The EPIC Lab advances sustainable datacenter and robust high‑performance computing (HPC) research by developing intelligent, resource‑aware infrastructures that reduce environmental impact while enhancing system reliability and performance. Lab researchers design AI‑driven management techniques that optimize energy consumption, carbon footprint, and water usage across large‑scale computing platforms, all while ensuring fault‑tolerant, highly available operation. This work integrates real‑time analytics, predictive modeling, and adaptive control into datacenter and HPC resource management frameworks, enabling infrastructures that dynamically respond to workload variability and system‑level failures. Through these innovations, the EPIC Lab is shaping next‑generation computing systems that are not only faster and more efficient, but also environmentally responsible and resilient under demanding operational conditions.

Ongoing Cloud Computing and High Performance Computing Projects:

• Energy Efficient and Robust Resource Management for Datacenters
• Fault Resilient Exascale Supercomputing