How telcos can achieve their sustainability goals

Telco sustainability is the global, industry-wide effort to reduce the environmental impact of telecommunications service providers (telcos). Service providers are mandated by their governments to meet sustainability targets, which is an important step in addressing the global challenge of climate change. Being a sustainable telco is also becoming an important business objective in order to improve reputation, gain a competitive advantage, increase operational efficiency, and spur innovation.

Most major service providers have already committed to becoming carbon neutral by 2050, a goal supported by the Global System for Mobile Communications Association (GSMA). Many have also set ambitious targets for emissions and waste reduction, reduced energy consumption, and support of their customers’ short-term sustainability goals.

Service providers are already improving sustainability by adopting strategies and best practices that aim to reduce their environmental impact. Some of these methods include:

• Sourcing energy from renewable energy sources.
• Moving workloads to the cloud, especially to clouds powered by renewable energy sources.
• Recycling network equipment and end-user devices.
• Retiring old, energy-hungry technologies, such as 2G and 3G networks.
• Moving from copper wiring to fiber, which uses significantly less power.
• Making sure unused equipment and network components are turned off.
• Procuring more energy-efficient equipment.

Though many factors are involved in improving telco sustainability further, the most important of them may be energy efficiency and eventually maximizing the use of renewable energy sources. 

According to the GSMA, telcos are responsible for 2-3% of all energy consumption worldwide. Energy consumption constitutes between 20% to 40% of network operating expenses, and energy prices continue to rise significantly, making reduced energy consumption a top board-level discussion and a primary goal of telco sustainability efforts.

To help telcos build this process of optimizing energy efficiency, Red Hat is working with the open source community to define and build a sustainable architecture in a cloud-native environment. 

This approach requires new tooling but, most importantly, it requires open industry standards for technology and data to ensure interoperability. Red Hat’s portfolio of solutions, which is integrated across service providers’ datacenters, clouds, and networks, provides a unified platform that forms the foundation for tools and capabilities that promote sustainability and energy efficiency.

A layered approach to energy optimization

Network and IT environments are complex because they include multiple vendor products, logical and physical domains, and boundaries. When approaching energy consumption and control, Red Hat proposes a holistic approach to sustainability for telco networks which optimizes energy use at 4 different levels:

1. Node-level optimizations can be achieved by innovating central processing units (CPUs) and their architectures; types of hardware, such as power consumption by smart network interface cards (SmartNICs), hardware accelerators and graphic processing units (GPUs); or by fine-tuning node level parameters, such as core and memory frequencies, and disabling cores.
2. Cluster-level optimizations are available or applied holistically to a cluster. For example, energy-aware schedulers and de-schedulers, energy-aware clusters, or pod auto scalers.
3. System-level optimizations affect all the clusters and elements involved in delivering a service. For example, all the clusters, switches, routers, and antennas that comprise a mobile network.
4. Domain-level optimizations refer to a specialized service or system functionality, for example, the Radio Access Network (RAN) domain, the multiaccess edge computing (MEC) domain, a content delivery network (CDN) domain, or the IT domain.

Each of these levels has optimization opportunities that are unique to the domain they are a part of. For example, in the case of RAN, the energy consumption of antennas can be reduced by adjusting the energy to individual sectors or enabling and disabling them based on utilization characteristics. In the case of MEC, the optimization can focus on the location of the workload based on the type of energy source, or the overall utilization of a node. 

These 4 dimensions act as a reference that can be used by telco service providers to explore options available today and plan how to improve their options in the future.

Red Hat® OpenShift® is a unified platform to build, modernize, and deploy applications at scale. Red Hat OpenShift includes a number of important features that can help telcos optimize their energy consumption.

• The node tuning operator can prioritize power-efficient profiles or create custom-tuned profiles to achieve service providers’ goals.
• Workload hints and runtime optimizations allow nodes to identify if they should be tuned for very low latency at the cost of increased power consumption, or if more power-efficient tunings can be applied to the node.
• Node network latency profiles can reduce the rescheduling of applications on environments experiencing high network latency, impacting power utilization thresholds of other nodes in the cluster.
• Application idling can allow platform administrators to temporarily disable applications not being used to reduce power consumption.
• Cluster hibernation allows administrators to deactivate clusters running in public clouds, reducing resource consumption during times when they are not being actively used.
• Serverless autoscaling can automatically scale applications to match incoming demand. An application that is not receiving traffic can be scaled to zero when the scale-to-zero capability is enabled.
• Autoscaling and rightsizing of applications can be used to scale workloads based on CO2 emission or power consumption metrics.

Power monitoring for Red Hat OpenShift 

Power monitoring for Red Hat OpenShift captures and monitors power-use metrics across the ecosystem, focusing on reporting, reduction, and regression so enterprises can better understand energy consumption. It is based on the community-driven, open source project Kepler, or Kubernetes-based Efficient Power Level Exporter, founded by Red Hat’s emerging technologies group with early contributions from IBM Research and Intel. 

The captured energy data enables multiple integration points and provides the capability to:

• Report power consumption at a variety of levels, including containers, pods, namespaces, or different compute nodes in the cluster.
• Estimate the carbon footprint of the workload by coupling energy consumption metrics with a datacenter’s power usage effectiveness (PUE) and electricity carbon intensity.
• Reduce cluster-level power consumption with power-aware workload scheduling and auto-scaling.
• Produce more sustainable software products by deploying Kepler in continuous integration and continuous development (CI/CD) pipelines for software testing and release.

The Red Hat certified partner ecosystem can provide additional functionality and improved capabilities such as machine learning models, smart automation, and further data aggregation. 

Red Hat works with its partners to expand their abilities to measure and control energy consumption and to integrate their built-in capabilities into a unified architecture.

Partner with Red Hat for a sustainable future

Red Hat is committed to partnering with service providers to modernize their business and deliver innovative, integrated solutions to their customers in a more sustainable way. Read the Sustainable service providers whitepaper to dive deeper into how Red Hat can help service providers increase sustainability and energy efficiency.