支持构件迁移的分布式系统容错算法

随着分布式应用技术的发展,构件的静态性限制了分布式系统在运行期间适应其可用资源变化的要求.为了动态适应系统可用资源,提高系统性能,支持构件可迁移的分布式应用研究受到广泛关注.针对提高构件可迁移分布式应用的可靠性,提出一种分布式系统容错算法.在准同步检查点算法的基础上,通过引入shadow-chain和通讯录机制,实现了构件间通信的可靠性和构件重定位的透明性,从而使得系统能够在较低代价下设置检查点,并能将出错后的系统回卷到一致性状态.     

基于CORBA的分布对象容错机制研究与实现

一、引言目前,以CORBA为代表的分布对象计算技术已日趋成熟,越来越多的分布式应用系统利用CORBA提供的标准服务和协议来实现。基于CORBA的新一代的分布式系统,如分布式实时控制系统、在线支付系统和股票交易系统,需要可靠性保证。容错技术是分布式系统运行过程中可靠性保证的重要手段,可以在分布式系统的每一个层次实现,利用底层的CORBA基础设施提供容错机制具有显著的优势,既能够避免系统层为支持容错而做的巨大改变,又能够简化应用软件的设计。因此,容错CORBA已经成为国外CORBA研究的重     

用VB开发应用软件时的错误预测及控制

本文阐述了在用VB开发应用软件时,如何预测和控制某些用户造成的错误。对某些程序功能进行限制,使用数据有效性检查以及良好的错误例程,使程序内部出错机会减到最小,增     

基于行为的构件组合及存在性检查

多个构件的组合可以解决单个构件的行为不能完全满足用户需求的问题。构件组合的关键是检查组合后复合构件的行为是否完全满足用户的需求。针对上述问题,提出一个基于行为的构件组合方法。该方法采用具有终止状态的接口自动机描述构件行为和用户需求。为了检查组合后复合构件的行为是否满足用户需求,提出基于行为映射图的组合存在性检查方法,给出从复合构件中提取用户所需行为的方法。     

基于行为的构件组合及存在性检查

多个构件的组合可以解决单个构件的行为不能完全满足用户需求的问题。构件组合的关键是检查组合后复合构件的行为是否完全满足用户的需求。针对上述问题,提出一个基于行为的构件组合方法。该方法采用具有终止状态的接口自动机描述构件行为和用户需求。为了检查组合后复合构件的行为是否满足用户需求,提出基于行为映射图的组合存在性检查方法,给出从复合构件中提取用户所需行为的方法。     

共用数据导向的分布式系统失效恢复缺陷检测

分布式系统的可靠性和可用性至关重要.然而,不正确的失效恢复机制及其实现会引发失效恢复缺陷,威胁分布式系统的可靠性和可用性.只有发生在特定时机的节点失效才会触发失效恢复缺陷,因此,检测分布式系统中的失效恢复缺陷具有挑战性.提出了一种新方法 Deminer来自动检测分布式系统中的失效恢复缺陷.在大规模分布式系统中观察到,同一份数据(即共用数据)可能被一组I/O写操作存储到不同位置(如不同的存储路径或节点).而打断这样一组共用数据写操作执行的节点失效更容易触发失效恢复缺陷.因此, Deminer以共用数据的使用为指导,通过自动识别和注入这类容易引发故障的节点失效来检测失效恢复缺陷.首先, Deminer追踪目标系统的一次正确执行中关键数据的使用.然后, Deminer基于执行轨迹识别使用共用数据的I/O写操作对,并预测容易引发错误的节点失效注入点.最后, Deminer通过测试预测的节点失效注入点以及检查故障征兆来暴露和确认失效恢复缺陷.实现了Deminer原型工具,并在4个流行的开源分布式系统ZooKeeper、HBase、YARN和HDFS的最新版本上进行了验证.实验结果表明Demine...     

一个分布式系统通信控制软件的设计

文中提出了一种严格分层管理的分布式系统模型，探讨了其体系结构、运行机制及其域管理方式对系统通信控制功能的要求。以基于该模型的分布式系统为原型，设计了一个相对独立、通用于系统中各种节点的通信控制功能软件，初步解决了通信双方的身份认证、通信可靠性和系统中结点运行状态监视等问题，并为上层分布式应用提供了通信控制功能接口。     

N个有限状态机的死锁和死环回避算法

死锁和死环是分布式系统或计算机网络中进程间通信时经常发生的逻辑错误，它们往往造成分布式系统中的部分主机或整个系统瘫痪，本文提出两个回避死锁和死环的算法，这些算法可被用来设计分布式系统的通信协议，从而提高分布式系统的可靠性和减少协议开发成本。     

基于CORBA的分布式系统自适应容错模型的研究

一、引言分布式系统需要可靠性保证,例如在线支付系统对安全性提出了很高的要求。因此,分布式系统必须提供可靠性机制,支持关键业务。容错技术是分布式系统运行过程中可靠性保证的重要手段,冗余资源是实现容错的根本保证。单一的容错策略仅适用于特定的应用和特定的系统,无法适应系统状态的动态变化,支持广泛的分布式应用。系统的容错模型应该能够智能地根据外部运行环境的变化,选择合适的容错策略,以便在保证系统可靠性的前提下提高系统资源利用率。自适应容错可以在分布式系统中的各个层次中实现。基于操作系统的自适应容错强烈地依赖于特定的操作系统,系统可移植性差;而在应用程序中实现自适应容错又加重了开     

基于对象中间件的自适应容错技术

一、引言近年来,计算机系统逐渐渗透到人类生活中的各个方面,对人类生活产生了不可估量的影响。目前,面向对象的分布式系统已经被广泛地应用于关键业系统中,如空中交通管制系统、在线支付系统和核电站控制系统等。显然,提供高可靠的服务是用户对系统最基本的要求,否则将造成重大财产,甚至人员损失。在过去的几十年间,针对分布式系统生命周期的不同阶段,研究人员提出了大量提高系统可靠性的方法。在分布式系统运行过程中,容错技术是保证分布式系统运行可靠性的重要手段。随着硬件可靠性的提高,软件设计错误成为系统的主要错误源,因此软件容错成为决定系统可靠性的极其重要的因素。目前,主要的软件容错模型包括版本复制、RB模型、NVP模型和NSCP模型。然而,由于软件容错实现的复     

Reliability factors in business software: volatility, requirements and end-users

Abstract. Many business‐oriented software applications are subject to frequent changes in requirements. This paper shows that, ceteris paribus, increases in the volatility of system requirements decrease the reliability of software. Further, systems that exhibit high volatility during the development phase are likely to have lower reliability during their operational phase. In addition to the typically higher volatility of requirements, end‐users who specify the requirements of business‐oriented systems are usually less technically oriented than people who specify the requirements of compilers, radar tracking systems or medical equipment. Hence, the characteristics of software reliability problems for business‐oriented systems are likely to differ significantly from those of more technically oriented systems.     

Emerging standards for component software

Component software benefits include reusability and interoperability, among others. What are the similarities and differences between the competing standards for this new technology, and how will they interoperate? Object-oriented technology is steadily gaining acceptance for commercial and custom application development through programming languages such as C++ and Smalltalk, object oriented CASE tools, databases, and operating systems such as Next Computer's NextStep. Two emerging technologies, called compound documents and component software, will likely accelerate the spread of objectoriented concepts across system-level services, development tools, and application-level behaviours. Tied closely to the popular client/server architecture for distributed computing, compound documents and component software define object-based models that facilitate interactions between independent programs. These new approaches promise to simplify the design and implementation of complex software applications and, equally important, simplify human-computer interactive work models for application end users. Following unfortunate tradition, major software vendors have developed competing standards to support and drive compound document and component software technologies. These incompatible standards specify distinct object models, data storage models, and application interaction protocols. The incompatibilities have generated confusion in the market, as independent software vendors, system integrators, in-house developers, and end users struggle to sort out the standards' relative merits, weaknesses, and chances for commercial success. Let's take a look now at the general technical concepts underlying compound documents and component software. Then we examine the OpenDoc, OLE 2, COM, and CORBA standards being proposed for these two technologies. Finally, we'll review the work being done to extend the standards and to achieve interoperability across them     

Visual Interactive Systems for End-User Development: A Model-Based Design Methodology

This paper is about the development of systems whose end users are professional people working in a specific domain (e.g., medicine, geology, mechanical engineering); they are expert in that domain, but not necessarily expert in nor even conversant with computer science. In several work organizations, end users need to tailor their software systems to better adapt them to their requirements and even to create or modify software artifacts. These are end-user development activities and are the focus of this paper. A model of the interaction between users and systems, which also takes into account their reciprocal coevolution during system usage, is discussed. This model is used to define a methodology aimed at designing software environments that allow end users to become designers of their own tools. The methodology is illustrated by discussing two experimental cases.     

基于序列的可靠性分析方法

随着软构件技术的快速发展，基于构件设计复杂软件系统的软件开发方法日趋成熟。如何利用系统架构和软构件的可靠性分析软件系统的可靠性成为一个亟待解决的问题。软件是静态的，而开发过程是动态的。为了在动态的开发过程中跟踪可靠性，本文提出了基于序列的场景模型，以便分析软件可靠性。与相关的其他方法不同的是，该方法更关注于动态开发过程中的可靠性分析。     

工业控制系统可靠性设计与测试评估技术研究

目前我国的工业控制系统发展迅速,但其可信性较低,为解决这一问题,我们开展了工业控制系统的可靠性设计和测试评估技术研究。研究形成了工业控制系统可靠性设计规范和工业控制系统可靠性测试评估规范,另外还生成了工业控制系统可靠性设计和测试评估的软件操作平台。     

Allocation and analysis of reliability: multiple levels: system, subsystem, and module

We model the reliability allocation and prediction process across a hierarchical software system comprised of modules, subsystems, and system. We experiment in modeling complex reliability software systems using several software reliability models to test the feasibility of the process and to evaluate the accuracy of the models for this application. This is a subject deserving research and experimentation because this type of system is implemented in safety-critical projects, such as National Aeronautics and Space Administration (NASA) flight software modules, that we use in our experiments. Given the reliability requirement of a software system in the software planning or design stage, we predict each module’s reliability and their relationships (e.g., reliability interactions among modules, subsystems, and system), Our critical interfaces and components are failure-mode sequences and the modules that comprise these sequences, respectively. In addition, we evaluate how sensitive the achievement of reliability goals is to predicted component reliabilities that do not meet expectations.     

锅炉控制系统采用WinCC冗余的可靠性研究

锅炉做为能量转换的压力容器,其工作过程涉及许多参数,是典型的多入多出(MIMO)系统,控制系统容错能力的高低、可靠性的强弱直接影响锅炉运行过程的安全。冗余容错技术是改善控制系统可靠性的有效途径,以WinCC自带的软件冗余功能实现服务器硬件的热备,避免控制系统的瘫痪和崩溃,高性价比的提升了运行操作的可靠性,在锅炉控制领域有较强的推广和借鉴意义。     

A new fuzzy rule based algorithm for estimating software faults in early phase of development

Estimation of reliability and the number of faults present in software in its early development phase, i.e., requirement analysis or design phase is very beneficial for developing reliable software with optimal cost. Software reliability prediction in early phase of development is highly desirable to the stake holders, software developers, managers and end users. Since, the failure data are unavailable in early phase of software development, different reliability relevant software metrics and similar project data are used to develop models for early software fault prediction. The proposed model uses the linguistic values of software metrics in fuzzy inference system to predict the total number of faults present in software in its requirement analysis phase. Considering specific target reliability, weightage of each input software metrics and size of software, an algorithm has been proposed here for developing general fuzzy rule base. For model validation of the proposed model, 20 real software project data have been used here. The linguistic values from four software metrics related to requirement analysis phase have been considered as model inputs. The performance of the proposed model has been compared with two existing early software fault prediction models.     

Special Feature A Survey of Methods of Achieving Reliable Software

As organizations become more dependent on computers, they become more sensitive to computer system failures. The importance of computer reliability in realtime control systems (e.g., communications systems^8,15and traffic control systems) has been recognized for some time. Many computer users are now becoming aware that they accomplish more on systems which seldom crash because of malfunctions than on systems which run very rapidly (and correctly) between frequent crashes. Consequently, increasing emphasis is being placed by users and vendors on the reliability of the total system, and particularly the system software. A notable example of this is the emphasis placed by IBM on reliability concerns in the design and implementation of OS/VS2 Release 2.^35,36     

一种针对可信计算平台的分布式可信验证机制

可信计算技术在提高系统安全性的同时,也给用户使用计算机带来了诸多限制。为了解决自由软件等未授权程序的可信验证问题,提出了一种基于可信计算的分布式可信验证机制（DTVMTC）。该机制以庞大的Internet用户群为基础,通过网络数据统计的方法,实现对应用程序的可信验证,从而解决了对无可信来源但实际可信的应用程序进行可信验证的问题,保障了用户使用可信计算平台的自由。在Windows平台实现了DTVMTC的原型,实验结果表明DTVMTC能够实现预定目标且具有良好的性能。