Colloquium - Neely

Pervasive and autonomic systems share many features in common: they are highly adaptive, involve a large and dynamically-changing population of components and services, must deal with a variety of sensors and information sources delivering partial and uncertain results and must deliver an overall experience which is simultaneously adaptive to changing context but stable enough to present a predictable and scrutable service to users and other systems. Building systems which meet these criteria is challenging, involves considerable infrastructural development work, and requires the designer to understand a wide range of quite subtle issues -- all of which interfere with the development of application-level services.

To support developers of pervasive systems we designed and built Construct. Construct differs from other pervasive systems platforms in a number of key respects. It is completely standards-based, using RDF as its data exchange model and ZeroConf for resource discovery. It supports a knowledge-centric model of interaction where clients' actions are driven by queries and triggers about the context of the system. It uses gossiping to maintain a consistent state across a distributed data structure, which maximises robustness and scalability and avoids many problems with hot-spots and hot-paths in communications. Finally, it treats all information sources uniformly as sensors acting as inputs to uncertain reasoning algorithms.

Dr. Steve Neely has been researching the construction and management of autonomous distributed systems since 1998. His current interests include programming models for global computation and architectures for capricious information systems. In particular: pervasive and ubiquitous computing, peer-to-peer networking, autonomics, sensor systems, context-awareness, location tracking, middleware, Web Services, persistence, code/process migration, semistructured data, XML datasets, data fusion and data provenance.

Hosted by Katie Siek.