Kyle Chard

Kyle Chard

Globus as a Case Study in Sustainable Research Software Development

Despite decades of work, the idea of sustainable research computing software in most cases remains an aspirational endeavor. Software is not a static artifact—it requires continuous maintenance in the form of bug fixes, security patches, and adaptation to shifting requirements, as well as strategic investments to develop new features to keep pace with evolving user needs. Globus, a department at the University Chicago, delivers a widely-used research IT platform that provides critical capabilities to manage data, compute, and automated workflows across research cyberinfrastructure. In addition to delivering impactful capabilities to researchers, sustainability has been a core goal for the Globus team. Over the last two decades, the team has considered different sustainability models, experimenting with open source models, commercial ventures, and consortia. Based on these experiences, Globus has pioneered a hybrid cloud-based model that reduces costs and exploits economies of scale as well as a “freemium” model that supports operational costs through tiered institutional subscriptions uniquely designed for the diverse research computing ecosystem. With the support of an active and engaged community and more than 250 subscribers, Globus is able to provide the highly available, secure, and performant capabilities necessary to support research computing at a global scale—more than 600,000 researchers and applications using 60,000 data collections across more than 80 countries. Despite its success, sustainability remains an ongoing endeavor, and Globus continues to adapt to shifting market dynamics to fulfill its mission of delivering sustainable research software.

Dr. Kyle Chard is a Research Associate Professor in the Department of Computer Science at the University of Chicago and a joint appointee at Argonne National Laboratory. He co-leads the Globus Labs research group, where he investigates foundational and applied problems in distributed computing, data-intensive computing, and scientific cyberinfrastructure. He leads the development of tools like Parsl and Globus Compute, which support secure and scalable science across diverse computing environments.