header image

Research Interests

Projects

CONNECT (Feb 2009 to September 2012)

The CONNECT Integrated Project aims at enabling continuous composition of networked systems to respond to the evolution of functionalities provided to and required from the networked environment. At present the efficacy of integrating and composing networked systems depends on the level of interoperability of the systems’s underlying technologies. However, interoperable middleware cannot cover the ever growing heterogeneity dimensions of the networked environment. CONNECT aims at dropping the interoperability barrier by adopting a revolutionary approach to the seamless networking of digital systems, that is, synthesizing on the fly the connectors via which networked systems communicate. The resulting emergent connectors are effectively synthesized according to the behavioral semantics of application- down to middleware-layer protocols run by the interacting parties. The synthesis process is based on a formal foundation for connectors, which allows learning, reasoning about and adapting the interaction behavior of networked systems at run-time. Synthesized connectors are concrete emergent system entities that are dependable, unobtrusive, and evolvable, while not compromising the quality of software applications. To reach these objectives the CONNECT project undertakes interdisciplinary research in the areas of behavior learning, formal methods, semantic services, software engineering, dependability, and middleware. Specifically, CONNECT will investigate the following issues and related challenges: (i) Modeling and reasoning about peer system functionalities, (ii) Modeling and reasoning about connector behaviors, (iii) Runtime synthesis of connectors, (iv) Learning connector behaviors,(v) Dependability assurance, and (vi) System architecture. The effectiveness of CONNECT research will be assessed by experimenting in the field of wide area, highly heterogeneous systems where today's solutions to interoperability already fall short (e.g., systems of systems).

 

Aspect Oriented Reflective Middleware (AORM)(May 2007 - December 2007)

This project was carried out at K. U. Leuven and investigated how aspects and reflection can be combined to better support: i) the development of adaptive middleware, and ii) the development of highly complex, dynamic applications.

 

Open Overlays (2004-2007)

The Open Overlays project was a collaboration between Lancaster University and Oxford Brookes University. The project explored network and middleware support for distributed collaborative visualisation in the Grid.

The aims of the project were to provide integrated middleware and networking support for future Grid applications with the following properties: highly heterogeneous; highly collaborative; employ multiple media-types; are large scale, complex and dynamic; and are QoS-sensitive (i.e., need to react to changes in environmental conditions).

A key result of the project was the Gridkit middleware platform; this leveraged the novel concept of application-level virtual networks as a core element of modern middleware, underpinning a range of middleware services.

 

ReMMoC (2000-2003)

The ReMMoC (Reflective Middleware for Mobile Computing) project was a collaboration between Lancaster University and Bell Labs UK, Lucent Technologies. The project examined the use of reflection and component technology to overcome the problems of heterogeneous middleware technology in the mobile environment.

Mobile applications and services are implemented upon a range of middleware platforms (e.g., RPC, message-oriented and event-based paradigms) and are advertised using different service discovery protocols. Therefore, developing mobile applications upon fixed middleware types is unsuitable; rather the middleware on the mobile device should be able to dynamically change its structure and behaviour so that it can discover and interoperate with all services available in the current environment.

The project produced a reflective middleware framework, named ReMMoC, that supports the capability to develop applications independently of middleware implementation. The framework itself combines the Lancaster approach of components, component frameworks and reflection with a Web Services based abstraction.

 

OpenCOM (Ongoing)

OpenCOM is an open, reflective component development platform that supports the creation of configurable and reconfigurable middleware. Originally, the platform was based upon the concepts of Microsoft's COM; however, we are evolving OpenCOM v2 as a platform independent component solution. OpenCOM has already been utilised to build many system platforms on desktops, portables and network routers.

 

Interests

Currently, I am interested in the following areas of research:
  • Systems support for mobile applications. In particular, adaptable systems that dynamically reconfigure to mirror device environments.
  • Service Discovery.
  • Web Services. How the web service architecture can be used to achieve improved systems interoperability.
  • Reflective Systems. The Application of reflection to produce configurable and reconfigurable systems software. Furthermore, the investigation of high-performance reflective systems.
  • Grid Computing. Next generation Grid Middleware to underpin complex e-Science scenarios.
  • Overlay networks and Peer-to-Peer systems. How such networks can support richer middleware binding semantics.
  • Reflective component devlepment platforms.
  • Aspect Oriented Programming and Dynamic Aspects.