Introduction: Automation and Changing Roles
With the advent of emerging technologies like edge cloud, 5G networking, Artificial Intelligence (AI), and the Metaverse, enterprise IT infrastructure and operations are expanding and accelerating to levels that tax the capabilities of even the most skilled network engineers and application developers. The need for rapid (often real-time) response, 24/7/365 availability, and the demands of a changing and ever more sophisticated market call for less reliance on human intervention for nuts and bolts operations, and increasing levels of automation.IT automation involves the creation of software and systems to replace repeatable processes and reduce manual intervention. Software tools, frameworks, and appliances conduct tasks that essentially replace the series of actions and responses that would otherwise occur between an administrator and the IT environment. Automated repeat instructions, processes, or policies save time and free up IT staff for more strategic work.
Why Automation Matters
Network automation is becoming imperative for organizations in the digital era. Since it reduces operational costs, prevents network outages, and enables agile IT operations, network automation can support mission-critical operations and has the potential to create a competitive advantage.
But although network automation is gradually gaining traction, it still lags behind DevOps, security, and other IT infrastructure domains in its automation maturity. This is due in large part to the continued reliance by network engineers on manual management models and practices dating back to the dawn of the Internet in the 1990s.
While the Command Line Interface (CLI) and spreadsheets largely remain the bedrock of network management, network engineers in the digital age must look beyond the manual techniques, to keep pace with the progress of technology.
This is especially relevant for commercial organizations. Forbes estimates that human error is the number one cause of network outages and is responsible for 70% of policy violations. As a result, enterprises are spending 75% of their network operations expenditure budgets in a year on troubleshooting alone. What’s more, network downtime is costing enterprises as much as $300,000 per hour, on average, and up to $540,000 per hour in worst-case scenarios -- in effect, killing business performance,
Embracing Automation with NetOps
NetOps, also known as NetOps 2.0 or NetDevOps, is an approach to networking operations that use Development & Operations (DevOps) tools and techniques for network configuration and management. NetOps enables agile, scalable, and programmable networks that support faster software development cycles. Its implementation, therefore, requires a degree of overlap between the skill sets of software programmers and network engineers.
In an app and platform-dependent economy, no system or piece of software operates in a vacuum. So, developers need an understanding of the building blocks of their application and how it functions on the underlying network. This in turn necessitates some level of comprehension of network engineering fundamentals -- including the concepts of latency, loss, congestion, and topology that characterize real production networks.
On the networking side, technologies like AI, cloud, machine learning, and data analytics all require some level of coding skill in their implementation, monitoring, and maintenance. The broader field of network automation also requires network engineers to make use of scripts that handle configurations. Consequently, for network engineers, learning programming skills is a great asset.
On either side of the development/networking divide, the outcome should be a NetOps engineer, who understands networking and its intricacies but is also able to program.
Looking to The Future
As we have noted, emerging technology trends like edge cloud, 5G, and the Metaverse will all require different degrees of network automation. In and of itself, automation will mean that we see very little manual network management in the future.
However, chances are that the future will be so much more network-intensive that while manual work gradually disappears, there will be huge opportunities for network engineers. This could be in areas such as developing automation such as workflows, personalization for enhanced user/customer experiences, and other specialist applications that must be programmed into the networks of tomorrow.
This kind of evolution is inevitable -- and not something that needs to be feared. After all, when the bank tellers of yesteryear were largely replaced by first ATMs and then online banking, humans remained in the loop. As FinTech assumes a greater role in financial services provision, people will remain in the system in various capacities: specialist advisors, platform creators, maintenance engineers, and troubleshooters.
So, although the networking community is understandably thinking about how automation and the new emerging technologies will impact their jobs, we see a very bright future ahead, for the industry as a whole. It’s just that to embrace it, network engineers will need to learn new skills and try to get their heads around new ways of thinking.
Once this happens and the art of network automation has been mastered, we will see levels of service agility and resiliency that will put today's networks to shame.