Next-generation data centers require immense, uninterrupted baseline power. Current models suggest gigawatt-scale facilities will soon become the standard, drawing more energy than small cities.
The target cluster consists of three specialized AI training facilities. Unlike traditional data centers that have predictable day/night load curves, these facilities run 100% utilization training runs lasting 90-120 days. They cannot be easily throttled without losing millions in compute efficiency.
| Facility | Type | MW Load (Continuous) | Water Usage (Gal/Day) |
|---|---|---|---|
| Cluster Alpha (LLM Training) | High-Density Liquid Cooled | 650 MW | 1.2M |
| Cluster Beta (Inference) | Air Cooled | 400 MW | 0.5M |
| Cluster Gamma (Multimodal) | Hybrid Cooled | 450 MW | 0.8M |
| TOTAL LOAD | - | 1.5 GW (1,500 MW) | 2.5M |
Note: 1.5 GW is roughly equivalent to the power consumption of 1.2 million residential homes. The municipal grid currently has a total baseline capacity of 4.2 GW with a peak load ceiling of 3.8 GW during summer months. The introduction of the AI cluster pushes the grid into a severe deficit.