The green credentials of your software

How green is your software? Mobile World Congress 2022 was the first time this question was posed in the context of a software solution for telecoms. It felt as though it had come out of left field because a large part of our marketing story focuses on differentiating solutions through the expertise of the various teams, the ease and speed of integration, flexibility, cloud-native microservice, and so on. The green element or the positive impact on scarce resources the software can affect was not a priority in building the 5G story. So, to answer any questions as to how software can be “green”, here are some considerations.

Simply put, it comes down to the level of control that can be exerted to deliver greater efficiency to run (and host) 5G telco core networks in the cloud. First and foremost, the explosion of devices and the associated data traffic enables 5G abilities to scale up and down. The core principle in telco is the resilience of the network and an important part of this was to build capacity for peak usage. This principle in legacy networks comes at a considerable expense seeing as the CSP is building network capacity for a peak event. On average, this might happen twice a year, so this additional capacity sits idle for the rest of the year which is incredibly wasteful. The resource in question is computing resources – often in the form of servers requiring enormous quantities of electricity to keep them operating effectively, and clean water to prevent overheating.

Now, looking at a 5G SA (standalone) core network – built in the cloud. The fact that it is cloud-native means CSPs can avail of cloud functionality such as scalability. So when a peak event occurs, the software running the network can automatically allocate computing resources – as needed, hence, reducing computing resources. As demand for network service reduces, in effect as does the auto-scaling. This drives far greater efficiencies for the CSP and means more sustainable use of the resources required to run the network. This best defines the situation today, and any CSP who is operating any form of automated network in the cloud can attest to this. However, there are further improvements that the new 5G functions (software) can facilitate CSPs in leveraging their networks. Close loop automation utilizing AI/ML (Artificial Intelligence/Machine Learning) to make decisions on the operation of the network removes any human interaction or dependency on a decision from the equation – dramatically improving network efficiency.

Consider the millions of extra devices to which the 5G network will connect. Due to the sheer volume, any reliance on human decision-making to manage network resources is not possible. This is largely why network automation is central to 5G cloud-native networks. NWDAF (Network Data Analytics Function) is a relatively new 5G function that provides control layer intelligence to 5G Network Functions to assist the automated decision process. Its role is to ensure optimizing the deployed resources to provide consistent and appropriate service levels particular to that function. Axing the need to build capacity for a peak event capacity Legacy system. With a continued focus on how we can use our scarce resources more effectively and how to reduce the impact on the environment, the positive effect of network automation and new 5G functions like NWDAF should be a top priority in how 5G network capabilities are built in a more sustainable manner.