Airsys Pushed for New Data Center Power Efficiency Metrics
The firm advocates for Power Compute Effectiveness standards to address massive energy waste in cooling systems.
Updated on Sept. 22, 2026 in Data Centers

Live Poll
Should the government mandate energy efficiency standards for the expansion of data centers?
Airsys has lobbied Congress to mandate Power Compute Effectiveness reporting within the Liquid Cooling for AI Act. The company argues that legacy cooling methods waste 22MW of power for every 100MW site, threatening data center expansion.
Why it matters
Rising rack densities driven by AI hardware are creating a performance discontinuity that threatens to strain grid capacity. By adopting standardized metrics, the industry aims to manage these constraints as demand, such as in Texas, surges.
Legacy cooling systems waste 22MW per 100MW site, whereas switching to a two-loop architecture allows for a 30 percent increase in compute capacity. NVIDIA's upcoming Rubin Ultra chips will require up to 600kW per rack, far exceeding current norms.
The players
Airsys
Airsys is a firm specializing in cooling and power management technologies for high-performance computing environments.
NVIDIA
NVIDIA is a global technology company that designs the specialized hardware and processors driving the current surge in AI computing density.
Uptime Institute
The Uptime Institute is a research and advisory organization that provides industry standards for data center uptime and efficiency.
The details
Airsys claims that existing data center designs are incompatible with the rapid escalation in hardware power requirements, which are outpacing traditional cooling infrastructure. Transitioning to advanced architectures could mitigate the potential 125 percent increase in Texas grid peak load.
Timeline
November 2025: The Liquid Cooling for AI Act was introduced.
April 2026: The Energy Subcommittee held a hearing on the act.
September 2026: Airsys officials met with congressional offices.
The Tech Race
This effort to standardize power efficiency aligns with the broader push to update infrastructure for the Liquid Cooling for AI Act. It represents a shift from legacy cooling methods toward specialized architectures capable of supporting the massive heat loads generated by modern AI.
Standardizing these efficiency metrics could influence how quickly new AI-focused facilities are approved and integrated into local power grids. Consumers and businesses may experience more reliable digital service availability as providers optimize their energy usage to meet regulatory demands.
The takeaway
The rapid move toward high-density AI hardware is forcing a reevaluation of cooling and power standards. Industry participants should monitor upcoming legislative sessions for potential requirements that could change standard build configurations for future data centers.
Further reading
Learn more about the latest innovations in Data Centers and their impact on global infrastructure.
Source note: This article includes information reported by Capacity.
Live Poll
Should the government mandate energy efficiency standards for the expansion of data centers?






