A Theory of Probabilistic Power Provisioning for Data Centers with Distributed Energy Storage
Abstract
The growing power demands and variability of AI workloads make electrical power delivery a critical constraint in data-center operation. Distributed energy storage can reduce the power capacity required to support stochastic loads, but its benefits depend fundamentally on the statistics and time scales of demand. This paper develops a probabilistic framework that jointly characterizes provisioned power, energy-storage capacity, and the probability of overdraw. We show that storage-assisted provisioning separates into two operating regimes. In the Small Battery Region, overdraw is dominated by short-lived demand excursions and storage provides nearly linear reductions in the required power margin. In the Large Battery Region, overdraw results from sustained demand fluctuations over longer spans of time, and the required margin exhibits diminishing returns with storage. For this regime we introduce effective power, an analogue of effective bandwidth that captures the temporal statistics of the demand and gives an asymptotically tight characterization of the required power. We further quantify how temporal correlation and spatial aggregation affect storage requirements and statistical multiplexing gains, and extend the analysis to loads with multiple demand time scales. Finally, we evaluate the framework using power-demand traces from three production data centers spanning HPC, GPU-training, and cloud-service workloads. Despite their heterogeneous, cyclo-stationary and multi-modal behavior, the measured workloads exhibit the predicted regimes, and a simple four-parameter two-state model captures the dynamics governing their storage-power tradeoffs. The resulting framework provides both a probabilistic foundation and practical dimensioning principles for storage-assisted power provisioning in next-generation AI data centers.
Related
- Optimisation of distributed storage in existing district heating networks to solve hydraulic bottlenecks
- Privacy-Preserving Distributed Online Dispatch of Low-Carbon Microgrids with TCN–BiLSTM Probabilistic Photovoltaic Forecasts
- Latency-Optimal Geo-Distributed Storage over Structured Networks
- Communication and Privacy Trade-Offs in Distributed Storage With Shared Secret Keys
- The Coverage Depth Problem in Distributed DNA Data Storage
BibTeX
@misc{koksal2026theory,
title = {{A Theory of Probabilistic Power Provisioning for Data Centers with Distributed Energy Storage}},
author = {Can Emre Koksal and Richard A. Barry and Artun Sel},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2608.12993},
eprint = {2608.12993},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.12993},
}
RIS
TY - UNPB
TI - A Theory of Probabilistic Power Provisioning for Data Centers with Distributed Energy Storage
AU - Can Emre Koksal
AU - Richard A. Barry
AU - Artun Sel
PY - 2026
DA - 2026/08/13
PB - arXiv
DO - 10.48550/arxiv.2608.12993
UR - https://arxiv.org/abs/2608.12993
AB - The growing power demands and variability of AI workloads make electrical power delivery a critical constraint in data-center operation. Distributed energy storage can reduce the power capacity required to support stochastic loads, but its benefits depend fundamentally on the statistics and time scales of demand. This paper develops a probabilistic framework that jointly characterizes provisioned power, energy-storage capacity, and the probability of overdraw. We show that storage-assisted provisioning separates into two operating regimes. In the Small Battery Region, overdraw is dominated by short-lived demand excursions and storage provides nearly linear reductions in the required power margin. In the Large Battery Region, overdraw results from sustained demand fluctuations over longer spans of time, and the required margin exhibits diminishing returns with storage. For this regime we introduce effective power, an analogue of effective bandwidth that captures the temporal statistics of the demand and gives an asymptotically tight characterization of the required power. We further quantify how temporal correlation and spatial aggregation affect storage requirements and statistical multiplexing gains, and extend the analysis to loads with multiple demand time scales. Finally, we evaluate the framework using power-demand traces from three production data centers spanning HPC, GPU-training, and cloud-service workloads. Despite their heterogeneous, cyclo-stationary and multi-modal behavior, the measured workloads exhibit the predicted regimes, and a simple four-parameter two-state model captures the dynamics governing their storage-power tradeoffs. The resulting framework provides both a probabilistic foundation and practical dimensioning principles for storage-assisted power provisioning in next-generation AI data centers.
ER -