Helsinki University, Main Building - Room 13
Galaxy jobs processed on the grid: what about the cloud?
As part of the labex CEBA (Center for the study of biodiversity in Amazonia), we set up a Galaxy based e-VirtualBiodiversityLab to make easy the use of biodiversity analysis tools to researchers. The increase of data production as a consequence of the progress in sequencing technologies (NGS) requires high computing resources like clusters, the EGI grid and the cloud. We started to write tools for Galaxy to distribute the processing of such big NGS files (phylogeny, global alignment, taxonomic annotation, ...) to the Avakas cluster (Mesocentre Aquitaine, 1000 cores, Torque) and to the EGI grid (France-Grilles) using the DIRAC interware. This is completely transparent for the user of Galaxy who doesn't need to have an expertise on such technologies and can focus on analyzing and interpreting the results. Software requirements (OS,version, dependencies, ...) could be a problem when using the grid. We propose for the hackaton to tackle this problem by distributing jobs on cloud resources from a Galaxy platform. This could also be a way to run jobs with parallelized softwares (MPI, ...) easily. The hackaton challenge would be to write a Galaxy tool to launch jobs on virtual machine (VM) instances on Academic cloud.