ST Telemedia Global Data Centres (STT GDC) is working with ABB to conduct an artificial intelligence (AI) energy optimisation pilot in Singapore as the former transitions its facilities to be net carbon-neutral by 2030.
The objective of the pilot is to leverage AI and machine learning (ML) technology and advanced analytics to optimise energy use within a data centre and reduce a facility’s overall carbon footprint.
“Our group’s AI roadmap will take a big leap forward with this pilot program and focus on optimising our data centre’s energy utilisation,” said Daniel Pointon, group CTO at STT GDC. “During the review of the pilot results, we will also assess future AI-based applications such as predictive maintenance, capacity management, security analytics and other business processes in our considerations.”
The pilot will span two phases – an initial data exploration, modelling and validation phase, followed by deployment of AI control logic in a live data centre environment. The pilot aims to achieve at least 10% reduction in energy usage by the cooling systems, which contributes most to the consumption of electrical power in a data centre after our customer’s IT equipment.
Optimising the cooling efficiency will in turn improve power usage effectiveness (PUE) and reduce the carbon footprint, with an estimated reduction of 1,100 tons of carbon in the atmosphere for every 10 megawatts of IT load utilised — equivalent to saving about 18,000 trees.
Going further than previous AI research work constrained to white-space of data centres, STT GDC’s pilot is taking a holistic approach and developing AI neural-network models for the entire DC cooling plant including the upstream chiller and distribution systems.
The AI project is also unlocking new opportunities for efficiency improvement at a granular level within the data centre. With the ability to glean AI-generated insights, STT GDC can effectively track and analyse the wealth of data generated by monitoring systems in the data centre, and better facilitate dynamic cooling optimisation to reap the desired outcomes.
“In view of growing data demands from end users and emerging business applications that require increasingly higher computing power, the resources that data centres consume have the potential to increase if not appropriately addressed,” said Pointon.
The AI project will be an on-premise solution that will tightly integrate to the existing ABB Ability Data Center Automation system on site, which provides comprehensive monitoring and control of the entire data centre infrastructure.