PCHES Researchers win a conference award for innovative workflow management research

A team of researchers from Penn State has been recognized for their innovative work in computational modeling with an award at the PEARC Closing Session in Providence, RI on July 25, 2024. Jeffrey Nucciarone and Joseph Perla were awarded the Best Abstract - Student for their submission “Novel Approach Using MPI to Manage a Coupled Workflow.”

The annual Practice and Experience in Advanced Research Computing (PEARC) Conference Series, organized by the Association of Computing Machinery (ACM), is a leading forum for discussing advancements in cyberinfrastructure and research computing and this year was centered around developments in Human Powered Computing (HPC). The award highlights the team’s significant contribution to the field through their inventive use of the Message Passing Interface (MPI) to manage complex coupled workflows through efficient job scheduling.

By leveraging the MPI as an overarching job workload manager, the team developed a robust and scalable workflow that enabled them to effectively couple various models of different spatial and temporal resolutions, including models of regional networks of watersheds and basins, the electric-power grid, and the regional economy. Research of these coupled systems requires multiple instances of multiple job runs. The MPI workflow made it possible to automate the intricate process of running multiple scenarios with these coupled models, which would have been nearly impossible to manage manually. This novel approach may be applicable to other coupled workflows.

The research, conducted as part of the Program for Coupled Human and Earth Systems (PCHES), focused on developing a state-of-the-art multi-sector dynamic framework to model climate change-induced resilience through understanding the interactions between climate change, water resources, and energy systems. This original MPI-based approach has proven essential in studies such as Webster et al. (2022), which examined the impact of exceeding water temperature thresholds on power generation in the western United States. It has also helped to identify vulnerabilities in the existing power system that could result from changes in supply and demand due to climate change (Perla et al., 2024, working paper).

Jeffrey Nucciarone, a Research Software Engineer at Penn State’s Institute for Computational and Data Sciences, provided the critical computational consulting and software engineering support necessary to bring this research to fruition. “The success of this work highlights the value of interdisciplinary collaboration in advancing computational solutions that can address the complexities of modern research challenges,” he noted. This research project, initiated during Joseph Perla’s time as a PhD candidate and completed during his postdoctoral appointment at Penn State, played an important role in his academic journey and career launch. Dr. Joseph Perla is now an Economist at the National Center for Environmental Economics at the United States Environmental Protection Agency.