This week Marc presented on his experience deploying Harvester as part of PanDA integration with NSF supercomputers (see Topical presentation above).