The wavelet transform has proven to be a valuable tool for image processing applications, like image compression and noise reduction. In this paper we present a scheme to process very large images that do not fit in t...
详细信息
ISBN:
(纸本)3540656413
The wavelet transform has proven to be a valuable tool for image processing applications, like image compression and noise reduction. In this paper we present a scheme to process very large images that do not fit in the memory of a single computer, based on the software library WAILI (Wavelets with Integer Lifting). Such images are divided into blocks that are processed quasi independently, allowing efficient parallel programming. The blocking is almost completely transparent to the user.
parallel programming of sparse data structure has been recognized to be hard to debug and to require an excessive amount of development time. The PARADEIS library (parallel Data Environment for Irregular Structure) is...
详细信息
ISBN:
(纸本)3540656413
parallel programming of sparse data structure has been recognized to be hard to debug and to require an excessive amount of development time. The PARADEIS library (parallel Data Environment for Irregular Structure) is an object library that aims at providing a data-parallel approach with object oriented languages to ease the treatment of the sparse data structure for SPMD-like programs on Massively parallel Processors (MPPs) with distributed memory. This paper focuses on the core of communication layout: the conversion of global addresses to communications, and its relationship with partitioning strategies.
One of the reasons why parallel programming is considered to be a difficult task is that users frequently, cannot predict the performance impact of implementation? decisions prior to program execution. This results in...
详细信息
ISBN:
(纸本)0769500595
One of the reasons why parallel programming is considered to be a difficult task is that users frequently, cannot predict the performance impact of implementation? decisions prior to program execution. This results in a cycle of incremental performance improvements based on run-time performance data. While gathering and analyzing performance data is supported by a large number of tools, typically interactive, the task of performance analysis is still too complex for. users. This article illustrates this fact based on the current analysis support on GRAY T3E. As a consequence, we are convinced that automatic analysis tools are required to identify frequently occuring and well-defined performance problems automatically! This article describes the novel design of a generic automatic performance analysis environment called KOJAK. Besides its structure we also outline the first component, EARL, a new meta-tool designed and implemented as a programmable interface to calculate more abstract metrics fr om existing trace files, and to locate complex patterns describing performance problems.
In this paper we propose to enhance concurrency of computation and communication by means of a coordination model we develop on the base of orthogonal parallel programming abstractions called forcing expressions. It i...
详细信息
ISBN:
(纸本)3540663630
In this paper we propose to enhance concurrency of computation and communication by means of a coordination model we develop on the base of orthogonal parallel programming abstractions called forcing expressions. It is shown that besides efficient synchronization forcing expressions can facilitate formal development of dedicated schemes of efficient data exchanges in multilevel memory environment yielding systematic reduction of slow memory accesses. The coordination programming facilities are illustrated by a case study to enhance data parallel paradigm of computation.
Comparable to the dining philosophers in, parallel programming we look for a paradigmatic problem for realtime systems. Our experimental system for balancing a ball on a flat board is described and assessed under its ...
详细信息
ISBN:
(纸本)0769501346
Comparable to the dining philosophers in, parallel programming we look for a paradigmatic problem for realtime systems. Our experimental system for balancing a ball on a flat board is described and assessed under its animative, curricular, and scientific aspects. Particularly under the curricular aspect the experimental system offers transferable solutions for various subjects in real-time systems education e.g. modeling, analysis, scheduling, software engineering, techniques of programming, operating systems and computer networks.
parallel programming can be made easier by means of a skeleton based methodology, such as P-3 L, which helps programmers to compose their applications by using a set of fixed parallel patterns. Such kind of approach i...
详细信息
ISBN:
(纸本)3540658211
parallel programming can be made easier by means of a skeleton based methodology, such as P-3 L, which helps programmers to compose their applications by using a set of fixed parallel patterns. Such kind of approach is also useful to obtain portability because the "structured" nature of the language can be used to devise a composable support for each parallel pattern so that the complexity of finding an "optimal" implementation on different parallel architectures can be reduced. In this work, we show how we can conjugate the BSP abstract model and its related cost analysis to provide an implementation strategy "abstract enough" for being also machine independent. We hope this can be a first step towards the idea of a portable set of optimization rules. The first results show how an implementation template for the Map constructor, able of to be tuned automatically, can be designed. A validation of the technique is given for a Gray T3E and a cluster of PC-Linux.
The OCM is a universally usable, distributed on-line monitoring system currently implemented for the PVM programming library. Due to the growing use of MPI in parallel programming, there is a need to enhance the OCM i...
详细信息
ISBN:
(纸本)3540658211
The OCM is a universally usable, distributed on-line monitoring system currently implemented for the PVM programming library. Due to the growing use of MPI in parallel programming, there is a need to enhance the OCM in order to support MPI applications development. This paper presents approaches to solve the problems of the mpich-oriented start-up mechanism and profiling MPI library calls in the OCM. While the existing transport layer for the OCM internal communication is preserved, a number of new MPI-related services are added.
作者：
Frumkin, MYan, JNASA
Ames Res Ctr Numer Aerosp Simulat Syst Div Moffett Field CA 94035 USA
We present an HPF implementation of ARC3D code along with the profiling and performance data on SGI Origin 2000. Advantages and limitations of HPF as a parallel programming language for CFD applications are discussed....
详细信息
ISBN:
(纸本)0769500870
We present an HPF implementation of ARC3D code along with the profiling and performance data on SGI Origin 2000. Advantages and limitations of HPF as a parallel programming language for CFD applications are discussed. For achieving good performance results we used the data distributions optimized for implementation of implicit and explicit operators of the solver and boundary conditions. We compare the results with MPI and directive based implementations.
Actual behaviour of parallel programs is of capital importance for the development of an application. Programs will be considered matured applications when their performance is under acceptable limits. Traditional par...
详细信息
ISBN:
(纸本)3540662286
Actual behaviour of parallel programs is of capital importance for the development of an application. Programs will be considered matured applications when their performance is under acceptable limits. Traditional parallel programming forces the programmer to understand the enormous amount of performance information obtained from the execution of a program. In this paper we propose an automatic analysis tool that lets the programmers of applications avoid this difficult task. This automatic performance analysis tool main objective is to find poor designed structures in the application. It considers the: trace file obtained from the execution oft he application in order to locate rite most important behaviour problems of the application. Then, the tool relates them with the corresponding application cone and scans the code looking for any design decision which could be changed to improve the behaviour.
In the field of parallel programming we notice a considerable lack of efficient an-line tools for debugging, performance analysis etc. This is due to the fact that the construction of those tools must be based on a co...
详细信息
ISBN:
(纸本)3540663630
In the field of parallel programming we notice a considerable lack of efficient an-line tools for debugging, performance analysis etc. This is due to the fact that the construction of those tools must be based on a complicated software infrastructure. In the case of such software being available tools from different vendors are almost always incompatible as they use proprietary implementations for it. We will demonstrate in this paper that only a common infrastructure will ease the construction of on-line tools and that it is a necessary precondition for eventually having interoperable tools. Interoperable tools form the basis for synergetic tool environments and yield an added value over just integrated environments.
暂无评论