In the exascale era, researchers are confronted with both enormous amounts of data and the complexities of data-intensive research spanning institutions. Data management across collaborations, and automation of data processing, including by bridging to various HPC systems, is needed. So too are intuitive interfaces for researchers to interact with such systems to ensure resources are well utilized.
Globus (www.globus.org), a research data management platform from the University of Chicago, can address many of these complexities, allowing research teams with web applications or science gateways to integrate with the platform seamlessly. In this session, we will highlight a joint effort between Globus and Argonne Leadership Computing Facility called Petrel, that provides a system for discovery and access of large scale data, with an intuitive portal interface. We’ll discuss the Globus platform capabilities for data management, and integration with data processing pipelines, that are used in Petrel.