A structural method for the design of fault tolerant distributed control systems

Distributed digital control systems provide alternatives to conventional, centralised digital control systems. Typically, a modern distributed control system will comprise a multi-processor or network of processors, a communications network, an associated set of sensors and actuators, and the systems and applications software. This thesis addresses the problem of how to design robust decentralised control systems, such as those used to control event-driven, real-time processes in time-critical environments. Emphasis is placed on studying the dynamical behaviour of a system and identifying ways of partitioning the system so that it may be controlled in a distributed manner. A structural partitioning technique is adopted which makes use of natural physical sub-processes in the system, which are then mapped into the software processes to control the system. However, communications are required between the processes because of the disjoint nature of the distributed (i.e. partitioned) state of the physical system. The structural partitioning technique, and recent developments in the theory of potential controllability and observability of a system, are the basis for the design of controllers. In particular, the method is used to derive a decentralised estimate of the state vector for a continuous-time system. The work is also extended to derive a distributed estimate for a discrete-time system. Emphasis is also given to the role of communications in the distributed control of processes and to the partitioning technique necessary to design distributed and decentralised systems with resilient structures. A method is presented for the systematic identification of necessary communications for distributed control. It is also shwon that the structural partitions can be used directly in the design of software fault tolerant concurrent controllers. In particular, the structural partition can be used to identify the boundary of the conversation which can be used to protect a specific part of the system. In addition, for certain classes of system, the partitions can be used to identify processes which may be dynamically reconfigured in the event of a fault. These methods should be of use in the design of robust distributed systems.

The design of robust protocols for distributed real-time systems

Modern distributed control systems comprise of a set of processors which are interconnected using a suitable communication network. For use in real-time control environments, such systems must be deterministic and generate specified responses within critical timing constraints. Also, they should be sufficiently robust to survive predictable events such as communication or processor faults. This thesis considers the problem of coordinating and synchronizing a distributed real-time control system under normal and abnormal conditions. Distributed control systems need to periodically coordinate the actions of several autonomous sites. Often the type of coordination required is the all or nothing property of an atomic action. Atomic commit protocols have been used to achieve this atomicity in distributed database systems which are not subject to deadlines. This thesis addresses the problem of applying time constraints to atomic commit protocols so that decisions can be made within a deadline. A modified protocol is proposed which is suitable for real-time applications. The thesis also addresses the problem of ensuring that atomicity is provided even if processor or communication failures occur. Previous work has considered the design of atomic commit protocols for use in non time critical distributed database systems. However, in a distributed real-time control system a fault must not allow stringent timing constraints to be violated. This thesis proposes commit protocols using synchronous communications which can be made resilient to a single processor or communication failure and still satisfy deadlines. Previous formal models used to design commit protocols have had adequate state coverability but have omitted timing properties. They also assumed that sites communicated asynchronously and omitted the communications from the model. Timed Petri nets are used in this thesis to specify and design the proposed protocols which are analysed for consistency and timeliness. Also the communication system is mcxielled within the Petri net specifications so that communication failures can be included in the analysis. Analysis of the Timed Petri net and the associated reachability tree is used to show the proposed protocols always terminate consistently and satisfy timing constraints. Finally the applications of this work are described. Two different types of applications are considered, real-time databases and real-time control systems. It is shown that it may be advantageous to use synchronous communications in distributed database systems, especially if predictable response times are required. Emphasis is given to the application of the developed commit protocols to real-time control systems. Using the same analysis techniques as those used for the design of the protocols it can be shown that the overall system performs as expected both functionally and temporally.

Comparing transport emissions and impacts for energy recovery from domestic waste (EfW): centralised and distributed disposal options for two UK Counties

Many local authorities (LAs) are currently working to reduce both greenhouse gas emissions and the amount of municipal solid waste (MSW) sent to landfill. The recovery of energy from waste (EfW) can assist in meeting both of these objectives. The choice of an EfW policy combines spatial and non-spatial decisions which may be handled using Multi-Criteria Analysis (MCA) and Geographic Information Systems (GIS). This paper addresses the impact of transporting MSW to EfW facilities, analysed as part of a larger decision support system designed to make an overall policy assessment of centralised (large-scale) and distributed (local-scale) approaches. Custom-written ArcMap extensions are used to compare centralised versus distributed approaches, using shortest-path routing based on expected road speed. Results are intersected with 1-kilometre grids and census geographies for meaningful maps of cumulative impact. Case studies are described for two counties in the United Kingdom (UK); Cornwall and Warwickshire. For both case study areas, centralised scenarios generate more traffic, fuel costs and emitted carbon per tonne of MSW processed.

The development of a multi-function computer-assisted instruction system using a hierarchical database structure as the lesson compendium.

Computer-Based Learning systems of one sort or another have been in existence for almost 20 years, but they have yet to achieve real credibility within Commerce, Industry or Education. A variety of reasons could be postulated for this, typically: - cost - complexity - inefficiency - inflexibility - tedium Obviously different systems deserve different levels and types of criticism, but it still remains true that Computer-Based Learning (CBL) is falling significantly short of its potential. Experience of a small, but highly successful CBL system within a large, geographically distributed industry (the National Coal Board) prompted an investigation into currently available packages, the original intention being to purchase the most suitable software and run it on existing computer hardware, alongside existing software systems. It became apparent that none of the available CBL packages were suitable, and a decision was taken to develop an in-house Computer-Assisted Instruction system according to the following criteria: - cheap to run; - easy to author course material; - easy to use; - requires no computing knowledge to use (as either an author or student) ; - efficient in the use of computer resources; - has a comprehensive range of facilities at all levels. This thesis describes the initial investigation, resultant observations and the design, development and implementation of the SCHOOL system. One of the principal characteristics c£ SCHOOL is that it uses a hierarchical database structure for the storage of course material - thereby providing inherently a great deal of the power, flexibility and efficiency originally required. Trials using the SCHOOL system on IBM 303X series equipment are also detailed, along with proposed and current development work on what is essentially an operational CBL system within a large-scale Industrial environment.

A study of adaptive load balancing algorithms for distributed systems

With the advent of distributed computer systems with a largely transparent user interface, new questions have arisen regarding the management of such an environment by an operating system. One fertile area of research is that of load balancing, which attempts to improve system performance by redistributing the workload submitted to the system by the users. Early work in this field concentrated on static placement of computational objects to improve performance, given prior knowledge of process behaviour. More recently this has evolved into studying dynamic load balancing with process migration, thus allowing the system to adapt to varying loads. In this thesis, we describe a simulated system which facilitates experimentation with various load balancing algorithms. The system runs under UNIX and provides functions for user processes to communicate through software ports; processes reside on simulated homogeneous processors, connected by a user-specified topology, and a mechanism is included to allow migration of a process from one processor to another. We present the results of a study of adaptive load balancing algorithms, conducted using the aforementioned simulated system, under varying conditions; these results show the relative merits of different approaches to the load balancing problem, and we analyse the trade-offs between them. Following from this study, we present further novel modifications to suggested algorithms, and show their effects on system performance.

The relative performance of scalable load balancing algorithms in loosely-coupled distributed systems. Available in 2 volumes.

The computer systems of today are characterised by data and program control that are distributed functionally and geographically across a network. A major issue of concern in this environment is the operating system activity of resource management for different processors in the network. To ensure equity in load distribution and improved system performance, load balancing is often undertaken. The research conducted in this field so far, has been primarily concerned with a small set of algorithms operating on tightly-coupled distributed systems. More recent studies have investigated the performance of such algorithms in loosely-coupled architectures but using a small set of processors. This thesis describes a simulation model developed to study the behaviour and general performance characteristics of a range of dynamic load balancing algorithms. Further, the scalability of these algorithms are discussed and a range of regionalised load balancing algorithms developed. In particular, we examine the impact of network diameter and delay on the performance of such algorithms across a range of system workloads. The results produced seem to suggest that the performance of simple dynamic policies are scalable but lack the load stability of more complex global average algorithms.

Some aspects of system design for production management in the rolling-mill industry

Case studies in copper-alloy rolling mill companies showed that existing planning systems suffer from numerous shortcomings. Where computerised systems are in use, these tend to simply emulate older manual systems and still rely heavily on modification by experienced planners on the shopfloor. As the size and number of orders increase, the task of process planners, while seeking to optimise the manufacturing objectives and keep within the production constraints, becomes extremely complicated because of the number of options for mixing or splitting the orders into batches. This thesis develops a modular approach to computerisation of the production management and planning functions. The full functional specification of each module is discussed, together with practical problems associated with their phased implementation. By adapting the Distributed Bill of Material concept from Material Requirements Planning (MRP) philosophy, the production routes generated by the planning system are broken down to identify the rolling stages required. Then to optimise the use of material at each rolling stage, the system generates an optimal cutting pattern using a new algorithm that produces practical solutions to the cutting stock problem. It is shown that the proposed system can be accommodated on a micro-computer, which brings it into the reach of typical companies in the copper-alloy rolling industry, where profit margins are traditionally low and the cost of widespread use of mainframe computers would be prohibitive.

20-km distributed temperature sensor based on spontaneous Brillouin scattering

We report on a novel experimental setup for distributed measurement of temperature, based on spontaneous Brillouin scattering in optical fiber. We have developed a mode-locked Brillouin fiber ring laser in order to generate the dual frequency source required for a heterodyne detection of the backscattered signal. This relatively simple system enables temperature measurements over 20 km with a spatial resolution of 7 m.

Distributed temperature sensor based on Brillouin loss in an optical fibre for transient threshold monitoring

A distributed temperature sensor for transient threshold monitoring with a 22 km sensing length, based on the Brillouin loss in standard communications fibre, is demonstrated. The system can be used for real-time monitoring of a preset temperature threshold. Good S/N ratios were achieved with only 8–16 sample averages giving a response time of 2 to 4 s with a temperature uncertainty of ±1 °C.

Experimental and theoretical studies on a distributed temperature sensor based on Brillouin scattering

Results are reported from recent research on the use of the Brillouin gain/loss mechanism for distributed sensing. A theoretical model of the interaction of the pulsed and CW beams is described and compared with experiments. Results from a system with a 51 km sensing length are presented. We finally investigate issues related to the variation within the sensing fiber of the polarizations of the two beams.

Distributed temperature change sensor, based on Rayleigh backscattering in an optical fibre

A frequency-modulated continuous-wave technique is used to detect the presence of frequency shifts in the Rayleigh-backscattered light in a single-mode optical fiber as a result of a changing temperature. The system is able to detect a rate of temperature change of 0.014 K/s, when a 20-cm length of fiber is heated. The system is also able to demonstrate a spatial resolution of better than 15 cm.

Meta level component-based framework for distributed computing applications

Adaptability for distributed object-oriented enterprise frameworks is a critical mission for system evolution. Today, building adaptive services is a complex task due to lack of adequate framework support in the distributed computing environment. In this thesis, we propose a Meta Level Component-Based Framework (MELC) which uses distributed computing design patterns as components to develop an adaptable pattern-oriented framework for distributed computing applications. We describe our novel approach of combining a meta architecture with a pattern-oriented framework, resulting in an adaptable framework which provides a mechanism to facilitate system evolution. The critical nature of distributed technologies requires frameworks to be adaptable. Our framework employs a meta architecture. It supports dynamic adaptation of feasible design decisions in the framework design space by specifying and coordinating meta-objects that represent various aspects within the distributed environment. The meta architecture in MELC framework can provide the adaptability for system evolution. This approach resolves the problem of dynamic adaptation in the framework, which is encountered in most distributed applications. The concept of using a meta architecture to produce an adaptable pattern-oriented framework for distributed computing applications is new and has not previously been explored in research. As the framework is adaptable, the proposed architecture of the pattern-oriented framework has the abilities to dynamically adapt new design patterns to address technical system issues in the domain of distributed computing and they can be woven together to shape the framework in future. We show how MELC can be used effectively to enable dynamic component integration and to separate system functionality from business functionality. We demonstrate how MELC provides an adaptable and dynamic run time environment using our system configuration and management utility. We also highlight how MELC will impose significant adaptability in system evolution through a prototype E-Bookshop application to assemble its business functions with distributed computing components at the meta level in MELC architecture. Our performance tests show that MELC does not entail prohibitive performance tradeoffs. The work to develop the MELC framework for distributed computing applications has emerged as a promising way to meet current and future challenges in the distributed environment.

Vibration analysis of submerged rectangular microplates with distributed mass loading

This paper investigates the vibration characteristics of the coupling system of a microscale fluid-loaded rectangular isotropic plate attached to a uniformly distributed mass. Previous literature has, respectively, studied the changes in the plate vibration induced by an acoustic field or by the attached mass loading. This paper investigates the issue of involving these two types of loading simultaneously. Based on Lamb's assumption of the fluid-loaded structure and the Rayleigh–Ritz energy method, this paper presents an analytical solution for the natural frequencies and mode shapes of the coupling system. Numerical results for microplates with different types of boundary conditions have also been obtained and compared with experimental and numerical results from previous literature. The theoretical model and novel analytical solution are of particular interest in the design of microplate-based biosensing devices.

All-optical OFDM and distributed Raman amplification : challenges to enable high capacities and extend reach

We review recent advances in all-optical OFDM technologies and discuss the performance of a field trial of a 2 Tbit/s Coherent WDM over 124 km with distributed Raman amplification. The results indicate that careful optimisation of the Raman pumps is essential. We also consider how all-optical OFDM systems perform favourably against energy consumption when compared with alternative coherent detection schemes. We argue that, in an energy constrained high-capacity transmission system, direct detected all-optical OFDM with `ideal' Raman amplification is an attractive candidate for metro area datacentre interconnects with ~100 km fibre spans, with an overall energy requirement at least three times lower than coherent detection techniques.

"Magic" dispersion maps with distributed Raman amplification

We analyze pulse propagation in an optical fiber with a periodic dispersion map and distributed amplification. Using an asymptotic theory and a momentum method, we identify a family of dispersion management schemes that are advantageous for massive multichannel soliton transmission. For the case of two-step dispersion maps with distributed Raman amplification to compensate for the fiber loss, we find special schemes that have optimal (chirp-free) launch point locations that are independent of the fiber dispersion. Despite the variation of dispersion with wavelength due to the fiber dispersion slope, the transmission in several different channels can be optimized simultaneously using the same optimal launch point. The theoretical predictions are verified by direct numerical simulations. The obtained results are applied to a practical multichannel transmission system.