PVM与网络并行计算

松散耦合的异构型并行处理系统近年来又有较大的发展，这种只能支持粗粒度并行的计算环境，由于通讯硬件的更新和软件环境的开发，其性能已经能够与某些ＭＰＰ机相媲美，因而形成并行处理领域一个强有力的分支－网络计算，本文围绕ＰＶＭ，讨论网络计算的软件环境，概述其特色和发展，并与其它基于消息传递的软件环境相比较，最后预测ＰＶＭ未来可能面临的问题和发展方向。     

PVM网络并行计算实现方案

使用双网卡方案实现了基于PVM平台的网络并行计算环境，并在该环境下测试了并行整体光照算法。对实际的测试结果进行了分析，讨论了利用PVM进行网络并行计算的机制和优化方法。     

LBS—基于PVM的动态任务负载平衡系统

负载平衡问题是影响工作站机群系统并行计算性能的一个重要因素。     

基于代理的网格计算中间件

WADE系统是基于代理技术实现的一个可屏蔽异构和分布性的动态自适应的校园计算网格，提出了基于代理技术在校园网络内实现并行计算的方法，详细论述了基于代理的网格计算中间件的体系结构和主要模块功能，阐述了利用代理实现异构编译、协同计算的过程，给出了代理的Java实现方法，利用软件代理实现网格计算中间件，可以解决异构计算平台下多种并行编程环境的协同计算问题，为用户提供统一的服务接口，这将大大增强系统的可用性。     

Algorithm-system scalability of heterogeneous computing

Scalability is a key factor of the design of distributed systems and parallel algorithms and machines. However, conventional scalabilities are designed for homogeneous parallel processing. There is no suitable and commonly accepted definition of scalability metric for heterogeneous systems. Isospeed scalability is a well-defined metric for homogeneous computing. This study extends the isospeed scalability metric to general heterogeneous computing systems. The proposed isospeed-efficiency model is suitable for both homogeneous and heterogeneous computing. Through theoretical analyses, we derive methodologies of scalability measurement and prediction for heterogeneous systems. Experimental results have verified the analytical results and confirmed that the proposed isospeed-efficiency scalability works well in both homogeneous and heterogeneous environments.     

异构型计算与并行程序设计环境

异构型计算能有效地利用多种不同的高性能计算机，以满足大型计算问题的不同计算需求。本文着重讨论了支持异构型计算的并行程序设计环境所需具备的功能，并以目前最流行的ＰＶＭ和Ｅｘｐｒｅｓｓ系统为例进行说明。     

一种包含Transputer的异构型分布式并行计算系统的架构

本文针对ＰＶＭ不支持Ｔｒａｎｓｐｕｔｅｒ的不足，介绍了基于ＴＣＰ／ＩＰ的Ｔｒａｎｓｐｕｔｅｒ异构型分布式并行计算系统（Ｔ－ＤＰＣＳ）的软硬件架构。从通信协议的选择、Ｔｒａｎｓｐｕｔｅｒ共享支持软件、分布式协同软件架构和通信函数库的实现等４个方面详细地阐述了实现ＴＤＰＣＳ的方案，并进行了原型实现。     

Performance of the NAS Parallel Benchmarks on PVM-Based Networks

The NAS parallel benchmarks are a set of applications that embody the key characteristics of typical processing in computational aerodynamics. Five of these, the kernel benchmarks, have been implemented on the PVM system, a software system for network-based concurrent computing, with a view to determining the efficacy of networked environments for high-performance computational aerodynamics applications. We present results of porting and executing the NPB kernels in three different duster environments using low- to medium-powered workstations on Ethernet and two types of FDDI networks. Our results indicate that mediocre to good performance could be obtained despite the communications-intensive nature of the applications. In most cases, we were able to achieve performance levels within an order of magnitude of a Cray Y/MP-1 on eight-workstation clusters via optimizations to the PVM infrastructure alone, i.e., with little or no algorithmic modifications. However, our results also indicate that further improvements are possible and that network-based computing has the potential to be a viable technology for high-performance scientific computing.     

基于高性能微机群集的可扩展性的研究与设计

可扩展性是并行计算系统的重要性能指标,虽然异构系统越来越普遍,但对其可扩展性的研究还很少。给出了一种既适合同构并行计算系统又适合异构并行计算系统的效率的定义,根据访定义对可扩展性进行了分析,得出了既适用于同构系统又适用于异构系统的等效率模型,并根据开销比得出了在某一效率常数保持一致的情况下系统规模和工作负载的变化情况。最后通过实验进行了分析,结果表明该模型可以对效率和可扩展性进行较好的评测,并能预测并行计算系统的高可扩展性。     

细胞自动机语言Cellang的扩展与应用

并行执行与并行描述是并行计算的两个方面,后者是并行软件技术的一个日益重要的问题,目前主流的并行语言都是针对特定的需求设计的。细胞自动机对许多问题具有自然而贴切的描述并行性,作为动力学系统仿真工具与并行离散计算模型近来受到广泛关注。通过对某些细胞自动机语言进行扩展,可以使其成为并行数值计算的良好工具,方便地描述与求解一大类并行数值计算问题。本文介绍我们对细胞自动机语言Cellang的扩展及在并行数值计算上的应用。     

基于集群的海洋遥感图像融合并行计算策略

集群体系下的大规模并行计算,是高性能计算的基础。遥感图像处理效率的提高,有赖于并行计算技术的应用。在分析已有网格计算环境下分布式任务分配方法的基础上,针对海上遥感图像目标物数量相对较少的特点,首先利用四叉树结构理念对目标区域进行划分,同时采用动态负载均衡的任务分配策略与并行计算思想,提出对目标区域图像进行融合处理的集群体系任务分配算法处理模型。通过对比验证,表明该集群体系下算法模型能有效地提高图像融合的速度。     

Application of Parallel Virtual Machine Framework to the Strong Prime Problem

This paper use the well-discussed PVM (Parallel Virtual Machine) software with several personal computers, and adopt the widespread Microsoft Windows '98 operating system as our operation platform to construct a heterogeneous PCs cluster. By engaging the related researches of PC cluster system and cluster computing theory, we apply our heterogeneous PC cluster computing system to generate more secure parameters for some public key cryptosystems such as RSA. Copes with each parameter's related mathematic theory's restriction, enormous computation power is needed to get better computation performance in generating these parameters. In this paper, we contribute heterogeneous PCs combined with the PVM software to cryptosystem parameters, which is conformed to today's safety specification and requirement. We practically generate these data to prove that computer cluster can effectively accumulate enormous computation power, and then demonstrate the cluster computation application in finding strong primes which are needed in some public key cryptosystems.     

微机网络环境下提高PVM并行程序性能的策略

网络并行计算是并行计算与分布式计算技术非常重要的发展方向之一,结合具体的数值试验,探讨了Windows操作系统下基于PVM的网络并行数值计算中影响PVM并行程序性能的几个重要因素,包括负载平衡、通信开销、网络性能、任务粒度、处理机个数、精度要求及处理机内存容量问题等,并提出了提高PVM并行程序性能的相应策略,以高效快速地实现问题的求解.     

基于PVM平台的并行数字图象处理算法研究

ＰＶＭ是目前最有影响的基于消息传递的并行软件,它为用户提供了一种以较小的代价实现高性能计算机的有效途径。本文提出了一种基于ＰＶＭ平台的数字图象处理算法的平行化方法,该算法充分考虑了数字图象处理的特点,使用“群集”模型,有效提高了数字图象处理的速度,达到理想效果。     

基于小波变换的图像压缩并行结构

本文结合小波图像压缩算法和VLSI并行处理的要求，提出了一种心动阵列与通用处理器组合的并行处理结构，具有易于VLSI实现、支持多种编码方案的优点，并在并行软件环境PVM上模拟其工作过程，证明该设计是可行的．     

基于GPU实时视频处理的多投影融合系统研究

介绍了GPU高速并行运算及其对数字图像、视频处理的重要作用。针对多通道环幕投影系统,采用CPU与GPU组合的异构计算结构,提出了一种视频实时处理方案。该方案通过DirectShow的链路模型保证了视频处理的灵活性,设计并采用可用于并行运算的几何校正、边缘融合算法,提升了视频处理的高效性。这一构架可以用于单通道4k格式视频的高质量效果展示,同时能有效降低构建成本,提高系统的经济实用性。     

Parallel multisource view maintenance

In a distributed environment, materialized views are used to integrate data from different information sources and then store them in some centralized location. In order to maintain such materialized views, maintenance queries need to be sent to information sources by the data warehouse management system. Due to the independence of the information sources and the data warehouse, concurrency issues are raised between the maintenance queries and the local update transactions at each information source. Recent solutions such as ECA and Strobe tackle such concurrent maintenance, however with the requirement of quiescence of the information sources. SWEEP and POSSE overcome this limitation by decomposing the global maintenance query into smaller subqueries to be sent to every information source and then performing conflict correction locally at the data warehouse. Note that all these previous approaches handle the data updates one at a time. Hence either some of the information sources or the data warehouse is likely to be idle during most of the maintenance process. In this paper, we propose that a set of updates should be maintained in parallel by several concurrent maintenance processes so that both the information sources as well as the warehouse would be utilized more fully throughout the maintenance process. This parallelism should then improve the overall maintenance performance. For this we have developed a parallel view maintenance algorithm, called PVM, that substantially improves upon the performance of previous maintenance approaches by handling a set of data updates at the same time. The parallel handling of a set of updates is orthogonal to the particular maintenance algorithm applied to the handling of each individual update. In order to perform parallel view maintenance, we have identified two critical issues that must be overcome: (1) detecting maintenance-concurrent data updates in a parallel mode and (2) correcting the problem that the data warehouse commit order may not correspond to the data warehouse update processing order due to parallel maintenance handling. In this work, we provide solutions to both issues. For the former, we insert a middle-layer timestamp assignment module for detecting maintenance-concurrent data updates without requiring any global clock synchronization. For the latter, we introduce the negative counter concept to solve the problem of variant orders of committing effects of data updates to the data warehouse. We provide a proof of the correctness of PVM that guarantees that our strategy indeed generates the correct final data warehouse state. We have implemented both SWEEP and PVM in our EVE data warehousing system. Our performance study demonstrates that a manyfold performance improvement is achieved by PVM over SWEEP.Received: 12 November 2001, Accepted: 18 December 2002, Published online: 31 July 2003This work was supported in part by the NSF NYI grant IIS-979624 and NSF CISE Instrumentation grant IRIS 97-29878 and NSF grant IIS-9988776.     

Parallel techniques for information extraction from hyperspectral imagery using heterogeneous networks of workstations

Recent advances in space and computer technologies are revolutionizing the way remotely sensed data is collected, managed and interpreted. In particular, NASA is continuously gathering very high-dimensional imagery data from the surface of the Earth with hyperspectral sensors such as the Jet Propulsion Laboratory's airborne visible-infrared imaging spectrometer (AVIRIS) or the Hyperion imager aboard Earth Observing-1 (EO-1) satellite platform. The development of efficient techniques for extracting scientific understanding from the massive amount of collected data is critical for space-based Earth science and planetary exploration. In particular, many hyperspectral imaging applications demand real time or near real-time performance. Examples include homeland security/defense, environmental modeling and assessment, wild-land fire tracking, biological threat detection, and monitoring of oil spills and other types of chemical contamination. Only a few parallel processing strategies for hyperspectral imagery are currently available, and most of them assume homogeneity in the underlying computing platform. In turn, heterogeneous networks of workstations (NOWs) have rapidly become a very promising computing solution which is expected to play a major role in the design of high-performance systems for many on-going and planned remote sensing missions. In order to address the need for cost-effective parallel solutions in this fast growing and emerging research area, this paper develops several highly innovative parallel algorithms for unsupervised information extraction and mining from hyperspectral image data sets, which have been specifically designed to be run in heterogeneous NOWs. The considered approaches fall into three highly representative categories: clustering, classification and spectral mixture analysis. Analytical and experimental results are presented in the context of realistic applications (based on hyperspectral data sets from the AVIRIS data repository) using several homogeneous and heterogeneous parallel computing facilities available at NASA's Goddard Space Flight Center and the University of Maryland.