DOE OSTI · 3012591
Machine learning-driven predictive resource management in complex science workflows
Abstract
Here, the collaborative efforts of large communities in science experiments, often comprising thousands of global members, reflect a monumental commitment to exploration and discovery. Recently, advanced and complex data processing has gained increasing importance in science experiments. Data processing workflows typically consist of multiple intricate steps, and the precise specification of resource requirements is crucial for each step to allocate optimal resources for effective processing. Estimating resource requirements in advance is challenging due to a wide range of analysis scenarios, varying skill levels among community members, and the continuously increasing spectrum of computing options. One practical approach to mitigate these challenges involves initially processing a subset of each step to measure precise resource utilization from actual processing profiles before completing the entire step. While this two-staged approach enables processing on optimal resources for most of the workflow, it has drawbacks such as initial inaccuracies leading to potential failures and suboptimal resource usage, along with overhead from waiting for initial processing completion, which is critical for fast-turnaround analyses. In this context, our study introduces a novel pipeline of machine learning models within a comprehensive workflow management system, the Production and Distributed Analysis (PanDA) system. These models employ advanced machine learning techniques to predict key resource requirements, overcoming challenges posed by limited upfront knowledge of characteristics at each step. Accurate forecasts of resource requirements enable informed and proactive decision-making in workflow management, enhancing the efficiency of handling diverse, complex workflows across heterogeneous resources.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Chowdhury, Tasnuva [Brookhaven National Laboratory (BNL), Upton, NY (United States)] (ORCID:0000000226818105), Maeno, Tadashi [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Akman, Fatih Furkan [Univ. of Massachusetts, Amherst, MA (United States)], Boudreau, Joseph [Univ. of Pittsburgh, PA (United States)], Dutta, Sankha [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Feng, Shengyu [Carnegie Mellon Univ., Pittsburgh, PA (United States)], Hoisie, Adolfy [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Hsu, Kuan-Chieh [Brookhaven National Laboratory, Upton, NY, USA], Khan, Raees [Univ. of Pittsburgh, PA (United States)], Kim, Jaehyung [Carnegie Mellon Univ., Pittsburgh, PA (United States)], Kilic, Ozgur O. [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Klasky, Scott [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)], Klimentov, Alexei [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Korchuganova, Tatiana [Univ. of Pittsburgh, PA (United States)], Outschoorn, Verena Ingrid Martinez [Univ. of Massachusetts, Amherst, MA (United States)], Nilsson, Paul [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Park, David K. [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Podhorszki, Norbert [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)], Ren, Yihui [Carnegie Mellon Univ., Pittsburgh, PA (United States)], Steele, John Rembrandt [Univ. of Massachusetts, Amherst, MA (United States)], Suter, Frédéric [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)], Vatsavai, Sairam Sri [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Wenaus, Torre [Brookhaven National Laboratory (BNL), Upton, NY (United States)], Yang, Wei [SLAC National Accelerator Laboratory (SLAC), Menlo Park, CA (United States)], Yang, Yiming [Carnegie Mellon Univ., Pittsburgh, PA (United States)], Yoo, Shinjae [Brookhaven National Laboratory (BNL), Upton, NY (United States)]. 2025-12-24. Machine learning-driven predictive resource management in complex science workflows. https://doi.org/10.1142/s0217751x26500259
Cite the original work for its findings. Save a collection to share your selection of sources.