Research Memos
These documents are formative research memos, working papers, and technical notes reflecting my ongoing work not yet submitted for formal peer review. They are intended to crystallize my early theoretical and empirical thinking and to catalyze rigorous discussion with peers. I welcome constructive comments, suggestions, and inquiries regarding potential research collaborations or joint publications by email.
[1] Andrew Sundstrom. “Some Concerns on the Prospect of Adding a Fleet of Micro Data Centers to Existing Grid Infrastructure”.
15 May 2026 → 20 Oct 2026. Zenodo.
Abstract: The emerging strategy of co-locating micro data centers with transmission substations is widely marketed as an elegant solution to AI power bottlenecks. However, this architecture couples two distinct optimization systems—algorithmic compute dispatch and transmission grid stability—without a unified control layer. Because data center workloads can shift in milliseconds while utility secondary frequency control operates on 4-second cycles, localized perturbations can trigger destabilizing power oscillations and synchronized algorithmic herd behavior