Speaker
Describe the added value of the Grid for the scientific/technical activity you (plan to) do on the Grid. This should include the scale of the activity and of the potential user community and the relevance for other scientific or business applications
MOTEUR is interpreting and enacting the Scufl workflow language developed in the context of the Taverna project. Scufl is a pure data flow language which eases the description of data-parallel applications in a workflow framework. It provides a large flexibility in data composition through the use of iteration strategies. MOTEUR defines a strict semantics for this data composition, relying on the implicit or explicit user input describing the connections between data fragments. While preserving this semantics, MOTEUR provides completely asynchronous parallelization of the workflow tasks and data segments to be processed, thus optimizing the degree of parallelism that can be achieved. It is also providing a legacy code wrapper enabling the execution of non-instrumented codes in a service-oriented framework, interfaced with the grid infrastructure.
Describe the scientific/technical community and the scientific/technical activity using (planning to use) the EGEE infrastructure. A high-level description is needed (neither a detailed specialist report nor a list of references).
The MOTEUR workflow manager is a generic workflow enactment service for grids. It targets a wide user community. It was designed inside the biomed community to fulfill the needs for efficient enactment of data-intensive workflow-based applications on the EGEE grid infrastructure, although it is not limited to this application area. MOTEUR is interfaced to the gLite workload manager and the code is open source.
Report on the experience (or the proposed activity). It would be very important to mention key services which are essential for the success of your activity on the EGEE infrastructure.
MOTEUR has been successfully used to efficiently enact data parallel applications to medical images processing. It has been exploited in production for more than one year and it has been evolving to include various execution time optimization strategies. The workflow engines shields the end user from the details of the application parallel execution and from the grid interface. It ensures consistent parallel execution and interface to the gLite middleware through its command line interface. Large scale experiments, involving thousands of medical image processings have been conducted with very good execution time as compared to other comparable workflow managers. Extensions to larger scale applications to test the scale-up capability of MOTEUR are planed in a near future.
