基于服务更新的网构软件可靠性模型研究(英文)

网构软件是在开放、动态和多变的Internet环境下软件系统基本形态的一种抽象。这种新的软件系统,它的构建依赖于开放环境中各异构、自治的软件服务实体之间的有效协同。其可靠性不单取决于拥有自主性的软件服务实体本身,还取决于外部环境的动态变化,主要表现为开放环境下服务实体元素的更新,所以传统的软件可靠性的度量模型已不能适应这种新的软件形态。能否在网构软件形态下建立一个好的可靠性度量模型成为其中一个较为核心的问题。文章以服务更新过程中失效数(failurecounts)为基础,将服务更新强度引入Musa-Okumoto(M-O)模型中,作为M-O模型在新的软件形态下的一个推广。最后讨论了网构软件退化的情形和退化条件,为开放环境下网构软件可靠性研究提供一种思路。     

面向基于场景规约的Web服务消息流分析与验证

采用UML顺序图构成基于场景的规约、WS-BPEL作为Web服务的描述语言,提出了一种面向基于场景规约对Web服务消息流进行分析与验证的方法:首先,对WS-BPEL消息流进行分析并将其自动抽象为基于Petri网的模型;同时,为了缩小状态空间、提高验证效率,在不影响消息交互顺序的前提下,对WS-BPEL源码和基于Petri网的模型分别进行化简,即面向基于场景规约将与验证无关的活动和元素删除;最后,通过遍历基于Petri网的模型以验证WS-BPEL消息流与基于场景的规约之间的一致性(消息交互顺序的存在/强制一致性).文中通过一个贯穿整个分析与验证过程的实例加以说明.该方法已经实现成为一个原型工具.     

一种面向动态软件体系结构的在线演化方法

为适应计算环境和用户需求在系统运行期间的变化,满足软件系统进行动态演化的需求,提出面向动态软件体系结构的在线演化方法,设计并实现了一种运行时刻的软件体系结构元模型,将原先运行时刻不可见的体系结构设计信息具体化为显式的体系结构实体,并与系统实现及系统规约之间保持因果关联.元模型的演化可通过反射实现对运行系统的修改和对规约的更新,所有演化行为都在良定义的体系结构元模型的指导下规范地进行,保证了演化前后系统的一致性、完整性和演化的可追溯性.基于该方法开发了可视化支撑平台Artemis-ARC系统,支持构件和服务的设计、开发、集成及动态演化,并通过简单的应用实例展示了在线演化的效果.     

基于JBoss AOP的软件动态演化

大型的长期运行的系统往往要求软件系统不能停止运行,而软件演化作为软件维护的重要部分是每一个软件系统不可或缺的,这就要求能够实现软件的动态演化.提出了一种基于AOP技术的软件动态演化机制,这种机制分为两种策略:运行时软件的功能更新方法和代码更新方法.根据对软件进行更新依赖的静态分析,按照一定的规则将一个个更新的动作作为一个系统的横切方面动态编织到系统中去,从而实现软件的动态演化.     

ConUp:一个支持构件动态更新的SCA中间件系统

中间件已经成为网络环境下构建复杂应用系统的核心基础支撑软件。Internet的发展促使应用环境从封闭、静态转变为开放、动态,这就要求中间件上的应用具有动态更新的能力。业界广泛应用的中间件多支持构件的热部署,但不能自动保证系统的一致性。ConUp是一个基于Tuscany的SCA中间件系统,它通过对构件间动态依赖的管理来保证构件动态更新后系统的一致性。本原型演示将展示ConUp的中间件上的构件进行动态更新的过程,它对多种动态更新算法、策略的支持,及其在动态更新安全性、及时性和低干扰性方面的优势。     

云环境下元数据弹性分级一致性保障机制研究 *

云存储环境下,元数据提供数据定位与资源描述服务,云存储系统利用副本技术保障元数据的可用性。为了解决元数据的动态更新导致的元数据副本的不一致,本文从云存储系统中不同元数据对一致性强度需求的差异性出发,提出了基于元数据弹性分级的一致性保障机制。根据元数据访问频率以及更新属性敏感度对元数据更新弹性分级,结合基于读优化的Quorum算法实现元数据更新的分级一致性保障。实验证明该机制可以有效权衡元数据副本可用性、一致性和系统开销,相比单一的一致性策略更为灵活。     

分布式系统动态配置技术的研究与实现

动态配置技术用于系统运行时刻改变系统配置,从而满足分布式系统对在线演化的需求。如何保证系统一致性,并尽可能提高动态配置性能,是动态配置研究需要解决的关键问题。本文提出了行为一致性,并将系统一致性总结为行为一致性、构件状态一致性、应用状态一致性和引用一致性。通过扩展静止状态理论,本文提出了采用等待方式和阻塞方式驱动构件进入静止状态的算法,保证了行为一致性。通过分析动态配置意图,本文提出了保证系统一致性的动态配置算法,确定了采用等待方式和阻塞方式的动态配置场景,从而在严格保证行为一致性的前提下,尽可能地提高了动态配置性能。最后,本文基于 CCM 构件平台实现了动态配置平台。性能测试的结果表明,该平台可满足应用对性能的要求。     

基于构件的软件演化波及效应分析

随着新技术的采用和系统环境的变化,构件和软件系统的演化在所难免,演化会影响系统的整体行为。分析了基于构件的软件系统中的耦合性对演化波及效应的影响,对软件系统中构件内部各种依赖关系、构件与连接件之间的各种耦合关系进行矩阵表示,提出一种能够防止波及效应扩大化的软件系统动态演化的波及效应方法,实现了对软件系统的构件和连接件的演化波及效应研究。在基于构件的软件系统的动态演化中,可以根据波及效应的分析获得需要进行重新修改或演化的构件和连接件,从而保证动态演化的一致性和连续性。     

分布式动态更新支持系统：研究综述

一些服务于关键应用领域、提供持续服务的分布式系统软件需要不断地进化，以修正软件故障、扩展服务功能、提高系统性能，从而产生动态更新的需求，即保证不中断系统服务的同时实现对系统功能的更新。由于分布式系统本身的复杂性，使得其上的动态更新比较复杂，一般需要做的工作包括：在适当的时候自动产生和传递更新消息；透明地实施更新、验证更新的合法性、一致性并处理非法更新下的容错问题；支持多版本节点之间的交互；对更新期间滞留的消息进行处理。本文围绕支持动态更新分布式系统的相关工作，从一般需求、功能目标、需要解决的关键问题、现有的工作成果以及将来的工作五个方面做了较为综合的阐述。     

资源约束下的基于类依赖关系的微服务识别方法

为有效提升基于微服务架构的遗留软件系统重构的自动化水平,根据存在依赖关系的两个类所操作的资源数据之间存在着一定相关性的原则,提出了一种资源约束下基于类依赖关系的微服务识别方法。首先,根据遗留软件程序中的类依赖关系构建类依赖关系图,并设置每个类的资源实体标签;然后,设计了基于资源实体标签的类依赖关系图划分算法,用以划分原软件系统和得到候选微服务;最后,合并依赖程度较高的候选微服务,从而得到最终的微服务集合。基于GitHub的4个开源项目的实验结果表明,所提方法具有高于90%的微服务划分准确率,证实了同时考虑不同类之间的依赖关系和资源约束对于微服务识别是合理和有效的。     

Maintaining Global Integrity Constraints in Distributed Databases

Given some integrity constraints over a distributed database, we consider the problem of incrementally checking global consistency in response to updates made to the base relations but without accessing all these base relations. In many application areas such as collaborative design, mobile computing and enterprise information systems, total data availability cannot be assumed. Even if all the base data is available, some of it may incur such a high cost that its use should only be considered as a last resort. Without looking at all the relations that participate in the constraint, how can one meaningfully check a constraint for violation? When the constraint is known to be satisfied prior to the update, the state of the relations that are available (aka local) can in principle be used to infer something about the relations that are not available (aka remote). This observation is the basis for the existence of tests that guarantee that global consistency is preserved under a given update, without looking at all the base data. In order to make consistency maintenance practical, the challenge is to find those tests that are most general (we call Complete Local Tests) and that are efficient to generate and execute. This paper addresses the problem of finding efficient complete local tests for an important class of constraints that are very common in practice: constraints expressible as conjunctive queries with negated subgoals. For constraints where the predicates for the remote relations do not occur more than once, we present complete local tests under insertions and deletions to the local relations. These tests can be expressed as safe, nonrecursive Datalog ¬ queries against the local relations. These results also apply to other constraints with negation that are not conjunctive.     

Triggered Updates for Temporal Consistency in Real-Time Databases

A real-time database systemhas temporal consistency constraints in addition to timing constraints.The timing constraints require a transaction to be completedby a specified deadline, and the temporal consistency constraintsrequire that temporal data read by a transaction be up-to-date.If a transaction reads out-of-date data, it will become temporallyinconsistent. A real-time database system consists of differenttypes of temporal data objects, including derived objects. Thevalue of a derived object is computed from a set of other objects,known as the read-set of the derived object. The derived objectmay not always reflect the current state of its read-set; a derivedobject can become out-of-date even if its read-set is up-to-date.Any subsequent transaction reading the derived object will thenbecome temporally inconsistent. In this case, in order to readup-to-date objects, a transaction will have to wait until someother transaction updates the out-of-date object. However, indoing so, the waiting transaction may miss its deadline, particularlyif the update is not periodic but instead arrives randomly. Wepropose to update the outdated objects so that not only is thetemporal consistency improved, but also the number of misseddeadlines does not increase significantly, and as a result thereis an overall improvement in the performance of the system. Wepropose, implement and study a novel approach, to be known astriggered updates, to improve temporal consistency in firm real-timedatabase systems when updates are not periodic. We identify propertiesof triggered updates and explain how they work by giving bothan intuitive and a probabilistic analysis. We present strategiesfor generating triggered updates, discuss their suitability invarious contexts and perform a detailed simulation study to evaluatetheir performance. Results show that it is possible to improvetemporal consistency without degrading the timeliness of real-time database systems to a great deal.     

Maintaining Strong Cache Consistency for the Domain Name System

Effective caching in the domain name system (DNS) is critical to its performance and scalability. Existing DNS only supports weak cache consistency by using the time-to-live (TTL) mechanism, which functions reasonably well in normal situations. However, maintaining strong cache consistency in DNS as an indispensable exceptional handling mechanism has become more and more demanding for three important objectives: 1) to quickly respond and handle exceptions such as sudden and dramatic Internet failures caused by natural and human disasters, 2) to adapt increasingly frequent changes of Internet Protocol (IP) addresses due to the introduction of dynamic DNS techniques for various stationed and mobile devices on the Internet, and 3) to provide fine-grain controls for content delivery services to timely balance server load distributions. With agile adaptation to various exceptional Internet dynamics, strong DNS cache consistency improves the availability and reliability of Internet services. In this paper, we first conduct extensive Internet measurements to quantitatively characterize DNS dynamics. Then, we propose a proactive DNS cache update protocol (DNScup), running as middleware in DNS name servers, to provide strong cache consistency for DNS. The core of DNScup is an optimal lease scheme, called dynamic lease, to keep track of the local DNS name servers. We compare dynamic lease with other existing lease schemes through theoretical analysis and trace-driven simulations. Based on the DNS dynamic update protocol, we build a DNScup prototype with minor modifications to the current DNS implementation. Our system prototype demonstrates the effectiveness of DNScup and its easy and incremental deployment on the Internet.     

利用WEB服务实现智能客户端应用

智能客户端是微软最近提出来的概念,指一种充分利用WEB服务所提供的信息的应用程序,依靠服务器端WEB服务的支持,智能客户端实现了强大的用户界面,数据的安全访问,便捷的部署和升级,并能实现数据离线操作。本文在一个具体系统的客户端设计过程中,简要探讨了智能客户端的特点和实现方式。     

高校参考咨询系统中专家调度模式的研究

针对高校参考咨询系统这一新兴的高校数字图书馆业务，分析和探讨了其中的专家调度模式，提出了一种基于层次分析法的专家调度模型，并且验证了它的一致性。针对传统参考咨询系统中的不足，引入提问向导的概念。在提问向导中将提问和检索很好地融合在一起，大大地降低了重复提问的概率，并能很好地避免以往对交叉学科提问时存在的疏漏。     

对等网维护数据副本一致性的乐观复制技术研究

在对等网环境中,保持数据对象的多个副本可极大地提高数据的可靠性和访问效率,但同时需要维护副本数据的一致性。乐观复制在保证数据一致性的同时放松了对操作顺序的要求。提出了数据乐观复制系统形式化模型,应用模型推导了数据达到一致的基本条件,提出了一种基于处理操作的最优化算法,并对实际应用中可能出现的问题提出了改进方法。     

基于计数副本的传统操作系统代码空闲状态监测技术

在系统不重启的情况下对操作系统进行缺陷修复对于保障服务器系统的性能、安全和可用性等方面具有重要意义。因传统操作系统并未针对在线更新进行实现,现有在线更新技术不能较好地处理被更新代码正处于"使用中"状态的问题,导致在有些场景下无法进行在线更新。本文针对这一问题,设计了一种用于传统操作系统的代码使用计数动态监测技术,通过将原始函数二进制码替换为与其对等的计数副本,以及修改相关线程的内核栈上返回地址,实现对于被更新代码执行情况的监测,从而发现可进行在线更新的时机。最后,本文基于开源的动态升级软件ksplice验证了上述工作的有效性。     

IRM: Integrated File Replication and Consistency Maintenance in P2P Systems

In peer-to-peer file sharing systems, file replication and consistency maintenance are widely used techniques for high system performance. Despite significant interdependencies between them, these two issues are typically addressed separately. Most file replication methods rigidly specify replica nodes, leading to low replica utilization, unnecessary replicas and hence extra consistency maintenance overhead. Most consistency maintenance methods propagate update messages based on message spreading or a structure without considering file replication dynamism, leading to inefficient file update and hence high possibility of outdated file response. This paper presents an Integrated file Replication and consistency Maintenance mechanism (IRM) that integrates the two techniques in a systematic and harmonized manner. It achieves high efficiency in file replication and consistency maintenance at a significantly low cost. Instead of passively accepting replicas and updates, each node determines file replication and update polling by dynamically adapting to time-varying file query and update rates, which avoids unnecessary file replications and updates. Simulation results demonstrate the effectiveness of IRM in comparison with other approaches. It dramatically reduces overhead and yields significant improvements on the efficiency of both file replication and consistency maintenance approaches.     

基于事件触发机制的多智能体网络平均一致性研究

文章研究了事件触发机制下具有固定和切换拓扑结构的多智能体网络的平均一致性问题。为了有效降低多智能体一致性控制协议的更新次数,文章提出了仅依赖于各智能体及其邻居节点信息的分布式事件触发机制。为了使多智能体网络渐近收敛至初始平均状态,文章提出了基于事件触发机制的多智能体平均一致性协议。然后,文章建立了事件触发机制下的闭环系统模型,并分别获得了固定和切换拓扑结构下多智能体网络平均一致收敛的充分条件。仿真表明,基于事件触发机制的平均一致性协议保证了多智能体网络的平均一致性,并且分布式事件触发机制有效降低了多智能体一致性协议更新次数。