New Plan to Power Data-Hungry AI: Sink Server Farms into the San Francisco Bay
NetworkOcean proposes submerging server farms in the San Francisco Bay to cut cooling costs for AI by 25%, but faces environmental concerns and regulatory scrutiny regarding permits and wildlife impact.
Read original article
A Bay Area startup, NetworkOcean, is proposing to address the energy demands of generative AI by submerging server farms in the San Francisco Bay. This innovative approach aims to reduce cooling costs, which currently consume significant resources, by utilizing the water for temperature regulation. The company claims that this method could lower operational costs for AI firms by 25%. However, concerns have been raised regarding the environmental impact of such underwater data centers, including potential disturbances to local wildlife and the risk of triggering toxic algae blooms. Regulatory agencies have questioned whether NetworkOcean has obtained the necessary permits for its planned test, which the co-founder asserts will take place in a privately owned area of the bay. Critics argue that while the idea of reducing water usage is appealing, the implications of adding numerous server farms to an already polluted ocean environment could be detrimental.
- NetworkOcean plans to sink server farms in the San Francisco Bay to reduce cooling costs for AI.
- The startup claims this method could lower operational costs by 25%.
- Environmental concerns include potential harm to local wildlife and toxic algae blooms.
- Regulatory agencies are questioning the company's compliance with necessary permits.
- Critics highlight the risks of further polluting the ocean with additional server farms.
Related
Microsoft shelves its underwater data center
Microsoft has ended its underwater data center experiment, noting improved server longevity underwater. Despite success, Microsoft shifts focus to other projects like AI supercomputers and nuclear ambitions, discontinuing further underwater endeavors.
Microsoft waves goodbye to underwater data centers
Microsoft has concluded its underwater data center experiment under Project Natick, showing lower failure rates than land-based centers. Despite discontinuing in 2024, Microsoft plans to apply findings to enhance data center reliability and sustainability amid increasing global data center demand driven by AI, cloud computing, and smart devices.
The Uneven Distribution of AI's Environmental Impacts
The article discusses AI's environmental impacts, emphasizing high energy and water use during model training. It highlights freshwater evaporation, uneven impact distribution, and suggests managing data centers responsibly to address environmental inequality.
Electricity grids creak as AI demands soar
The demand for generative AI is straining electricity grids, with data centers projected to double their consumption by 2026, raising concerns about energy efficiency and sustainability in various regions.
Aging US grid can't handle power and water requirements of generative AI
The rise of generative AI is increasing electricity and water demand, straining the U.S. power grid. Data centers' energy consumption may reach 16% of total electricity by 2030, prompting sustainability efforts.
3 commentsRelated
Microsoft shelves its underwater data center
Microsoft has ended its underwater data center experiment, noting improved server longevity underwater. Despite success, Microsoft shifts focus to other projects like AI supercomputers and nuclear ambitions, discontinuing further underwater endeavors.
Microsoft waves goodbye to underwater data centers
Microsoft has concluded its underwater data center experiment under Project Natick, showing lower failure rates than land-based centers. Despite discontinuing in 2024, Microsoft plans to apply findings to enhance data center reliability and sustainability amid increasing global data center demand driven by AI, cloud computing, and smart devices.
The Uneven Distribution of AI's Environmental Impacts
The article discusses AI's environmental impacts, emphasizing high energy and water use during model training. It highlights freshwater evaporation, uneven impact distribution, and suggests managing data centers responsibly to address environmental inequality.
Electricity grids creak as AI demands soar
The demand for generative AI is straining electricity grids, with data centers projected to double their consumption by 2026, raising concerns about energy efficiency and sustainability in various regions.
Aging US grid can't handle power and water requirements of generative AI
The rise of generative AI is increasing electricity and water demand, straining the U.S. power grid. Data centers' energy consumption may reach 16% of total electricity by 2030, prompting sustainability efforts.