| Summary: | Data centers have long been known as one of the most power-hungry facilities in the world. The explosion of data in the past decades have led to a tremendous demand for data center capacity, thus further drawn attention to the data center’s industry ability to improve energy efficiency as it scales. Energy consumption by data centers can be broken down into two main components: energy consumed by IT equipment and energy consumed by infrastructure facilities (cooling system and power management). Energy savings largely depend on the later, as the former depends solely on the workload of the data center. A recent survey shows that most data centers are spending much more on cooling than on actual work, which suggests there are still opportunities for further enhancements. This project proposes a data center infrastructure management system on top of an existing AI-based energy optimizer to improve energy efficiency in data centers and thus reduce overall cooling costs. This includes a web-based admin interface and other backend services to support the interface.
|
|---|