No description available.

This thesis addresses the formal veri cation of a support infrastructure for resilient dynami- cally recon gurable systems. A component-based system, whose architectural con guration may change at runtime, is classed as dynamically recon gurable. Such systems require a support infrastructure for the control of recon gurations to provide resilience. The veri cation of such recon guration support increases the trust that developers and stakeholders may place on the system. The thesis de nes an architectural model of an infrastructure of services for the support of dynamic recon guration and takes a formal approach to the de nition and veri cation of one aspect of the infrastructure. The execution of recon guration policies in a recon guration infrastructure provides guidance to the architectural change to be enacted on a recon gurable system. These recon guration policies are often produced using a language with informal syntax and no formal semantics. Predicting properties of these policies governing recon guring systems has yet to be attempted. In this thesis, we de ne RPL { a recon guration policy language with a formal syntax and semantics. With the use of a case study, theories of RPL and an example policy are developed and the veri cation of key proof obligations and validation conjectures of policies expressed in RPL is demonstrated. The contribution of the thesis is two-fold. Firstly, the architectural de nition of a support infrastructure provides a lasting contribution in that it suggests a clear direction for future work in dynamic recon guration. Secondly, through the formal de nition of RPL and the veri cation of properties of policies, the thesis provides a basis for the use of formal veri cation in dynamic recon guration and, more speci cally, in policies for dynamic recon guration.

No description available.

Tracing the history of software engineering reveals a series of abstractions. In early days, software engineers would construct software using machine code. As time progressed, software engineers and computer scientists developed higher levels of abstraction in order to provide tools to assist in building larger software systems. This has resulted in high-level languages, modelling languages, design patterns, and software architecture. Software architecture has been recognised as an important tool for designing and building software. Some research takes the view that the success or failure of a software development project depends heavily on the quality of the software architecture. For any software system, there are a number of individuals who have some interest in the architecture. These stakeholders have differing requirements of the software architecture depending on the role that they take. Stakeholders include the architects, designers, developers and also the sales, services and support teams and even the customer for the software. Communication and understanding of the architecture is essential in ensuring that each stakeholder can play their role during the design, development and deployment of that software system. Software visualisation has traditionally been focused on aiding the understanding of software systems by those who perform development and maintenance tasks on that software. In supporting developers and maintainers, software visualisation has been largely concerned with representing static and dynamic aspects of software at the code level. Typically, a software visualisation will represent control flow, classes, objects, import relations and other such low level abstractions of the software. This research identifies the fundamental issues concerning software architecture visualisation. It does this by identifying the practical use of software architecture in the real world, and considers the application of software visualisation techniques to the visualisation of software architecture. The aim of this research is to explore the ways in which software architecture visualisation can assist in the tasks undertaken by the differing stakeholders in a software system and its architecture. A prototype tool, named ArchVis, has been developed to enable the exploration of some of the fundamental issues in software architecture visualisation. ArchVis is a new approach to software architecture visualisation that is capable of utilising multiple sources and representations of architecture in order to generate multiple views of software architecture. The mechanism by which views are generated means that they can be more relevant to a wider collection of stakeholders in that architecture. During evaluation ArchVis demonstrates the capability of utilising a number of data sources in order to produce architecture visualisations. Arch Vis' view model is capable of generating the necessary views for architecture stakeholders and those stakeholders can navigate through the views and data in order to obtain relevant information. The results of evaluating ArchVis using a framework and scenarios demonstrate that the majority of the objectives of this research have been achieved.

The dynamics of crowd movements are self-organising and often involve complex pattern formations. Although computational models have recently been developed, it is unclear how well their underlying methods capture local dynamics and longer-range aspects, such as evacuation. A major part of this thesis is devoted to an investigation of current methods, and where required, the development of alternatives. The main purpose is to utilise realistic models of pedestrian crowds in the design of fitness functions for an evolutionary approach to architectural design. We critically review the state-of-the-art in pedestrian and evacuation dynamics. The concept of 'Multi-Agent System' embraces a number of approaches, which together encompass important local and longer-range aspects. Early investigations focus on methods-cellular automata and attractor fields-designed to capture these respective levels. The assumption that pattern formations in crowds result from local processes is reflected in two dimensional cellular automata models, where mathematical rules operate in local neighbourhoods. We investigate an established cellular automata and show that lane-formation patterns are stable only in a low-valued density range. Above this range, such patterns suddenly randomise. By identifying and then constraining the source of this randomness, we are only able to achieve a small degree of improvement. Moreover, when we try to integrate the model with attractor fields, no useful behaviour is achieved, and much of the randomness persists. Investigations indicate that the unwanted randomness is associated with 2-lattice phase transitions, where local dynamics get invaded by giant-component clusters during the onset of lattice percolation. Through this in-depth investigation, the general limits to cellular automata are ascertained-these methods are not designed with lattice percolation properties in mind and resulting models depend, often critically, on arbitrarily chosen neighbourhoods. We embark on the development of new and more flexible methodologies. Rather than treating local and global dynamics as separate entities, we combine them. Our methods are responsive to percolation, and are designed around the following principles: 1) Inclusive search provides an optimal path between a pedestrian origin and destination. 2) Dynamic boundaries protect search and are based on percolation probabilities, calculated from local density regimes. In this way, more robust dynamics are achieved. Simultaneously, longer-range behaviours are also specified. 3) Network-level dynamics further relax the constraints of lattice percolation and allow a wider range of pedestrian interactions. Having defined our methods, we demonstrate their usefulness by applying them to lane-formation and evacuation scenarios. Results reproduce the general patterns found in real crowds. We then turn to evolution. This preliminary work is intended to motivate future research in the field of Evolutionary Architecture. We develop a genotype-phenotype mapping, which produces complex architectures, and demonstrate the use of a crowd-flow model in a phenotype-fitness mapping. We discuss results from evolutionary simulations, which suggest that obstacles may have some beneficial effect on crowd evacuation. We conclude with a summary, discussion of methodological limitations, and suggestions for future research.

No description available.

No description available.

No description available.

No description available.

The influence of embedded systems is felt in many aspects of our daily lives; being particularly apparent in consumer electronics and automotive products. Customer demand and rapid advances in the complexity of the underlying technology has enabled the introduction of new systems and services that were simply not feasible just a few years ago. Although the cost of embedded systems is an important design parameter, their development is also affected by performance and functionality. The performance issue is traditionally addressed by the design of faster microprocessors, but more recently by the exploitation of parallelism (for example, vector units and very long instruction word processors), as well as special purpose hardware architectures, such as graphics processing units and network cards. In such systems, however, the main performance bottleneck is often the memory hierarchy, particularly in systems with complex memory access arbitration, where read or write operations to the main memory could result in delays of thousands of cycles. AI though the widespread use of cache memories aims to alleviate this effect to some extent, memory access penalties remain a significant drain on performance. Functionality is closely related to the memory capacity available, particularly in portable systems such as mobile phones and handheld games consoles. The work described in this thesis includes a comprehensive analysis of code size and performance issues of embedded reduced instruction set computer architectures.