Open source software plays a significant role in environmental sustainability and reducing the data center industry's energy footprint, Red Hat Principal Sustainability Community Architect Cara Delia said during a KubeCon Europe 2023 keynote address.

"Sustainability has taken the world's stage, and rightfully so," she argued. "It's through open source we can provide critical transparency, traceable and inclusive decision-making and collaboration for innovations."

Emerging technology at Red Hat tends to focus on the cloud-native aspect of sustainability, including projects, methodologies and technologies. In today's era of cloud, cloud-native sustainability has the potential to touch a variety of areas from chip design to application development.

The vendor claimed its open source technologies and contributions are accelerating research, decarbonization and conservation efforts in the public and private sectors, Delia noted.

Energy conservation and carbon emissions reductions are particularly attractive goals for data center operators considering that industry's role in climate change; according to the International Energy Agency, data centers are responsible for 1% of global energy use.

"It is very important for all of us to support open source software and close the knowledge gap to develop better technical interfaces between platforms, data, models and open source tools to help with climate alignment," Delia explained.

Delia cited the vendor's Project Kepler, short for Kubernetes-based Efficient Power Level Exporter, which uses eBPF and machine learning technologies to "exploit workload energy consumption across a wide range of computing platforms."

Senior Principal Software Engineer Huamin Chen, who leads the Kepler project, explained the tool measures the energy use associated with running workloads in an on-premises data center environment or a public cloud.

Using eBPF technology, Project Kepler is able to "crack information from the hardware, software and user level," Chen said. That information is then used to build up machine learning models that "later on will be used to calculate the energy" used by Kubernetes workloads, containers and clusters, he explained.

"This is so exciting because with these metrics it enables a lot of innovation" in terms of tuning, scheduling or scaling Kubernetes workloads to optimize energy and carbon use.

Project Kepler was developed in collaboration with Intel, Weaveworks and IBM, and touts contributing members from all corners of the globe. Kepler is also a proposed Cloud Native Computing Foundation (CNCF) Sandbox project awaiting approval.

"With all these contributions, big or small, we hope to give ourselves, our industry, your industry and our future generations a sustainable environment," Chen added.