Amazon Web Services, Inc. (AWS), an Amazon.com, Inc. company, has announced findings from the Carbon Reduction Opportunity of Moving to the Cloud for APAC report by 451 Research, part of S&P Global Market Intelligence.
The report found that, on average, organizations moving their business applications from on-premises data centers to cloud infrastructure in APAC can expect to reduce their energy use—and associated carbon footprint—by 78%.
Furthermore, if cloud operations in APAC could be powered by 100% renewable energy, moving compute workloads to the cloud would further boost emissions savings up to 93% for average APAC companies in a region where procuring renewable energy remains challenging. Commissioned by AWS, the study surveyed over 500 companies across APAC, spanning a variety of industries across Australia, India, Japan, Singapore, and South Korea.
“Customers in APAC who move compute workloads to the AWS Cloud can significantly reduce their carbon footprint, benefiting from the net effect of all our sustainability efforts,” said Ken Haig, Head of Energy Policy for Asia Pacific & Japan at AWS.
“Our scale and focus on innovation allow us to improve the energy efficiency of our data center operations faster than traditional on-premises infrastructure. Apart from maximizing efficiency in our operations to reduce the amount of energy needed to power our data centers, we’re also working toward procuring 100% renewable energy for all of our energy needs by 2030 and are on a path to reach that milestone globally by 2025,” he said.
Asia Pacific markets remain among the most challenging in the world for businesses seeking to source 100% renewable energy.
The study found that cloud-based data centers have significant efficiency advantages at both the server and facility levels. Further, it found that cloud servers are responsible for more than 67% of energy reduction due to the use of advanced technology and high utilization.
In line with these findings, AWS server systems are designed for power optimization and use the latest component technology. For example, custom-built AWS Graviton2 processors—offered exclusively to AWS customers—provide better performance per watt than any other AWS processor. AWS also runs servers with high levels of utilization and power efficiency, sharing and dynamically allocating resources across customers’ workloads on the cloud.
To execute on this front, AWS designs data centers to lower energy losses with optimized electrical infrastructure and energy-efficient cooling methods. AWS is also innovating the design of cooling systems to reduce water use and utilizes real-time sensor data to adapt to changing weather conditions.
451 Research stated that cloud providers have more aggressive renewable energy goals that many companies in APAC, and the scale of their investments is such that each new project adds tens of megawatts of new renewable power to the grid. However, there are significant barriers to renewable energy opportunities in APAC, as evidenced by the relatively low number of corporate renewable power purchase agreements (PPAs) executed there to date. According to Bloomberg New Energy Finance (BNEF), as of the end of 2020, there have been only 75
The key findings from the 451 Research report include:
- The cloud is five times more energy efficient than the average on-premises data centers of companies and public sector organizations in APAC.
- Moving to the cloud from on-premises data centers can improve energy efficiency and reduce the carbon footprint of compute workloads by 78% for typical enterprises and public sector organizations in APAC.
- Enabling cloud operators to source 100% renewable energy would reduce related emissions by another 15%, which, when combined with server and facility-level efficiency gains, could reduce IT infrastructure emissions by up to 93% for organizations across APAC.
- The average server utilization for APAC companies was just under 15%. By contrast, 451 Research estimates that cloud operators utilize servers well above 50% to find the right balance between efficiency and application performance.
