OSG Site Documentation¶
If you are a researcher interested in accessing OSG computational capacity, please consult our user documentation instead.
The OSG Consortium provides common service and support for capacity providers and scientific institutions (i.e., "sites") using a distributed fabric of high throughput computational services. The OSG Consortium does not own computational capacity but provides software and services to users and capacity providers alike to enable the opportunistic usage and sharing of capacity.
This documentation aims to provide HTC/HPC system administrators with the necessary information to contribute computational capacity to the OSG Consortium.
Contributing to the OSG¶
We offer two models for sites to contribute capacity to the OSG Consortium: one where OSG staff hosts and maintains capacity provisioning services for users; and the traditional model where the site hosts and maintains these same services. In both of these cases, the following will be needed:
- An existing compute cluster running on a supported operating system with a supported batch system: Grid Engine, HTCondor, LSF, PBS Pro/Torque, or Slurm.
- Outbound network connectivity from your cluster's worker nodes
- Temporary scratch space on each worker node
Don't meet the requirements?
If your site does not meet the above conditions, please contact us to discuss your options for contributing to the OSG Consortium.
To contribute computational capacity with OSG-hosted services, your site will also need the following:
- Allow SSH access to your local cluster's login host from a known IP address
- Shared home directories on each cluster node
If you are interested in OSG-hosted services, please contact us for a consultation, even if your site does not meet the conditions as outlined above!
If you are interested in contributing capacity by hosting your own OSG services, please continue with the site planning page.