AIMCI: AI-Guided Resource Management for Advanced Cyberinfrastructure

Applies predictive AI models, reinforcement learning, and human-in-the-loop simulation to coordinate a facility's heterogeneous computing resources as a unified pool for a broad spectrum of applications.

As computing facility workloads shift beyond traditional simulations, managing increasingly complex, heterogeneous infrastructure with heuristic or manual methods struggles to keep up. AIMCI (Artificial Intelligence for Managing Cyberinfrastructure) combines predictive models of resource usage and user behavior, reinforcement learning for dynamic resource allocation, and a human-in-the-loop simulation tool for testing management strategies, moving a facility from managing individual clusters toward coordinating its full pool of resources for a broad spectrum of applications. As a Minority-Serving Institution, UIC’s education plan for the project also aims to broaden diversity and inclusion in the field.

Team:

  • Students
    • Yash Kurkure [PhD Student]
    • Yiheng Tao [PhD Student with Zhiling Lan]
  • Faculty
    • Zhiling Lan (UIC) — PI
    • Michael E. Papka (UIC) — Co-PI

This work was supported by the National Science Foundation under Grant No. OAC-2402901 (OAC Core: AIMCI: AI-Guided Resource Management for Advanced Cyberinfrastructure). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.