For the last two decades, independent software vendors have been able to serve operators with operational support systems (OSS) that have continually improved processes. OSS vendors have managed to stay ahead of the curve, says Tim McElligott, senior analyst at TM Forum, because the underlying fundamentals of networking have remained intact.
But fast forward to today and networking is rapidly changing. The cloud does not operate like a switched network and virtualised architectures no longer behave like physical networks. 5G networks will demand a scale and receptivity that OSS will struggle to deliver.
This has led communications service providers (CSPs) to weigh up how they transform operational and business support systems (OSS/BSS). There are many benefits to this transformation, but it’s important that CSPs don’t underestimate the requirements for making this a success. Legacy methods and ways of working are currently aligned with traditional OSS and redefining this means redefining entire internal cultures.
Emulating success stories
The good news for CSPs is that service providers across the spectrum have already made similar wide-reaching transformation projects a success. Virtualisation and cloud architecture are commonly implemented in the IT networks of today. Examining how the IT industry successfully transformed networks holds clues for CSPs looking to do the same for their own networks.
Even with this insight, stumbling blocks remain. In a TM Forum report, Digital Transformation Tracker 2: How to fix the cultural divide, 75% of CSP respondents said they believe cultural and organisational issues are a very serious or moderately serious challenge to digital transformation. A key concern under this umbrella is around a lack of necessary software skills. CSPs’ existing workforces don’t have the skills needed to install and manage the hardware that comes after traditional OSS.
Rather than making a raft of new hires to fill the skills gap, CSPs should instead think about reskilling their existing workforce. Doing so will come at a great cost, but there are additional benefits to investing in the workforce. When reskilling employees, CSPs can implement ways of working that are compatible with agile development, namely DevOps practices.
Applying DevOps
Like cloud architectures and virtualisation, DevOps got its start in IT. Increasingly, however, CSPs are experimenting with DevOps, adapting a collaborative approach towards end-to-end service delivery where development, production and business teams collaborate to common goals.
DevOps means continuous integration, testing and deployment. Arranging small, cross-functional teams to maintain this steady, constant output stream isn’t easy – particularly for CSP teams that are accustomed to legacy development practices such as the waterfall model. The difficulties, however, shouldn’t distract CSPs from the advantages of implementing DevOps, with one of its key benefits being the drastic reduction in time to market for product rollouts.
Managing the 5G networks of the future requires the kind of speed and seamlessness that current OSS practices don’t support. DevOps, on the other hand, speeds things up considerably by implementing a continuous cycle of development, deployment, assurance and improvement in which no group lets go of their responsibility for the success of a service.
Working in fast-paced cycles also allows for more experimentation and testing. Those using DevOps in the IT world talk about the idea of “failing fast” – this concept absolutely applies to the telecoms industry. Through increased testing, CSPs can quickly see what does and doesn’t work with a product or service – as opposed to blindly hoping that current developments will be the next “killer app”.
The agile future of OSS
When thinking about the future of OSS, it’s helpful to imagine an entirely new concept rather than a simple upgrade of current OSS. For TM Forum’s recent Future OSS: Towards an open digital architecture report, operators and vendors said that their top two attributes for future OSS are automated, closed-loop service fulfilment and assurance, and automated, closed-loop network optimisation. (See chart below for the top 10 attributes.)
For those accustomed to traditional OSS, knowing what future OSS needs to improve on is easy. The hard part is knowing how these improvements can be made. Implementing open application program interfaces (APIs) is a good starting point. Through utilising component-based architecture, like TM Forum’s Open Digital Framework, business services of components can be exposed as a set of interoperable Open APIs. These APIs can be further decomposed into a set of services and microservices. The advantage of using microservices is that they can be managed on scalable infrastructure using agile development practices.
Operators have conflicting views about which form future OSS should take. Some are for implementing cloud-native technologies to develop applications built with services packaged in containers. Other CSPs, however, are more wary about embracing the public cloud. There’s a desire to keep networks and related operations on-premise or in a private cloud because networks are viewed as a strategic differentiator, and then there are concerns around the reliability and security of the public cloud.
While the debate around the use of public cloud rages on – it’s evident that the future of OSS is agile. Working in DevOps cycles will enable teams to more quickly implement and upgrade the technologies needed to automate and make networks more data driven. DevOps also allows teams to embed security and privacy into the development cycle of any application or service—a massive improvement on traditional OSS, where security and privacy is often a bolted-on afterthought.
It’s positive that most CSPs understand the need to transform OSS, but these CSPs shouldn’t underestimate the requirements of doing so. Retraining workforces is a significant investment, but if it results in successfully implementing agile DevOps internally then the investment will pay for itself. Shifting internal culture when transforming OSS doesn’t need to be a chore. It’s a golden chance for CSPs to speed up service delivery, leverage emerging technology and ultimately improve the experience for customers.
The author is Tim McElligott, senior analyst, TM Forum
About the author
Tim McElligott is senior analyst at TM Forum. McElligott joined TM Forum in January 2019, after working as senior analyst at Frost & Sullivan, editor in chief of Billing & OSS World (B/OSS magazine), and senior editor at Telephony. He is currently focused on: industry transformation and collaboration, open architectures, digital service delivery and monetisation, automated service development and assurance.
Comment on this article below or via Twitter: @VanillaPlus OR @jcvplus
