Rapid is software for rapidly creating domain-specific portlet interfaces that allow jobs to be submitted to resources (e.g. Grid resources), including a wide range of features to "give computational science a friendly face".
In this seminar I will demo the results of my work on the FireGrid project since 1st Jan. in which I have been:
1.) Applying the Rapid software to FireGrid project scenarios
2.) Extending the above to include an initial provenance system - so the Rapid system archives all the information required to at a click re-run a particular experiment run for a particular experiment for a particular paper - and to compare the results to the archived original results.
3.) Providing a path for scientists with limited IT literacy to be exposed to the Rapid functionality step by step; from a mode that only provides basic Rapid functionality to Rapid in its entirety
4.) Providing a Rapid-based system that can be used when the FireGrid software is running in Urgent Computing mode. For example when the start time of a real-time job depends on the technical difficulties encountered setting up a real furnace, or when the start time depends on when a real fire starts.