Don’t stop innovating - Automate the risk out of new service deployment
While typing this blog, I’m sitting in my home office, which is imposed on many of us by the COVID19 pandemic, and thinking about how increasingly relevant the topic automation became in a fortnight.
Originally, I wanted to start to talk about how much automation is shortening Time to Market (TTM) or what beneficial impact it has on the Total Cost of Ownership (TCO) of telecom infrastructure. But then life presented a very real and terrifying use case.
The pandemic showcased a situation, when operators have to react in days or hours and not in weeks or months to an unforseen event. First of all they need to do the basics, like coping with the increased data/voice/video traffic, and then going further, to introduce relevant new or improved services to their subscribers, like messaging alerts, chatbot helplines and so on. Well, how to do that while respecting social distancing? The answer is very simple. Automation.
From vendor to operator – automation of the whole service lifecycle
The very real need for automation in the whole service lifecycle was made obvious even in the early days of the pandemic. It is also clear, that automation does not start at the premises of the Communcation Service Providers, but it starts with the vendors who need to develop and deliver their products and services in an agile manner; doing iterative development; continuous integration and continuous release. To be able to roll-out software as frequently as it is increasingly required, automation of the whole delivery process is a must. What should you look for so that you can automate the delivery process?
First, DevOps tooling, or a DevOps automation platform, with products that are pre-integrated to it. This essentially means that, there are pre-crafted VNF lifecycle workflows being released with the products, and all products are coming with their respective test-VNFs.
Then, once in service, digitally automated customer-tailored, yet reusable end-to-end workflows should be created for network design, deployment and integration.
Next, another crucial part of the process is software acceptance testing that also needs to be automated and provide end-to-end system regression, performance and robustness testing.
Finally, after deployment, a predictive method for continuous network monitoring is put in place to assure continuous Quality of Service.
Minimizing the risk
But how would this work in practice? Let’s say, we just deployed a new network service or a new version of a single VNF.
- We do verification and acceptance testing, where with the help of a test automation framework (TAF), the vendor executes and analyses hundreds of functional testcases, in a matter of hours, without any human interaction. When these automated end-to-end tests are passed, you can be sure that the network service did not break any earlier functionality and that the new service has been activated and works correctly. Hence, rollout to your live network can happen.
- Rollout is executed again by automated workflows minimizing human errors and significantly improving speed and the reliability of the rollout.
- Following the rollout, predictive network monitoring (PNM) continuously tracks network services, infrastructure, VNF KPIs and alarms as well as predictively look for anomalies of the network. If such an anomaly is detected or predicted, a ticket is raised with a proposed corrective action. Once the operator of the network approves the corrective action, the execution is triggered, closing the loop in assuring undisturbed Quality of Service.
With automation in place throughout the whole service delivery chain - from development to operation - CSPs can minimize the risk of network incidents and react to any unforseen situation in a very agile manner.
In our experience, we estimate a 60% leadtime savings and 90% reliability improvement over time by automating the whole service lifecycle, but most importantly and this is very timely, operators can make a difference in a crisis situation like COVID19. They can continue to reliably roll out relevant, innovative services to end customers in an increased speed by simply automating the process. So there is no need to stop innovating, but there is a dire need to start automating.
Visit our Nokia Service Lifecycle Automation for 5G Core webpage to find out more.
Share your thoughts on this topic by joining the Twitter discussion with @nokia or @nokianetworks using #5G #cloud #DevOps #ICT #innovation