Interaction between data-center flexibility products and on-site resources

Quantify how data-center flexibility products designed for individual hyperscale data centers interact with on-site resources, including co-located generation and storage.

Background

The paper models data-center flexibility at the grid interface through firm, flexible, and interruptible load tiers, without representing the internal scheduling decisions of individual data centers or explicitly coupling those tiers to on-site generation and storage. The authors identify direct design of flexibility products for hyperscale data centers as a natural planning and regulatory lever because individual facilities are large enough to influence grid-wide generation, storage, and transmission investments.

The unresolved issue is how such externally procured or designed flexibility products would interact with resources located at the data center itself. Quantifying this interaction would help determine whether workload shifting, curtailment, and on-site supply are complementary or substitutable, and would support more realistic valuation and design of large-load flexibility products.