Bejoy Pankajakshan, Executive Vice-President and Chief Technology and Strategy Officer, Mavenir
Service providers are becoming increasingly aware of the need to migrate their networks to a cloud-native environment. However, they have raised concerns about the challenges in the network transformation process. Their apprehensions are well founded since embracing a cloud-native architecture entails significant changes and reorganisation, not only in technological aspects but also in the traditional operations of telecommunication companies. The vast scale of the telecommunication network, combined with complex configurations, further complicates the process. Nevertheless, telcos can improve their chances of achieving favourable outcomes in their transition to a cloud-native environment by being mindful of a few key factors:
Mitigating workforce challenges
The proficiencies required to manage a modern cloud-native network are significantly different from the current skill sets. In conventional service providers, the network and IT teams typically operate independently, but the cloud-native environment requires the integration of both skill sets. To ensure the teams work in tandem with the evolving demands, telcos must amalgamate and restructure the two teams. They may have to enlist additional resources as IT skill sets become more imperative in a cloud-native environment, and adopt DevOps practices.
Infrastructure and tooling
The deployment of a cloud-native network typically requires a variety of tools to support the development, deployment and management of network infrastructure and services. Some of the key tools required for cloud native network deployment are continuous integration/continuous delivery pipelines, network security tools such as firewalls, intrusion detection and prevention systems, monitoring and observability tools such as Prometheus and Grafana, service mesh technologies like Istio and, most importantly, network function virtualisation platforms.
Adopting novel models
Traditionally, telcos opt for one or two network equipment vendors. However, in a cloud-native environment, they would be collaborating with new stakeholders such as Wind River, Red Hat and VMware, which provide the underlying container software platform, as well as public cloud providers. Telcos should exercise caution and prudence regarding the commercial models associated with support and licensing, whether they are based on perpetual or recurring revenue.
Legacy infrastructure
Existing legacy systems and applications may not be designed for cloud-native environments, which can make it challenging to integrate them into a cloud-native network. A way to mitigate this challenge is implementing interworking functions, which interwork open interfaces with legacy protocol, allowing operators to undertake a gradual transition from legacy applications or interfaces.
Design for failover scenarios
Superior design practices for cloud-native functions (CNFs) should be adopted so that application state information is not stored on a single instance. Since cloud-native applications are designed to run in distributed environments and can be easily replicated across multiple geographic locations, it is crucial to ensure high availability and geo-redundancy. In public clouds, high availability can be achieved by deploying redundant CNFs in separate availability zones within a region. Geo-redundancy can be achieved by deploying redundant CNFs in separate availability zones across multiple regions.
Security
Cloud-native applications are deployed in a containerised environment that provides security at the container layer, cloud layer and cluster layer. However, additional security measures must be considered during the early stages of the application development life cycle to mitigate potential security breaches. The open interfaces implemented in cloud-native applications will pose new security challenges. These interfaces must ensure privacy protection, protection against external threats and compliance with regulatory requirements.
Leveraging network visibility and analytics
One of the primary advantages of transitioning to a cloud-native environment for telcos is gaining visibility into the network, enabling them to leverage analytics to enhance network performance and provide a genuinely unique experience to subscribers. In addition, telcos can leverage full slice assurance to monitor the performance of network slices and optimise the network in real time to meet the requirements of each slice.
Service-level agreement requirements
The adoption of a cloud-native network opens up new possibilities for service providers to provide a differentiated experience and expedite the launch of new services while simultaneously reducing the total cost of network ownership. Mavenir is currently assisting several service providers in realising the complete potential and benefits of migrating their networks to a cloud-native environment.
