Development Of Large Systems Research Articles

Distributing Risks and Responsibilities

Coordinated responses to flooding in the New Orleans area date back to at least the early 18 century when the Company of the Indies built a mile long bulwark on a natural levee. Since that early project, corporations, NGOs, and local, state, and federal governments have taken numerous steps to limit the risk of flooding including building extensive levee systems, redirecting rivers, and developing evacuation plans. Like many risks that have been perceived and dealt with for centuries – including crime, disease, and other natural disasters – modern science, engineering, and institutions have worked in concert with one another to address the problem through the development of large socio-technical systems. As with most large socio-technical systems, there was no single group or organization charged with overseeing all facets of the system to limit the risks of flooding in New Orleans. While some parts of the strategy to address the risks were meticulously planned, tightly coupled, and carefully coordinated, other components and institutions were only loosely connected. There was a general sense of who was responsible for what and even a few systems set up to assess whether the specific tasks were being adequately carried out. But the very nature of enormous and complex socio-technical systems is that they are too big to be centrally coordinated in any specific manner. The components of these systems are often developed for a myriad of reasons and their role in any particular mitigation system or strategy may be quite ancillary to their primary purpose. Thus the methods by which responsibilities were distributed amongst the various components were negotiated in an unsystematic and often unspoken way. Despite the fact that the mitigation efforts were not meticulously coordinated, there was certainly some general agreement over what the overall strategy for New Orleans looked like. By tracing the ways in which tasks were distributed around it one can begin to see the ways in which ideas of responsibility hold together complex systems and perhaps why these systems sometimes fail in their mission.

The productivity of large business information system development

We examine key factors that influence the productivity of large business information system application development. A survey method has been selected focusing on large business information system developers in eleven organisations in Hong Kong. The purposes of this study are to identify those factors which are considered by practitioners to have the potential to enhance productivity and also to identify the factors actually adopted in everyday practice. The research indicates that significant factors which affect large business information system development productivity include creating good developmental environments, employing capable and committed IT developers, sound project management and managing IT department activities effectively. We believe that the findings are universal and may be useful to any organisation seeking to improve the productivity of its system development.

Towards efficient web engineering approaches through flexible process models

After more than a decade of web developments and some deafening fiascos, it has become clear that it is not possible to face the development of large scale web systems without following a systematic and well-defined process to guarantee quality, measurability, maintainability and reusability. There are a number of hypermedia/web engineering methods that provide mechanisms to specify the product requirements, including those concerning structure, navigation, interaction, presentation and access. But apart from product requirements there are also process requirements which in the web arena are constantly changing. Hence, to be put in practice without disturbing the project goals nor compromising its success, methods have to rely on empirical and flexible process models that can be easily adapted to fit process requirements. Little attention has been paid to the process model in most hypermedia/web methods which usually apply a classical iterative process based on the use of prototypes that are tested with users. In this paper we describe how we did apply the ADM web engineering method following a flexible star life cycle in the context of a specific web project highlighting the main benefits of this approach. In particular we will describe the different cycles applied in the ARCE project, illustrating the application of a usability engineering life cycle in a real case.

A FRAMEWORK FOR THE EVALUATION OF REAL-TIME SPECIFICATION TECHNIQUES

Real-time and embedded software systems are becoming increasingly prevalent in everyday life. Well-defined and robust methods and methodologies are needed to support development and maintenance of such large and complex systems. Specification of a system is a critical phase of the development life cycle and has a major impact on the quality of a system. Formal specification languages and techniques are needed to be able to comprehensively and correctly define and prototype large and complex systems. The quality of a specification directly depends upon the capability of the language or technique used to express the specification. There is a need to determine language appropriateness for real-time systems. This paper establishes a real-time specification technique evaluation framework based on the features of real-time systems. The features are presented and the paper summarizes the result in terms of major features and their sub-features in a tabular form. This detailed list of features should serve not only to compare the techniques using a common framework, but also to determine the extent of real-time support that these techniques provide for specification. Application of the framework is demonstrated by testing a set of specification techniques for the specification of a real-time embedded cell phone software system.

Towards a Subject-Oriented Model-Driven Framework

Model-Driven Architecture is an approach which tackles such problems as: the high availability that a software product requires to be ready for use, the high degree of evolution that a software system has nowadays, etc. However, in the development of large complex systems, the benefits of that approach have been diminished due to the size and complexity of models that describe these kinds of systems. At this point Aspect-Oriented Software Development (AOSD) appears to improve the understanding, reusability and adaptation of the software artefacts. Its mechanism is based on modularization of crosscutting concerns in well-identified isolated entities called aspects. For this reason we propose to use together AOSD and MDA in the hope of reducing the shortcomings of the latter. Thus, aspects like security, replication, real-time constraints, etc., will be modelled by specialist modellers independently throughout the MDA framework. Our proposal exploits a tool for checking the consistency between different models (aspects) at the same level of abstraction; supporting the traceability of UML elements, requirements, and concerns; and controlling the impact of changes throughout the MDA framework.

Java based middleware for embedded systems

Current tendency in software development of large control systems Icads to products built up from components working in different nodes of a heterogeneous network environment. Particular components becomes an important aspect, since a program composed of components does not guarantee in itself reliable function after the system is extended or a component is replaced. To ensure the system reliability, it is necessary to create a middleware layer. which is able to guarantee a reliable cooperation of components in a heterogeneous distributed environment at application design level. The present paper describes an approach to the design of configurable software architecture focused in the field of control systems. It introduces the technical realization of the middleware layer for reliable and predictable collaboration of components in a distributed heterogeneous industrial environment

An Introduction to Software Testing

The development of large software systems is a complex and error prone process. Faults might occur at any development stage and they must be identified and removed as early as possible to stop their propagation and reduce verification costs. Quality engineers must be involved in the development process since the very early phases to identify required qualities and estimate their impact on the development process. Their tasks span over the whole development cycle and go beyond the product deployment through maintenance and post mortem analysis. Developing and enacting an effective quality process is not a simple task, but it requires that we integrate many quality-related activities with product characteristics, process organization, available resources and skills, and budget constraints.This paper discusses the main characteristics of a good quality process, then surveys the key testing phases and presents modern functional and model-based testing approaches.

Hot-pressed ceramic Cr2+:ZnSe gain-switched laser

The technology of hot-pressed Cr(2+):ZnSe ceramic preparation is reported. Comparative gain-switched lasing of hot-pressed ceramic and CVD grown Cr(2+):ZnSe samples with slope efficiencies up to 10% and output energies up to 2 mJ were demonstrated. This is a milestone in the development of future large scale mid-IR laser systems based on ceramic TM(2+):II-VI materials. This evolving technology has excellent potential as media for practical low cost, high power mid-IR laser applications.

The risks of large organizations in developing complex systems

The risks to the large organization of being able to complete the development of large software intensive systems on time and in budget include not just the defining and maintaining of proper requirements and work processes, but also in the defining and maintaining of the proper organizational work structures. Bureaucratic structures prevalent in large organizations can be inefficient and irrational especially in regards to unplanned surprises, exigencies, contingencies, and the emergent properties that are the normal part of the development and integration of first-of and one-of-a-kind large systems. All large software intensive systems, at their true core, are experiments, and experimentation and bureaucracy have proved to be, over time, destructively antithetical in basic values. Conway's law, first stated in 1968, has by the lights of 2005 provided only a superficial explanation of the antipathy of bureaucracy towards the unknown and risk. The belief that "[o]rganizations which design systems...are constrained to produce designs which are copies of the communication structures of these organizations", can be easily refuted by noting that hierarchical organizations have been quite successful in designing and implementing networked systems subsequent to the invention of Ethernet and TCP/IP. The example provided by Conway was not sufficient by itself to draw the conclusion was made, a common fallacy of cause and effect, but he was certainly on an interesting path of investigation. Large system development failures cannot be prevented soley through improved planning, requirements, processes and software. Improvement must also come from the recognition of the limits of organization, planning and process; the recognition of the basic antipathy between bureaucracy and risk; the recognition of the need for organizational structures that scale yet remain responsive; the recognition of the need for better attitudes towards faults and failures, and the recognition that in order to reduce the risks that are inherent when the technical becomes political, that one must reduce the risks of speaking truth to power. The short article that follows is intended to raise more questions than provide answers, and the references are intended to provide a starting point for further research.

Permission-based ownership

Today's module systems do not effectively support information hiding in the presence of shared mutable objects, causing serious problems in the development and evolution of large software systems. Ownership types have been proposed as a solution to this problem, but current systems have ad-hoc access restrictions and are limited to Java-like languages.In this paper, we describe System F own , an extension of System F with references and ownership. Our design shows both how ownership fits into standard type theory and the encapsulation benefits it can provide in languages with first-class functions, abstract data types, and parametric polymorphism. By looking at ownership in the setting of SystemF, we were able to develop a design that is more principled and flexible than previous ownership type systems, while also providing stronger encapsulation guarantees.

Software engineering for large-scale multi-agent systems - SELMAS'05

Software is becoming present in every aspect of our lives, pushing us inevitably towards a world of distributed, context-aware computing systems. SELMAS'05, "Software Everywhere - Context-Aware Agents", was built on the success of precedent SELMAS workshops, but with a special emphasis on the impact of the agent technology in the development of large context-aware systems. SELMAS has a track record of bringing together researchers and practitioners with a variety of perspectives in order to engage in lively discussion and debate. A particular interest of this workshop was to understand those issues in the agent technology that make it difficult and/or improve the production of context-aware systems.

Problems of the development and quality control of large software systems

Problems of the development and quality control of large software systems

Architectural support in industry: a reflection using C‐POSH

AbstractSoftware architecture plays a vital role in the development (and hence maintenance) of large complex systems (containing millions of lines of code) with a long lifetime. It is therefore required that the software architecture is also maintained, i.e., sufficiently documented, clearly communicated, and explicitly controlled during its life‐cycle. In our experience, these requirements cannot be met without appropriate support.Commercial‐off‐the‐shelf support for architectural maintenance is still scarcely available, if at all, implying the need to develop appropriate proprietary means. In this paper, we reflect upon software architecture maintenance taken within three organizations within Philips that develop professional systems. We extensively describe the experience gained with introducing and embedding of architectural support in these three organizations. We focus on architectural support in the area of software architecture recovery, visualization, analysis, and verification.In our experience, the support must be carried by a number of pillars of software development, and all of these pillars have to go through a change process to ensure sustainable embedding. Managing these changes requires several key roles to be fulfilled in the organization: a champion, a company angel, a change agent, and a target. We call our reflection model C‐POSH, which is an acronym for Change management of the four identified pillars of software development: Process, Organization, Software development environment, and Humans. Our experiences will be presented in terms of the C‐POSH model. Copyright © 2005 John Wiley & Sons, Ltd.

922 超高解像度大型平面入力スキャナの開発に関する考察(GS-16 センシング,研究発表講演)

Digital archiving of graphic art works can significantly contribute to the conservation of these cultural heritage items. Scanner is suitable device to get digital image of graphic art works. The final purpose of this study is development of large high precision scanner system. In this paper, the relation between color information and its material are investigated as the first step of this study. Color information of an object has a lot of information, especially about material. Color information can be shown by spectral reflectance. Spectral reflectance is measured by spectrophotometer and the relation between spectral reflectance and its main component are investigated. The result is as follows. Objects that have same main component show same spectral shape. As the grain size of the object is small, spectral intensity rises.

Double‐stranded secondary structures on mRNA induce type I interferon (IFN α/β) production and maturation of mRNA‐transfected monocyte‐derived dendritic cells

The development of dendritic cell (DC)-based vaccines using antigen-encoding mRNA requires identification of the critical parameters for efficient ex vivo loading of DCs. Exogenously delivered mRNA can induce DC activation, but the molecular mechanisms involved are unknown. The aim of the present study was to identify the means by which mRNA-dependent activation of DCs occurs. In vitro transcribed mRNA molecules were delivered into porcine monocyte-derived DCs (MoDCs) using different non-viral gene transfer procedures. Using the green fluorescent protein (GFP) as reporter gene, as well as rhodamine-labeled RNA, intracellular delivery and transfection efficiency were assessed by confocal microscopy and flow cytometry. DC activation was monitored in terms of MHC class II and CD80/86 upregulation, as well as the production of type I interferon (IFN-alpha/beta). mRNA-lipofected MoDCs produced type I IFN and upregulated MHC class II and CD80/86. Computational analysis of the mRNA molecules predicted highly ordered secondary structures forming double-stranded RNA (dsRNA). This dsRNA was also detectable by immunofluorescence in mRNA-lipofected cells, using antibody specific for dsRNA. Digestion of the mRNA prior to lipofection with a double-strand-specific RNase, but not a single-strand-specific RNase, abrogated DC activation. Impairment of protein kinase R (PKR) with 2-aminopurine also interfered with the activation. Double-stranded secondary structures on mRNA delivered by lipofection can activate MoDCs. This could have important implications for mRNA-based immunomodulation of DCs, DC-based immunotherapy, and formulation of RNA-based vaccines. In addition, this report describes the first in vitro steps towards development of a novel large animal model system to evaluate DC-based vaccines against infectious diseases.

Assertion Based Verification and Analysis of Network Processor Architectures

System level modeling with executable languages such as C/C++ has been crucial in the development of large electronic systems from general processors to application specific designs. To make sure that the executable models behave as they should, the designers often have to "eye-ball" the simulation traces and at best, apply simple "assert" statements or write simple trace checkers in some scripting languages. The problem is the lack of a concise and formal method to specify and check desired properties, whether they be functional or performance in nature. In this paper, we apply assertion checking methodology to the system design of network processors. Functional and performance assertions, based on Linear Temporal Logic and Logic of Constraints, are written during the design process. Trace checkers and simulation monitors are automatically generated to validate particular simulation runs or to analyze their performance characteristics. Several categories of assertions are checked throughout the design process, such as equivalence, functionality, transaction, and performance. We demonstrate that the assertion-based methodology is very useful for both system level verification and design exploration.

컴포넌트 상호작용 명세기법

컴포넌트 기반 플랫폼이 갖는 주 한계점은 컴포넌트 기반 시스템이 갖는 컴포넌트간 상호작용과 관련된 행위에 대한 기술과 추론에 대한 적당한 수단을 제대로 제공하지 못한다는 것이다. 현 컴포넌트 기반 플래폼이 CORBA의 IDL과 같은 컴포넌트에 대한 시그네쳐 수준에서의 기술을 위한 방법을 제공하기는 하지만, 이는 컴포넌트들간 상호작용과 같은 행위 기술을 저 수준에서 제공하는 것에 불과하다 컴포넌트 기반 시스템에서 중요한 과제 중 하나는 바로 컴포넌트들간 상호작용과 같은 행위를 기술하는 것과 이를 바탕으로 행위에 대한 검증이 필요하다. 본 논문에서는 프로세스 대수를 이용하여 컴포넌트 인터페이스의 명세 정의와 추상화된 소프트웨어 아키텍처를 어떻게 만족시킬 것인가를 보장받기위한 role의 결합방법, 생성된 아키텍처에 대한 전개규칙, 시각화방법을 연구한다. 또한 사례연구를 통해 본 논문에서 제안한 명세기법 및 정의의 타당함을 보였다. A major limitation of available component-based platforms Is that they do not provide suitable means for describing and reasoning on the concurrent behaviour of interacting component-based system. Indeed while these platforms provide convenient ways to describe the typed signatures of components, e.g. like CORBA's IDL, they offer a quite low-level support to describe the concurrent behaviour of component. The ability to describe and verify the concurrent behaviour of interacting components is key aspect In the development of large component-based software system. This study propose a component interface specification using process algebra and configuration's role which allows one to prove correctness of software architecture generated at design level as well as to define compatibility relations by our evolution rule and <TEX>$\pi$</TEX>-graph. Also, we shown on an appropriateness of a specification techniques and definitions proposed in this paper by case-study.

Towards a theory of multimedia metacomputing

Multimedia metacomputing is a new approach to the management and processing of multimedia data in web-based information systems. It offers high flexibility and openness while shielding the applications from any system internals. Starting with the vision of a completely open and globally distributed multimedia information system, we consider abstraction concepts required, especially transformation independence, and an appropriate semantic model.Thus, the major focus of this paper is on the abstract data and processing model called VirtualMedia,11This work is supported by the Deutsche Forschungsgemeinschaft (DFG) as part of the Sonderforschungsbereich (SFB) 501 “Development of Large Systems with Generic Methods”. which provides a transformation independence framework for multimedia processing. In particular, we describe how transformation requests are represented and processed, exploiting semantic equivalence relations on filter graphs and redundant materialization, finally yielding instantiatable plans for materializing the requested media object(s) at the client.

A UML-based language for specifying domain-specific patterns

Cost-effective development of large computer-based systems can be realized through systematic reuse of domain-specific design experience. Such experience can be captured by domain-specific design patterns, that is, patterns specifying design solutions for well-defined families of applications. This paper presents a pattern specification notation called the Role-Based Metamodeling Language (RBML) and shows how it can be used to express domain-specific patterns. An RBML pattern defines a domain-specific sub-language of the Unified Modeling Language (UML). Developers can use the sub-language to create UML diagrams for applications in the domain addressed by the pattern. An RBML pattern characterizing checkin–checkout (CICO) applications is presented. Applications in the CICO domain manage checking in and out of items. Examples of such applications are car rental, library, and video rental systems. The paper also describes how a UML model of a library system can be obtained from the RBML CICO pattern.

TOUGH TIMES FOR 2005 R&amp;D BUDGET

ECORD FEDERAL BUDGET DEFICITS and a continued emphasis on spending for national defense and homeland security combine to keep funding increases for nondefense research and development spending essentially zero next year. President George W Bush's record $2.4 trillion federal budget for fiscal 2005 includes just $ 132 billion dedicated to R&D. Although this is a 5% increase from 2004, only about $60 billion of that is supporting traditional research programs. Much of the rest goes to development of large weapons systems by the Department of Defense. Even if large increases for R&D funding at the Department of Homeland Security (DHS) are taken into account, overall federal research support has fallen almost 1% from Bush's proposal a year ago (C&EN, Feb. 9, page 8). The Administration tried to put a good face on the low numbers. John H. Marburger III, science adviser to the President and director of the Office of Science & Technology Policy told ...