An endpoint UUID is needed to submit work to an endpoint. Endpoint UUIDs can be configured in a YAML, or they can be hard-coded in a workflow program. However, it is probably most useful for users to think in terms of machines, not endpoint UUIDs. Therefore a utility that uses the Globus Compute SDK to query available endpoints and extract UUIDs associated with names supplied by the user is needed so that UUIDs never need to be stored or shared.
An endpoint UUID is needed to submit work to an endpoint. Endpoint UUIDs can be configured in a YAML, or they can be hard-coded in a workflow program. However, it is probably most useful for users to think in terms of machines, not endpoint UUIDs. Therefore a utility that uses the Globus Compute SDK to query available endpoints and extract UUIDs associated with names supplied by the user is needed so that UUIDs never need to be stored or shared.