EPSRC PhD Studentships
Distributed system evaluation through fault injection (3 years)
Evolutionary Service-Oriented Architecture (3 years)
This project is appropriate for a software engineering graduate with a strong validation and verification background or a computer science graduate with a strong programming background. The student will join the Distributed Systems and Services research group located within the School of Computing at the University of Leeds (see http://www.comp.leeds.ac.uk/distsys/).
The studentships are part of the EPSRC/BAE Systems funded research project in system engineering addressing the question: “Are you prepared for NEC (Network Enabled Capability)?” The NECTISE (Network Enabled Capability Through Innovative Systems Engineering) project is a five and half year research endeavour involving the Universities of: Loughborough, Bath, Cambridge, Cranfield, Leeds, Leicester, Manchester, Queen’s Belfast, Strathclyde, and York, with a value of around £8.4M. The University of Leeds is one of the four academic leaders for the entire programme, leading research activities on System Architectures and contributing to the development of Through Life System Management. These posts will be linked with the Systems Architecture topic group.
Studentship 1. Distributed system evaluation through fault injection (3 years)
The first studentship will be linked to the evaluation of architectures work package, investigating the evaluation of system dependability through the use of fault injection. Noted by a number of projects in academic and industrial circles, fault injection is a useful tool for assessing the dependability and robustness of systems. Fault injection is an important evaluation tool but its use requires detailed knowledge of the system architecture being applied to along with available fault models.
This studentship will examine the application of heuristic rules to automate fault injection placement. Whilst it is relatively easy to implement fault injection tools or hand insert fault injection code, it requires a great deal of skill and detailed knowledge of the system under test to place the fault injection points in a system so that they provide optimal coverage.
Research projects have implemented fault injection frameworks that semi-automate the creation of fault injection campaigns for a system. Unfortunately these systems fail to automate the actual placement of fault injection points within a system and thus require experienced domain experts to operate them with detailed knowledge of system under test. Published work indicates that a method based on heuristic rules may allow automation of the process by guiding fault injection point placement.
The aim of this studentship is to devise a method to guide automated fault placement enabled by organizing heuristic rules, their associated fault models and coverage statistics as well as the system descriptions into a number of ontologies which will determine appropriate injection points. The method will be implemented as an example tool, either as a standalone tool or as an add-on to an existing tool, for example a plug-in for the Eclipse integrated development environment.
Studentship 2. Evolutionary Service-Oriented Architecture (3 years)
The second studentship will be linked to the NECTISE work on evolutionary architectures and contribute to the understanding of architectural requirements in the pursuit of robust, dependable solutions over the development cycles of distributed systems.
The objectives of this studentship are:
to develop an evolutionary service-oriented model for the provision of dependable and sustainable capability,
to relate the model to realistic systems for the provision of such capability, and
to evaluate the performance of this model through mathematical analysis and simulations.
This project will focus on evolutionary methods for managing service development, service discovery and service integration in a critical and dynamic environment. This will include exploring novel architectures that can adapt to changes due to, for example: faults; customer need; technology developments and obsolescence.
In this doctoral programme, mathematical and simulation-based evaluations will be used to scrutinise new models of service integration and discovery to deliver dependable capability. Evaluation areas will include optimisation of resources, architecture agility, managing dynamic availability and minimising the impact on affordability.
Entry Requirements: Applicants should have a BSc 2(i) level (or equivalent) background in Software Engineering, Computer Science or other relevant discipline.
The studentship will start imminently. Details on how to apply can be found at: http://www.engineering.leeds.ac.uk/comp/pg/PGRapply.shtml
This studentship will provide maintenance and fees for UK students or EU students with three year’s UK residence. The studentship will follow the standard EPSRC studentship rate (the standard rate can be found at http://www.epsrc.ac.uk/PostgraduateTraining/default.htm ).
Applications will be considered up to 31 December 2008. Early applications will be considered immediately.
For more information about the posts please contact:
Dr Lu Liu, School of Computing, University of Leeds, Leeds, LS2 9JT Email: [email protected]
Dr Duncan Russell, School of Computing, University of Leeds, Leeds, LS2 9JT Email: [email protected] Tel. +44 (0)113 343 1707