Ronald H. Perrott

An improved parallel thinning algorithm

An iterative thinning algorithm reduces a two-dimensional pattern of strokes to its skeleton by removing layers of edge elements until each stroke has unit thickness. A parallel solution requires the independent calculation of new values for each iteration, using a window of nearest neighbors for each element. The traditional need for at least two subiterations can be avoided by modifying the window to permit the availability of intermediate calculations. Timings on an ICL DAP (an array processor) indicate an improvement of over 40 percent. Additional refinements are suggested to reduce noise in the final skeleton.

A Language for Array and Vector Processors

The scientific community has consistently demanded from computing machines an increase in the number of instructions executed per second. The latest increase has been achieved by duplication of arithmetic units for an array processor and the pipelining of functional units for vector processors. The high level programming languages for such machines have not benefited from the advances which have been made in programming language design and implementation techniques. A high level language is described in this paper which is appropriate for both array and vector processors and is defined without reference to the hardware of either type of machine. The syntax enables the parallel nature of a problem to be expressed in a form which can be readily exploited by these machines. This is achieved by using the data declarations to indicate the maximum extent of parallel processing and then to manipulate this, or a lesser extent, in the course of program execution. It was found to be possible to modify many of the structured programming and data structuring concepts for this type of parallel environment and to maintain the benefits of compile time and run time checking. Several special constructs and operators are also defined. The language offers to the large scale scientific computing community many of the advances which have been made in software engineering techniques while it exploits the architectural advances which have been made.

Provider-Independent Use of the Cloud

Utility computing offers researchers and businesses the potential of significant cost-savings, making it possible for them to match the cost of their computing and storage to their demand for such resources. A utility compute provider enables the purchase of compute infrastructures on-demand; when a user requires computing resources a provider will provision a resource for them and charge them only for their period of use of that resource. There has been a significant growth in the number of cloud computing resource providers and each has a different resource usage model, application process and application programming interface (API)---developing generic multi-resource provider applications is thus difficult and time consuming. We have developed an abstraction layer that provides a single resource usage model, user authentication model and API for compute providers that enables cloud-provider neutral applications to be developed. In this paper we outline the issues in using external resource providers, give examples of using a number of the most popular cloud providers and provide examples of developing provider neutral applications. In addition, we discuss the development of the API to create a generic provisioning model based on a common architecture for cloud computing providers.

Syntax-directed editing

The conventional approach to the construction of a program is to view it as a sequence of characters. However, it is now becoming more meaningful to perceive and develop a program in terms of its inherent syntactic structure. This paper introduces the topic of syntax-directed editing in general and then considers some of the problems and design decisions that are encountered when implementing a syntax-directed system. An overview of a range of structure-oriented editors exhibiting varying degrees of syntax direction is also included.

A constraints-based resource discovery model for multi-provider cloud environments

AbstractIncreasingly infrastructure providers are supplying the cloud marketplace with storage and on-demand compute resources to host cloud applications. From an application user’s point of view, it is desirable to identify the most appropriate set of available resources on which to execute an application. Resource choice can be complex and may involve comparing available hardware specifications, operating systems, value-added services (such as network configuration or data replication) and operating costs (such as hosting cost and data throughput). Providers’ cost models often change and new commodity cost models (such as spot pricing) can offer significant savings. In this paper, a software abstraction layer is used to discover the most appropriate infrastructure resources for a given application, by applying a two-phase constraints-based approach to a multi-provider cloud environment. In the first phase, a set of possible infrastructure resources is identified for the application. In the second phase, a suitable heuristic is used to select the most appropriate resources from the initial set. For some applications a cost-based heuristic may be most appropriate; for others a performance-based heuristic may be of greater relevance. A financial services application and a high performance computing application are used to illustrate the execution of the proposed resource discovery mechanism. The experimental results show that the proposed model can dynamically select appropriate resouces for an application’s requirements.

An application-centric model for cloud management

The cloud model is increasingly popular as a means of creating dynamic, flexible and cost effective networkcentric application infrastructures. The model separates the applications, or application cloud, from the resources, or resource cloud, upon which the applications will be hosted. There are an increasing number of utility resource providers that aim to provide cloud infrastructure on demand to users and libraries that aim to manage owned infrastructure as a resource cloud. There is, unfortunately, no common API for cloud resources and it is unlikely that one will emerge soon given the immaturity of the area and the competing commercial interests in the domain. In this paper, we outline our commodity and application-centric approach to resource management, and describe our integration framework for cloud application management-illustrating its use in a field deployed application and a particular dynamic component within that application.

A comparison of two parallel algorithms for the symmetric eigenproblem

Two algorithms for the eigensolution of real symmetric matrices [8] are discussed. The first is a variant of the classical Jacobi method and the second is also based on orthogonal transforms. Both algorithms may be readily expressed in forms which allow the power of SIMD (single instruction multiple data) machines to be exploited. The performances of the algorithms on the ICL DAP (distributed array processor) and the CRAY-1 are compared.

The programming language ACTUS

This paper describes, in an informal manner, the programming language ACTUS which was designed to facilitate programming array processing and vector processing ‘supercomputers’.

A statistical multiresolution approach for face recognition using structural hidden Markov models

This paper introduces a novel methodology that combines the multiresolution feature of the discrete wavelet transform (DWT) with the local interactions of the facial structures expressed through the structural hidden Markov model (SHMM). A range of wavelet filters such as Haar, biorthogonal 9/7, and Coiflet, as well as Gabor, have been implemented in order to search for the best performance. SHMMs perform a thorough probabilistic analysis of any sequential pattern by revealing both its inner and outer structures simultaneously. Unlike traditional HMMs, the SHMMs do not perform the state conditional independence of the visible observation sequence assumption. This is achieved via the concept of local structures introduced by the SHMMs. Therefore, the long-range dependency problem inherent to traditional HMMs has been drastically reduced. SHMMs have not previously been applied to the problem of face identification. The results reported in this application have shown that SHMM outperforms the traditional hidden Markov model with a 73% increase in accuracy.

Gridcast: a grid and Web service broadcast infrastructure

The Gridcast project is pioneering the use of grid and Web technologies to prototype the next generation of broadcast media infrastructure. The project has a physical network infrastructure that connects BBC Northern Ireland, BBC R&D in London, the Belfast e-Science Centre and the emerging UK grid infrastructure. This physical network infrastructure is being used to test grid and Web services that manage a television broadcast infrastructure; that is, an infrastructure that manages a collection of broadcast schedules to be transmitted to viewers and that contains a mix of live and recorded video. Broadcasting is a highly demanding industry with high levels of reactivity and robustness required in its infrastructure - it is a significant test of current grid and Web technologies. The project has implemented schedule-based transport of recorded video between broadcast locations that assumes a highly reactive broadcasting environment, the remote, secure use of technical resources and the automation of broadcast and production workflows.