User: Glatard

I am using Taverna to design workflows running on the EGEE grid. EGEE jobs are wrapped into Web-Services and can then be  embedded in a Taverna workflow. During my PhD, I studied workflows for medical image analysis applications and we really enjoyed Scufl’s iteration strategies that allow to define complex data patterns with simple operators. Because of performance limitations of Taverna (in particular concerning pipelining among the services), we developped MOTEUR, a Scufl enactor dedicated to the execution of Taverna workflows on EGEE. We also faced some limitations of the semantics of the ‘dot product’ and we extended it.

I am currently working at the University of Amsterdam in Silvia Olabarriaga’s group. We are implementing the "feat" fMRI application from the FSL package in Scufl. We set up an interface between the VBrowser data management tool and the MOTEUR engine to be able to  execute workflows processing files located in any grid storage (gridftp, srb, EGEE storage elements).