3 steps for delivering network automation at scale
My previous post looked at how taking a ‘brick-by-brick’ approach to network automation can help you get quick wins compared to lengthy, expensive deployments that span multiple systems and departments. It also showcased some examples of key use cases that deliver instant value for your business, including zero-touch provisioning for optical lambdas to support capacity requirements in the IP-MPLS layer using Ciena Blue Planet Multi-Domain Service Orchestration (MDSO) software.
The next step on your automation journey: deployment at scale
As we’ve seen already, the incremental approach to rolling out automation use cases is a great way to start, but you’ll need to scale up your efforts pretty fast if you want to extend the benefits out to all your divisions and operating companies.
But deploying automation projects at scale is never going to be all smooth sailing, especially when you think that probably the majority of most operators’ service revenues are generated by legacy network equipment and platforms. That means that you have to decide which assets are still valuable for your business for revenue generation, over time.
Deploying automation projects at scale is never going to be all smooth sailing, especially when you think that probably the majority of most operators’ service revenues are generated by legacy network equipment and platforms.
Although this sounds hard – and it can be – there are three steps you can take to help speed up your deployment of network automation use cases at scale:
1) Rationalise your infrastructure assets before you do anything else - Because every one of your legacy network assets is at a different stage in its lifecycle, you need to know what you have, which services touch which equipment, and how much value every piece of your infrastructure is adding to your revenue streams. Based on this in-depth analysis, you’ll be able to make better decisions about which assets to bring into your automation environment, which ones to leave where they are, and which need to be retired ASAP.
The only problem is that your assets will be running in multiple, unconnected Operational Support Systems (OSS), with records held in multiple inventory systems. That means obtaining a single view can be difficult or impossible, especially as documentation for older equipment is probably on paper in filing cabinets across multiple locations. There’s also the fact that a single service touches multiple network components and layers, and those connections are not always easy to unravel.
To give you a clear picture of your existing environment, Ciena’s Blue Planet Inventory (BPI) solution federates data from across the multiple OSS sources to provide a centralised view of all your network components, from underlying fibre and optical equipment, to multi-vendor IP-layer equipment, to virtualised network elements. Furthermore, it provides visualization of the logical customer services that are running over the underlying infrastructure. By providing a simple, automated solution for mapping your network in terms of service topologies, BPI helps you bring the necessary legacy equipment into your automated environment, and also helps you to eliminate product variants that are no longer needed – helping you simplify your network footprint and achieve major cost savings.
2) Build a centre of excellence – and a passionate team to run it - Even to deliver small, focused automation projects, you’ll need highly motivated, passionate people with the right skills. By bringing these kinds of people together in a centre of excellence, and empowering them to build use cases that can be deployed at scale, you can accelerate your journey to network automation.
Creating a centre of excellence in house also means you can take ownership of your automation journey and reduce the need for costly external consultants. This ensures you can deliver automation projects at scale when the time is right, maximise cost efficiency for new development projects, and stay in control of your transformation strategy.
While the centre of excellence is the epicentre of your automation operations, delivering automation at scale requires collaboration with teams and departments across the business. As well as building a culture of automation with strong top-down leadership, you will also need to foster knowledge and skills transfer between your centre of excellence team and other tech professionals responsible for network projects in their respective areas of the business.
3) Optimise your development and delivery capabilities with ‘agile sprints’ - When we talk to customers, we find that most are still using a ‘waterfall’ method to deliver automation projects and capabilities. This involves a large program of work with multiple milestones, for new product development, with tangible outcomes often not delivered for 12 months or more. These kinds of projects also tend to be extremely restrictive in terms of governance and workflows, and extended timelines and high costs are common.
One way you can deliver automation use cases faster is by borrowing from ‘Dev Ops’ software development techniques, and especially agile sprints for delivering new automation features in short periods of time. Speed of development is facilitated by the open interfaces that are provided by modern software systems. This continual approach to development, which needs to be driven by the centre of excellence, allows you to deploy new automation use cases incrementally and to speed up returns on investment. To support migration to this kind of development approach, Ciena Blue Planet DevOps Exchange and Community runs hackathons, co-development events and engages with telcos at all level to help you increase your agility and move away from slow, monolithic projects that slow down your ROI.
Creating the right environment for automation success with Ciena
At Ciena, we have the automation know-how and smart network technologies to help you develop and deploy automation use cases at scale in your business.
We can help you create the right environment for automation based on a modern network architecture, our leading platform for network orchestration and automation, and our skills and experience of helping operators around the world to automate key processes and accelerate time to market for new, revenue-generating services.
Thank you. Your comment has been received and should appear on the blog shortly.