一个小型网格原型系统的设计与实现

这个小型网格原型系统是基于OGSA(开放网格服务架构)体系结构和Web 服务架构而构建的网格服务的基础平台,按照OGSI网格服务规范标准,用GT3(Globus Toolkit 3.0)网格开发工具包开发实现。程序设计符合通常的分布式计算编程模型,它包括服务器端编程和客户端编程两个部分,它们之间的联系是通过WSDL服务描述文件来完成的。最后通过实验展示了其实用性。     

LVT: a layered verification technique for distributed computing systems

This paper presents a layered verification technique, called LVT, for the verification of distributed computing systems with multiple component layers. Each lower layer in such a system provides services in support of functionality of the higher layer. By taking a very general view of programming languages as interfaces of systems, LVT treats each layer in a distributed computing system as a distributed programming language. Each relatively higher‐level language in the computing system is implemented in terms of a lower‐level language. The verification of each layer in a distributed computing system can then be viewed as the verification of implementation correctness for a distributed language. This paper also presents the application of LVT to the verification of a distributed computing system, which has three layers: a small high‐level distributed programming language; a multiple processor architecture consisting of an instruction set and system calls for inter‐process message passing; and a network interface. Programs in the high‐level language are implemented by a compiler mapping from the language layer to the multiprocessor layer. System calls are implemented by network services. LVT and its application demonstrate that the correct execution of a distributed program, most notably its inter‐process communication, is verifiable through layers. The verified layers guarantee the correctness of (1) the compiled code that makes reference to operating system calls, (2) the operating system calls in terms of network calls, and (3) the network calls in terms of network transmission steps. The specification and verification involved are carried out by using the Cambridge Higher Order Logic (HOL) theorem proving system. Copyright © 1999 John Wiley & Sons, Ltd.     

HLA技术在虚拟样机开发中的应用

虚拟样机系统目前还没有一个统一、规范、标准的体系平台,导致系统在后续扩展时较困难,而且在开发系统时开发人员亦不能将全部精力放在专业领域的仿真程序设计上.而目前广泛应用于军事领域的分布式仿真系统的HLA技术,具有清晰的逻辑层次、良好的扩展性、高效的数据传输速率等诸多优点.正基于此,该文首次将HLA技术框架引入机械系统虚拟样机系统设计中,主要讨论了符合HLA标准的分布式机械仿真系统的系统结构和仿真节点的设计方法,从而解决了机械系统虚拟样机扩展性差等问题.     

Improving Resource Management in Distributed Systems using Language-Level Structuring Concepts

Currently, a shift of paradigm from sequential to distributed computing can be observed. Tremendous efforts are needed to cope with the challenging demands that are inherent to this transition. The two most important issues concern programming support for developing large-scale distributed applications and the efficient execution of such applications on top of a distributed hardware configuration. This paper presents a new language and object-based approach called MoDiS to cope with these demands. Distributed systems are developed and transformed into executables following a systematic, top-down driven method. Language-level is intended to mean that a high-level programming language is used to develop operating system services as well as user-level applications. The language-level concepts employed are based on objects and support advanced structuring features. Structural dependencies between objects are implicitly determined at the application level and exploited by the distributed resource management system to transform applications into efficient executables.     

Middleware infrastructure for parallel and distributed programming models in heterogeneous systems

We introduce a middleware infrastructure that provides software services for developing and deploying high-performance parallel programming models and distributed applications on clusters and networked heterogeneous systems. This middleware infrastructure utilizes distributed agents residing on the participating machines and communicating with one another to perform the required functions. An intensive study of the parallel programming models in Java has helped identify the common requirements for a runtime support environment, which we used to define the middleware functionality. A Java-based prototype, based on this architecture, has been developed along with a Java object-passing interface (JOPI) class library. Since this system is written completely in Java, it is portable and allows executing programs in parallel across multiple heterogeneous platforms. With the middleware infrastructure, users need not deal with the mechanisms of deploying and loading user classes on the heterogeneous system. Moreover, details of scheduling, controlling, monitoring, and executing user jobs are hidden, while the management of system resources is made transparent to the user. Such uniform services are essential for facilitating the development and deployment of scalable high-performance Java applications on clusters and heterogeneous systems. An initial deployment of a parallel Java programming model over a heterogeneous, distributed system shows good performance results. In addition, a framework for the agents' startup mechanism and organization is introduced to provide scalable deployment and communication among the agents.     

Super－Object模型：在分布存储器多计算机上实现共享存储器程序设计模式

Ｓｕｐｅｒ－Ｏｂｊｅｃｔ模型提出了一种新的方法，在分布存储器多计算机上实现语言级虚拟共享存储器以支持共享存储器通信模式．Ｓｕｐｅｒ－Ｏｂｊｅｃｔ模型引入新的概念ｓｕｐｅｒ－ｏｂｊｅｃｔ，不同于其它模型，基于ｓｕｐｅｒ－ｏｂｊｅｃｔ，它提出了新的共享数据定位方法，全局地址标识（ｎａｍｅ，ｏｆｆ－ｓｅｔ）．Ｓｕｐｅｒ－Ｏｂｊｅｃｔ模型与Ｆｏｒｔｒａｎ７７结合，我们实现了一个运行时间系统和库调用，支持程序员使用Ｆｏｒｔｒａｎ语言编写并行程序，最后介绍了系统的实现和取得的性能．     

基于情境演算理论的情境服务研究

本文针对一个与移动服务相关的概念即基于情境的移动服务（简称情境服务）展开研究。首先对情境服务及其相关研究进行了说明，然后提出了一个基于情境演算理论和高级逻辑程序语言Golog的情境服务平台。该平台能够方便地嵌入各种情境服务，并且能够方便地将用户的偏好约束集成到逻辑服务程序中。本文给出了一个情境服务原型系统示
例，并在最后提出了几个今后值得继续研究的方向。     

A message-driven programming system for fine-grain multicomputers

This paper describes an experimental message-driven programming system for fine-grain multicomputers. The initial target architecture is the J-machine designed at MIT. This machine combines a unique collection of architectural features that include fine-grain processes, on-chip associative memory; and hardware support for process synchronization. The programming system uses these mechanisms via a simple message-driven process model that blurs the distinction between processes and messages: messages correspond to processes that are executed elsewhere in the network. This model allows code and data to be distributed across the computers in the machine, and is supported at every stage of the program development cycle. The prototype system we have developed includes a basic set of programming tools to support the model; these include a compiler, linker, archiver, loader and microkernel. Although the concepts are language independent, our prototype system is based on GNU-C.     

Parallel Pattern-Based Systems for Computational Biology: A Case Study

Computational biology research is now faced with the burgeoning number of genome data. The rigorous postprocessing of this data requires an increased role for high-performance computing (HPC). Because the development of HPC applications for computational biology problems is much more complex than the corresponding sequential applications, existing traditional programming techniques have demonstrated their inadequacy. Many high level programming techniques, such as skeleton and pattern-based programming, have therefore been designed to provide users new ways to get HPC applications without much effort. However, most of them remain absent from the mainstream practice for computational biology. In this paper, we present a new parallel pattern-based system prototype for computational biology. The underlying programming techniques are based on generic programming, a programming technique suited for the generic representation of abstract concepts. This allows the system to be built in a generic way at application level and, thus, provides good extensibility and flexibility. We show how this system can be used to develop HPC applications for popular computational biology algorithms and lead to significant runtime savings on distributed memory architectures.     

A Java-centric Distributed Object-based Paradigm for Surveillance Services and Visual Message Exchange

Building an effective surveillance system on the Internet is an interesting application involving the interoperability of heterogeneous computing environments and the challenge of orchestrating the exchange of different visual message types in a unified manner. In this paper, we propose a distributed object-based paradigm to achieve this by applying the so-called bridging technologies for supporting distributed surveillance services. This paradigm integrates emerging technologies of the next-generation Internet-based applications ranging from wireless transmission at the network infrastructure level, to multi-tier, client/server, and RMI/CORBA at the architecture level, and to Java-centric, delegation-based visual user interfaces, and object web-based user interfaces at the system design level. The experimental results show that objects in the paradigm can efficiently cross the boundaries of different platforms, operating systems, programming languages, and visual message types. The DSS paradigm establishes a firm basis for supporting a variety of potentially cost-effective applications in a plug-and-play manner.     

Parallel computing in networks of workstations with Paralex

Modern distributed systems consisting of powerful workstations and high-speed interconnection networks are an economical alternative to special-purpose supercomputers. The technical issues that need to be addressed in exploiting the parallelism inherent in a distributed system include heterogeneity, high-latency communication, fault tolerance and dynamic load balancing. Current software systems for parallel programming provide little or no automatic support towards these issues and require users to be experts in fault-tolerant distributed computing. The Paralex system is aimed at exploring the extent to which the parallel application programmer can be liberated from the complexities of distributed systems. Paralex is a complete programming environment and makes extensive use of graphics to define, edit, execute, and debug parallel scientific applications. All of the necessary code for distributing the computation across a network and replicating it to achieve fault tolerance and dynamic load balancing is automatically generated by the system. In this paper we give an overview of Paralex and present our experiences with a prototype implementation     

MVSS: an active storage architecture

This paper presents MVSS, a storage system for active storage devices MVSS offers a single framework for supporting various services at the device level. It provides a flexible interface for associating services to a file through multiple views of the file Similar to views of a database in a multiview database system, views in MVSS are generated dynamically and are not stored on physical storage devices. MVSS represents each view of an underlying file through a separate entry in the file system namespace. MVSS separates the deployment of services from file system implementations and, thus, allows services to be migrated to storage devices. The paper presents the design of MVSS and how different services can be supported in MVSS at the device level. To illustrate our approach, we implemented a prototype system on PCs running Linux. We present results from example applications implemented on the prototype and discuss a variety of architectural issues including mixed workloads.     

MMM: a Web-based system for sharing statistical computing modules

This prototype method management system (MMM), implemented with Web technologies, supports distributed authoring and execution of computational software modules among an interdisciplinary group of developers and users. Heterogeneous data formats, programming languages, and computing platforms pose various challenges whenever researchers are sharing and combining software modules, especially when the collaboration occurs across different traditions of scientific computing. To meet these challenges, we designed and implemented MMM. MMM is a collection of middleware services to support the interaction between software users and developers, and to facilitate the sharing of software modules across heterogeneous networks. The system design follows the World Wide Web paradigm: developers (providers) install their modules on the network in a way that allows users (consumers) to access and execute them. A prototype is available on the Web     

基于UDDI的知识服务

采用UDDI协议标准来解决动态知识的发布和集成问题。UDDI不仅在结构上适应松耦合的知识服务系统的要求,而是在技术上也适应知识服务网络的分布性及异构性特点。作为应用,该文给出了一个供应链知识服务的原型系统。     

REED: Flexible rule based programming of wireless sensor networks at runtime

Wireless Sensor Networks (WSNs) have emerged as an enabling technology for a variety of distributed applications. WSN middleware eases the development of these applications by providing a uniform programming environment. In this paper we present a rule based approach called REED (Rule Execution and Event Distribution) and describe how it supports flexible programming of WSNs at runtime. Indeed REED is required by the nature of its project setting to allow runtime programming. We demonstrate that by combining this runtime programmability with rules in an event, condition, action format we can support a range of paradigms, including Publish-subscribe and data aggregation algorithms. Current WSN middleware solutions have limited on-line programmability support so the applications cannot re-configure their WSNs while operational. Yet the runtime nature of the prototype requires both the distribution of rules and the events that trigger them so we also describe the rule management approach used to support the rule distribution; in particular a novel rule merging and filtering algorithm is described. The paper reports on the results gained from a REED prototype system constructed in our laboratory using Gumstix.     

ES-Kit: An object-oriented distributed system

This paper describes the design, implementation, and performance of ES-Kit, a distributed object-oriented system being developed by the Experimental Systems Project at the Microelectronics and Computer Technology Corporation. The operating system consists of a kernel and a set of Public Service Objects which dynamically extend the functionality of the kernel by providing several traditional operating system services when required by application objects. Applications for the ES-Kit environment are written in GNU C++ and do not require additional language primitives for distributed execution. Initial performance results from a representative set of applications indicate that the object-oriented paradigm provides a powerful solution to distributed programming.     

Key Concepts of the INCAS Multicomputer Project

This paper gives an overview of the INCAS (INCremental Architecture for distributed Systems) multicomputer project, which aims at the development of a comprehensive methodology for the design and implementation of locally distributed systems. A structuring concept for distributed operating systems has been developed and integrated into the system implementation language LADY. The concurrent high-level programming language CSSA, based on the actor model, has been designed for the implementation of distributed applications. A substantial effort in the INCAS project is directed towards the development of a distributed test methodology. An experimental system has been implemented on a network of ten MC68000 microcomputers. Preliminary experience with the methodology has been gained from a small number of prototype applications.     

A Mechanism Supporting the Client/Server Relationship in the Operating System of Distributed System “THUDS”

This paper presents a distributed operating system modeled as an abstract machine that provides all the distributed processes with the same set of services.The kernel of our operating system supports services which are achieved by a remote procedure call on requests by parallel processes.Therefore,a scheme for solving the client-server relationship is required.In our system there are more than one clients and,at least,a receive would be required for each.Similarly,there are more than one servers such that the send in a client should produce a message that can be received by every server.Consequently,a mechanism well suited for programming multiple-clients/single-server and single-client/multiple-servers interactions is proposed.     

可重构片上系统设计流程中的动态重构问题研究

近年来,可重构片上系统已成为科学研究及嵌入式应用领域中应对复杂计算需求的有效技术解决方案.针对目前缺少一个从系统级设计到应用实现,统一、综合规划动态重构问题的系统设计流程,以及动态重构过程对系统设计人员不透明等问题,在系统设计层给出了一种过程级软硬件统一编程模型.在此框架内,设计人员通过调用已根据应用特性进行优化的软硬件协同函数,即可利用高级语言完成系统功能描述;在细节设计层提出了基于单位面积加速比的软硬件任务调度算法,实时管理动态可重构资源;在应用实现层,以可重构专用图形加速卡为原型系统,论述动态可重构系统实现中的关键技术.实验及测试结果验证了通过将动态重构问题置于整个系统设计流程中予以考虑,能够达到提升系统开发效率之目的.     

Distrbution and Abstract Types in Emerald

Emerald is an object-based language for programming distributed subsystems and applications. Its novel features include 1) a single object model that is used both for programming in the small and in the large, 2) support for abstract types, and 3) an explicit notion of object location and mobility. This paper outlines the goals of Em-erald, relates Emerald to previous work, and describes its type system and distribution support. We are currently constructing a prototype implementation of Emerald.