We have learned a fair amount today: how to submit vanilla jobs,
standard universe jobs, java universe jobs, and DAGMan jobs. But we
have only scratched the surface. There is so much more to learn. Here
are a few topics to whet your apetite.
- Our jobs relied on the fact that hal shares a filesystem with the
execution computers, but they didn't have to do that. They could have
told Condor to transfer files to the execute computer.
- Condor worries about job priorities and user priorities.
- Condor can run MPI jobs.
- Condor can submit jobs to remote grid sites.