Globus Alliance

Globus Alliance
Name	Globus Alliance
Type	Research consortium
Founded	2001
Headquarters	University of Chicago
Fields	High-performance computing, Data management, Cyberinfrastructure

Globus Alliance is a collaborative consortium that developed middleware and services for secure, reliable, and high-performance research data management across distributed computing resources. The Alliance combined expertise from national laboratories, universities, and commercial partners to build production-grade tools used in science and engineering domains, supporting data transfer, identity management, and metadata services for large-scale projects. Its efforts influenced cyberinfrastructure practices at major facilities and projects, integrating with grid, cloud, and supercomputing ecosystems.

History

Founded in the early 2000s by researchers associated with Argonne National Laboratory, University of Chicago, and California Institute of Technology, the Alliance emerged amid efforts connected to National Science Foundation cyberinfrastructure initiatives and the TeraGrid project. Early contributors included teams from Fermilab, Lawrence Berkeley National Laboratory, and Oak Ridge National Laboratory, collaborating on middleware prototypes inspired by the Globus Toolkit and concepts from Andrew W. Appel-era distributed systems research. The Alliance expanded through partnerships with European Grid Infrastructure, Enabling Grids for E-sciencE (EGEE), and project consortia such as Open Grid Forum and Renaissance Technologies-adjacent research groups. Major milestones included deployments for Large Hadron Collider collaborations, integrations with XSEDE resources, and support for data-intensive experiments like National Center for Supercomputing Applications initiatives and Human Genome Project-style sequencing pipelines.

Architecture and Services

The Alliance designed a layered architecture combining secure authentication, reliable data movement, and metadata management that interoperated with systems like Amazon Web Services, Google Cloud Platform, and on-premises clusters at Los Alamos National Laboratory. Core design patterns drew on standards from Simple Object Access Protocol-era web services and later RESTful approaches used by National Institutes of Health data platforms. Services included identity federation compatible with InCommon and Shibboleth, delegation models akin to those used by OAuth implementers, and transfer protocols interoperable with GridFTP deployments at facilities such as Princeton Plasma Physics Laboratory. The architecture emphasized scalability for use with petascale resources like Oak Ridge Leadership Computing Facility and integration with archival systems such as National Archives and Records Administration technologies.

Software and Tools

Key software developed or maintained by Alliance participants encompassed data transfer utilities, catalog services, and access control components used alongside tools from Apache Software Foundation projects and CERN-derived software stacks. Notable components interoperated with Globus Toolkit-derived libraries and implemented features complementary to Apache Hadoop and HTCondor job management. The software suite supported provenance capture consistent with models from PROV-DM communities and metadata schemas employed by Dublin Core and domain repositories like PANGAEA. Developers collaborated with teams behind Jupyter and GitHub-hosted research workflows, and integrated with identity systems such as ORCID and CILogon.

Use Cases and Applications

The Alliance’s technologies powered a range of scientific applications, including data movement for collaborations at CERN experiments, multi-institutional genomics workflows tied to National Human Genome Research Institute, earth science data distribution for NASA missions, and climate modeling efforts linked with Intergovernmental Panel on Climate Change data centers. Other deployments included large-scale microscopy repositories associated with National Institutes of Health centers, seismology data sharing coordinated with United States Geological Survey, and social science data curation for projects at Harvard University and Stanford University. The platform also supported cross-institution workflows for astronomy surveys associated with National Optical Astronomy Observatory and computational chemistry simulations used by teams at MIT and Caltech.

Governance and Funding

Governance structures incorporated university principal investigators, representatives from national laboratories such as Brookhaven National Laboratory and Sandia National Laboratories, and program officers from National Science Foundation directorates. Funding sources included grants from National Institutes of Health, cooperative agreements with Department of Energy, and contributions from academic partners like University of Illinois Urbana-Champaign and University of Wisconsin–Madison. Steering committees coordinated with standards bodies including Internet Engineering Task Force and Open Grid Forum while engaging with commercial stakeholders from firms like IBM and Microsoft Research.

Community and Adoption

Adoption grew through collaborations with initiatives such as XSEDE, Extreme Science and Engineering Discovery Environment (XSEDE), and international projects like PRACE and ELIXIR. Training and outreach occurred at conferences including SC Conference, International Conference for High Performance Computing, Networking, Storage and Analysis, and workshops hosted by Society for Industrial and Applied Mathematics. The user community comprised researchers from institutions such as Stanford University, University of California, Berkeley, Princeton University, and national laboratories, with ecosystem integrations facilitated via partnerships with DataONE and domain repositories like Dryad (repository). Continued adoption relied on interoperability with emerging platforms from Google Research and Amazon Web Services research programs, and on standards engagement through organizations like World Wide Web Consortium.

Category:Cyberinfrastructure Category:Research projects