Provenance in Web Services and Grid Computing

The provenance of a resource has been defined as "the relationship between records and the organisations or individuals that created, accumulated and/or maintained and used them in the conduct of personal or corporate activity." For digital resources, provenance information includes the person who produced the data, how and when it was produced, what tool produced it subject to what parameters etc. Prof. Luc Moreau of ECS is a partner in the EU Provenance project.

In significant distributed computing projects, such as are supported by a Grid, or by Web Services, many data and computational resources are produced and consumed. For example, a significant web service-supported set of chemical experiments might involve input and output data from experiments and lab equipment, identification and skill level of personnel involved, dates and times of activities, the sequence in which activities were performed. There are many circumstances where there is a need to validate this information trail (or provenance) of the project: a regulatory authority might require an audit trail, or it might be necessary to repeat the experiments.

This research theme is speculative and is concerned with investigating how we might validate and verify provenance in applications, and how such V&V activities might be tool-supported.

Updated 24/2/05