The telecommunications industry is undergoing a major transformation with the adoption of fifth generation mobile technology (5G) that is heralding a massive shift for public, private, and societal use cases. Unlike earlier generations of network upgrades, 5G expands the supplier ecosystem to embrace software-defined network functions that bring with it agility and innovation.
With this shift, industry leaders such as Deutsche Telekom (DT) and others recognise that old ways of building, managing, and upgrading networks must adapt to modern realities. By working closely with leaders within testing and assurance such as Spirent, Deutsche Telekom has pioneered a new strategy, built on a notion of highly automated environments with integrated testing. These greater levels of automation underpin a new operating model able to deliver on the true potential of both 5G and interconnected telecommunications systems.
To deliver on this vision, DT has re-invented how it runs its operational workflows. This moves away from a legacy approach of a few vendors providing in essence the full vertical stack, and towards an approach where potentially dozens of vendors may be involved in the underlying infrastructure or functions of the largely software-based architecture. The operator then internally needs to integrate and operate a diverse stack of software, hardware, and management elements to gain more control and agility over its environment.
Integrator and operator
From being an operator managing fixed stacks, innovators such as DT are now taking ownership of both Integration and Operations – defined internally as IntOps and supported by established methodologies and technologies such as GitOps. IntOps is not an industry standard, but it addresses the flexibility versus stability dilemma which every telecommunications operator faces in the context of managing software-defined network functions.
A crucial aspect of the IntOps approach requires consistency and reproducibility across the change processes. This includes many elements such as modelling software-defined network functions in a declarative way and establishing a well-defined approach to customize software network functions so they reflect the environment-specific context. Another aim is to manage the lifecycle of all software process instances via closed-loop automation. Here test automation is the required complementary technology for ensuring end-to-end service quality for the customers of telecommunications operators.
To capitalise on the potential of IntOps requires automated continuous testing (CT). This means each stage of the pipeline must include CT methodologies rather than creation of additional pipeline stages for testing. As an analogy, to make a train go faster than its design allows requires more than just a new engine. It will likely require a redesign from the ground up to ensure all the elements are able to cope with the new demands. Success also requires equipping the pipeline with the right tooling to facilitate continuous testing.
10 steps of IntOps and test automation
One of the fundamental principles of IntOps is to help create structured and repeatable stages to simplify what is potentially an incredibly complex workflow. To do this, DT broke down the process into ten steps. Each step may have its own functional elements, but this framework helps to articulate the concept and maps against real-world stages of onboarding of network solutions. These are broadly categorised within 5 key areas: vision, architecture, process, implementation, and closed-loop automation. The vision paper “” outlines all 10 steps and the role of test automation.
Embracing the industry
IntOps has been adopted by DT and is demonstrating practical benefits and the shift towards its intrinsic methods is an emerging trend across the industry. The goal is to help operators benefit from the advantages of software-defined network services and ultimately deliver the best customer experience levels. The longer-term aim is to continue to excel as the scale and scope of products expands into emerging areas such as smart cities, connected cars, private enterprise networks and IoT.
The role for CT as defined by IntOps is also shifting in line with the needs for a more dynamic, agile, and adaptive network. Concepts such as “Design for Failure”, once considered radical, are now accepted as best practice and vital to enable the levels of scale that modern 5G networks will need to embrace to meet new demands. In practical terms, this means that CT is deploying new capabilities such as network digital twins based on the needs of IntOps and similar strategies.
The last crucial point is that IntOps is a guide not a prescription and it is almost certain that IntOps, like DevOps and NetOps before it, will evolve over time – as will the selection of tools and technologies that are considered as “best in breed” to deliver on its vision. However, by sharing the IntOps vision, we hope to foster knowledge exchange between the various stakeholders across the operator, supplier, and partner communities – and ultimately help to make the 5G transition to the next generation of services better for all.
To learn more, read the vision paper:.
Vitus Zeller, a Spirent Solutions Architect focused on delivering network virtualization solutions in the EMEA region, contributed to this blog.