised for a probabilistic model which models tasks' time complexities as realisations of a random variable with known mean and variance. Then a deterministic model is presented which requires the knowledge of the minimal and maximal tasks' complexities. While the goal in the probabilistic model is the minimisation of the expected makespan, the goal in the deterministic model is the minimisation of the worstcase makespan. We give a novel analysis of chunking and factoring for the deterministic model. In the context of demanddriven parallel ray tracing, tasks' time complexities are unfortunately unknown until the actual computation finishes. Therefore we propose automatic selftuning procedures which estimate the missing information in runtime. We experimentally demonstrate for an "everyday ray tracing setting" that chunking does not perform much worse than factoring on up to 128 processors, if the parameters of these strategies are properly tuned. This may seem surprising. However, the experimentally measured efficiencies agree with our theoretical predictions.
ObjectOriented Parallel Programming with Objective Linda
"... In this paper we present Objective Linda, a coordination model in which objectorientation is combined with uncoupled, generative communication in order to enable objectoriented parallel programming in networked computing resources. Objective Linda provides suitable abstractions for structuring lar ..."
Abstract
 Add to MetaCart
In this paper we present Objective Linda, a coordination model in which objectorientation is combined with uncoupled, generative communication in order to enable objectoriented parallel programming in networked computing resources. Objective Linda provides suitable abstractions for structuring large software systems, supports interoperability between different programming languages and parallel architectures and simplifies the development of parallel applications. Its use is illustrated by presenting programming examples, a prototype implementation is described, and measurements for evaluating the implementation efficiency and the performance of parallel applications are presented.
RadioWave Propagation Prediction using RayTracing Techniques on a Network of Workstations (NOW) ∗
, 2004
"... Raytracing based radio wave propagation prediction models play an important role in the design of contemporary wireless networks as they may now take into account diverse physical phenomena including reflections, diffractions, and diffuse scattering. However, such models are computationally expensi ..."
Abstract
 Add to MetaCart
Raytracing based radio wave propagation prediction models play an important role in the design of contemporary wireless networks as they may now take into account diverse physical phenomena including reflections, diffractions, and diffuse scattering. However, such models are computationally expensive even for moderately complex geographic environments. In this paper, we propose a computational framework that functions on a network of workstations (NOW) and helps speed up the lengthy prediction process. In raytracing based radio propagation prediction models, orders of diffractions are usually processed in a stagebystage fashion. In addition, various source points (transmitters, diffraction corners, or diffuse scattering points) and different raypaths require different processing times. To address these widely varying needs, we propose a combination of the phaseparallel and manager/workers paradigms as the underpinning framework. The phaseparallel component is used to coordinate different computation stages, while the manager/workers paradigm is used to balance workloads among nodes within each stage. The original computation is partitioned into multiple small tasks based on either raypathlevel or sourcepointlevel granularity. Dynamic loadbalancing scheduling schemes are employed to allocate the resulting tasks to the workers.