Overview Tools & solutions Data centers Customer success Resources

AWS Sustainable Infrastructure

Designing, building, and operating more efficient data centers

Sustainable infrastructure for every AI workload

AWS is building data centers to support the next generation of artificial intelligence (AI) innovation and customers’ evolving needs. These capabilities combine innovations in power, cooling, and hardware design to create a more energy efficient data center that will underpin further customer innovation. These new capabilities will be implemented globally in our data centers, and many components are already deployed in our existing data centers.

New data center components to support AI innovation

AWS' latest innovative data center design delivers 12% higher compute power with an expected power usage effectiveness (PUE) rating of 1.08. PUE is one way we measure the efficiency of our data center operations. A lower PUE indicates a more efficient data center and a PUE score of 1.0 is perfect. The enhanced design improves availability and maximizes operational efficiency across our infrastructure.

Accelerate innovation with AWS Processors

Customers can accelerate their sustainability innovation using AWS Silicon chips and leverage them to run their cloud workloads more sustainably. A 2024 Accenture study shows that customers running compute-heavy workloads can see carbon emissions reduce by up to 99%, with reduction of up to 94% by moving from on-premises to AWS, and an additional reduction up to 81% using AWS’s purpose-built silicon.

AWS-designed silicon for efficient AI/ML workloads

AWS Graviton

AWS Graviton-based Amazon EC2 instances use up to 60% less energy than comparable EC2 instances for the same performance. They also provide the best price performance for cloud workloads running on Amazon EC2. Over 70,000 customers have used AWS Graviton to build efficient and performant workloads.

AWS Trainium

Trn2 instances are designed to be three times more energy efficient than Trn1 instances. Trn1 instances are up to 25% more energy efficient than comparable accelerated computing EC2 instances. These instances help you meet your sustainability goals when training ultra-large models.

Optimizing efficiency across our operations

Advanced liquid cooling for AI processors

AWS has developed novel mechanical cooling solutions providing configurable liquid-to-chip cooling in both new and existing data centers. Some AWS technologies utilize network and storage infrastructure that does not require liquid cooling, so updated cooling systems will seamlessly integrate air and liquid cooling capabilities for the most powerful AI chipsets, like AWS Trainium3 and rack-scale AI supercomputing solutions like NVIDIA GB200 NVL72, as well as AWS’s network switches and storage servers. This flexible, multimodal cooling design allows us to provide maximum performance and efficiency at the lowest cost, whether running traditional workloads or AI models.

Support for high-density AI workloads

AWS is maximizing how power is used by optimizing how we position racks in a data center. This was achieved through software, powered by data and generative AI, that predicts the most efficient way to land servers. AWS will now reduce the amount of stranded power – energy that is available but unused or underutilized – and make more efficient use of the energy available.

Reduced energy consumption

AWS continues to innovate to increase efficiency with each generation of our data center designs. The new data center components in AWS data center design are projected to reduce mechanical energy consumption by up to 46% and expected to reach a power usage effectiveness (PUE) rating of 1.08. PUE is one way we measure the efficiency of our data center operations. A lower PUE indicates a more efficient data center and a PUE score of 1.0 is perfect. This rating is below the IDC estimated PUE of 1.84 in 2023 for on-premises data centers.

Circular economy: conserving natural resources and reducing biodiversity

13%

of spare parts, sourced from AWS own reuse inventory in 2024. And we expect this number to grow as we expand our efforts

23.5M

components recycled or sold on the secondary market

30%

of plastic server rack components made of recycled or bio-based plastic

+2yrs

extended lifespan of Amazon hard disk drives

Becoming water positive at AWS

In 2022, we announced our commitment to being water positive by 2030. That means we’ll return more water to communities and the environment than we use in our data center operations. By the end of 2023, AWS was 41% of the way towards water positive.

There are four pillars of our water positive commitment:

Water efficiency
Sustainable sources
Water reuse in communities
Water replenishment

Learn more