Prior to the emergence of modern, programmable networks, legacy network management principles such as manual CLI and scripting were enough to keep businesses up and running. However, programmability has proven to be critical in order to interface with the complex ecosystem of technologies that are involved in today’s network operations.
Many organizations are struggling to reconcile the disconnect between traditional network engineering, with deep knowledge of the physical network, primarily managed from the CLI and the emergence of an API-driven, software-defined world that heavily relies on DevOps and agile software development skills. As engineers have started their journey into network automation using scripts with Ansible and others, they tend to focus on basic network operational use cases such as software upgrades, device onboarding, or port turn-ups. The problem here is the automation focus is only on the specific execution of the network task. In fact, by only focusing on the individual network change executions, they're effectively ignoring 80% to 90% of the automation opportunity.
Stop Ignoring 80% of the Process
The reality today is that traditional methods of managing the network only address between 30 minutes to 2 hours of a typical network change process that can span weeks of time, including ten to 20 hours of manual effort overall. On average, network operators implementing automation only end up addressing between 10% and 20% of the possible work involved and ignoring 80% to 90% of the process that they could be automating.
So, what does the ignored 80% of the automatable network processes look like? Beyond execution, organizations are ignoring the opportunity to optimize and automate the entire network change process, from planning and scheduling, all of the pre and post network health checks, and inventory and change request updates. The resources required and transition time between these various stages requires an extensive amount of swivel chairing between systems and between the various people across the organization that have to manually interact with disparate network management systems. Each of these manual steps has a cost and increases the error rate of network changes, driving fear and uncertainty in the organization every time a change needs to be made.
In order to achieve end-to-end automation, organizations need to broaden the scope of their automation efforts to better align and increase the velocity at which they can deliver change.
Expand Automation Participation
On top of the prescriptive process, network automation is all about the people responsible for building, maintaining, and operating these networks. As networks become increasingly distributed and complex, this means accommodating a wider range of groups and skills, and their respective systems, across an organization in order to move a change forward. A typical network change process involves prep and pre-checks all the way through to post-checks can involve groups that span security, network operations, network planning, network engineers, and IT.
With over ten to 20 people in those groups actively engaged in the change, as well as the five to ten systems those people have to access in order to request changes, reserve resources, update network monitoring, etc., it’s critical that organizations move beyond spreadsheets and scripts and toward a self-service framework that automates all the interactions and handoffs.
Imagine, network tasks and activities that previously required a 60-page step by step manual (which detailed what to do, who to contact and how to handle errors at each stage of a network change) could be replaced by automation.
End-to-End Network Automation
Automating the end to end process of network management addresses functions like:
- How to assign human resources at the right time of an operation
- How to allocate network resources to ensure business applications and services will continue to run
- How to schedule network changes within appropriate maintenance windows
- How to complete change requests
- How to verify that the network is in a proper state to accommodate the change before you run that script
- How to validate that the change didn’t break anything after the change is made
An end-to-end perspective towards network automation ultimately needs to traverse all operational silos and integrate with all of the necessary external systems and network technologies involved in the entire operational process. This will enable organizations to rapidly automate today's complex networks that span traditional physical networks, next-generation programmable networks, SD-WANs, and cloud networks.