5 papers · 1 filter
DCEmbed: Scalable Optimization over Neural Surrogates
Akshay Sreekumar, Nicolas Christianson, Priya L. Donti +2
Neural surrogates can accelerate large-scale optimization by replacing expensive or intractable model components with efficient learned approximations, but solving the resulting em…
Multi-Scale Datacenter Power Modulation
Akshay Sreekumar, Nicolas Christianson, Fiodar Kazhamiaka +1
Cloud datacenters must increasingly modulate power in response to time-varying grid and infrastructure constraints. We study this problem as finite-horizon control of a networked h…
Optimizing Energy Efficiency and Grid Stability via Public EV Charging Flexibility
Marek Miltner, Artem Bryksa, Ondřej Štogl +4
This study evaluates the potential of electric vehicle (EV) charging flexibility to enhance both energy efficiency and power grid stability. Using real-world data from public charg…
EasyRider: Mitigating Power Transients in Datacenter-Scale Training Workloads
Dillon Jensen, Obi Nnorom, Grant Wilkins +4
Large-scale AI model training workloads use thousands of GPUs operating in tightly synchronized loops. During synchronous communication, start-up, shut-down, and checkpointing, GPU…
From Servers to Sites: Compositional Power Trace Generation of LLM Inference for Infrastructure Planning
Grant Wilkins, Fiodar Kazhamiaka, Ram Rajagopal
Datacenter operators and electrical utilities rely on power traces at different spatiotemporal scales. Operators use fine-grained traces for provisioning, facility management, and…