Computing with Globus

The Globus compute platform delivers the same “fire-and-forget” capabilities for computation as our core platform does for data management.

Globus offers a distributed Function as a Service (FaaS) platform that enables reliable, scalable, and high performance remote function execution. Unlike centralized FaaS platforms, the Globus compute platform enables you to execute functions on diverse remote systems, from laptops to campus clusters, clouds, and supercomputers.

Globus Compute showing different remote systems

Why Globus Compute?

Remote computing is notoriously complicated—the Globus compute platform takes care of complex infrastructure management tasks so you can focus on running the code you want, when and where you want. Using Globus means you no longer waste time…

  • …figuring out credentials and different authentication mechanisms
  • …configuring and managing batch jobs and schedulers
  • …interacting with resource managers, waiting in queues and scaling nodes
  • …configuring the execution environment for different compute systems
  • …retrieving and sharing computation results

Once your code is ready, the Globus compute platform allows you to run it anywhere you have access to compute resources—and move it to other systems, without having to overcome the same infrastructure management obstacles.

The Globus compute platform incorporates work from the University of Chicago and the University of Illinois Urbana-Champaign, supported by funding from the National Science Foundation and an Argonne National Laboratory Directed Research and Development award.