EU project investigates artificial intelligence augmented cooling system for small data centres

Share
Link copied! Copy link
EU project investigates artificial intelligence augmented cooling system for small data centres

 

The Sustainable Digital Infrastructure Alliance (SDIA) is collaborating in a first large-scale research project which will aim on increasing the energy-efficiency of existing data center facilities. The project is funded by the EU under the Horizon 2020 project and has reached the highest score in the project call.

Artificial-Intelligence-Augmented Cooling System for Small Data Centres or ECO-Qube as it is known is a holistic management system that aims to enhance energy efficiency and cooling performance by orchestrating both hardware and software components in edge computing applications.

This is a data driven approach that utilises valuable unused data from active data centre components. Created big data is being used by an artificial intelligence augmented system which detects cooling and energy requirements instantaneously.

ECO-Qube differentiates from conventional cooling systems that keep operating temperatures within a strict interval and do not evaluate measurable cooling performance. Unmeasured cooling performance leads to underperformed airflow, thermal disequilibrium, and high energy consumption. By contrast ECO-Qube offers a zonal heat management system that benefits from computational fluid dynamics (CFD) simulations to adapt cooling system for the best airflow and cooling performance with minimum energy consumption.

In addition it utilises smart workload orchestration to keep the CPUs at their most energy efficient state and maintain the thermal equilibrium to reduce overheating risk.

Sustainability is another priority for ECO-Qube’s smart energy management system (EMS), which is designed to track the energy demand and operate the energy supply in cooperation with building or district’s EMS. This synergy maximises the energy supplied from renewable energy sources and minimizes the energy supplied from sources with big carbon footprint.

The solution will be assessed in three different pilots from different climatic conditions to validate energy efficiency under different external variables.

“Many cooling and energy-efficiency technologies are emerging, particularly adiabatic, liquid- as well as immersion-cooling,” Lasse Schneppenheim of the SDIA said. “These technologies dramatically increase the efficiency of newly built data centres. However, to successfully transform the digital infrastructure sector towards having no negative impact on the environment, new and existing data center facilities are required to improve their energy and resource footprint.”

Existing data centres often operate on IT utilization levels as low as 15 per cent. Despite such rates, the cooling and electrical infrastructure operates on a near full load configuration – irrespective of the actual IT load. 

As part of the ECO-Qube project, the SDIA and a consortium of organisations from across the value chain, aim to interconnect the data center building, electrical infrastructure and the IT workloads, in order to reduce resource consumption to a level that matches the operational IT capacity.

For this purpose the consortium brings together established data center suppliers, building management software, fluid dynamics experts, universities and pilot facilities to explore how existing air-cooled data centres can be transformed through precise measurements, simulations and machine-learning-based optimization.

“Together, the consortium will equip new and existing data center facilities with ECO-Qube sensors in three different locations, build fluid dynamics models for each and execute optimization strategies using software,” Schneppenheim concluded. “Additionally, the project will research the integration of renewable energy and waste heat valorisation solutions.

“The Sustainable Digital Infrastructure Alliance is a proud partner of the project and will lead the dissemination of the results and the progress of the projects. It is one of the many important technologies which are part of our Roadmap towards Sustainable Digital Infrastructure by 2030.”

 

 

Talks