Dan DeBacker is the Director of Data Center Product Management at Extreme Networks.
Digital transformation is redefining how businesses work. We’ve all heard the noise – from cloud mandates to IoT devices to innovative uses of artificial intelligence. But those shiny objects obscure what should be the priority in any transformation effort: the data center.
IT staff, who are under incredible pressure, need to find a way to embrace digital transformation initiatives, up their skill sets in a way that will help the business move forward, and not just keep the lights on, and ensure that the data center – the enterprise’s Digital Nervous System – modernizes along with the company. This is a tall order, and though there is no single answer or silver bullet, there is one strategy that can help in nearly all instances: automation. But as with any buzzword du jour, there’s a lot of misconceptions about this topic. It’s time to separate fact from fiction when it comes to data center automation.
DevOps/NetOps is the Future
There is little debate left here: The time to embrace DevOps/NetOps is now. Don’t be too alarmed here, as this is a journey and not simply a destination. The goal of these methodologies is to streamline various aspects of the data center. There is a need to increase the velocity of service delivery which is a direct requirement of digital transformation. Agility within IT is a cornerstone to making this a reality. Automation is a basis for these frameworks to be successful, yet another major component is having an organizational structure that supports the effort. By simply piling automation on an organization and not utilizing it up properly just adds more tools and overhead.
One Size Fits All
When it comes to automation, this couldn’t be further from the truth. A big benefit of automation is the ability to customize. Why should we fit our business around a generic tool meant for the masses? This approach will never yield results anywhere near what is needed. The ability to streamline automation workflows in the data center is where the value lies. We should be able to take a workflow that is manual today and simply translate that into a process that is automated.
For example, when a new virtual machine is created, we could automatically provision the network connectivity. From there, we can add workflows to configure load balancers, firewalls or other services required for application roll out. Sounds simple, but you must choose wisely and not get duped by tools that claim this but require a PhD in programming to operate.
Automation Requires an 'All-or-Nothing' Approach
Although “everything automated” is an admirable goal, it’s not realistic. We need to take automation in bite-sized chunks and attack areas that will provide the most rate of return in terms of time saved. You can think of this as a building block approach, and it’s why we call it a journey. Several small wins can add up to major victories in a rather short amount of time.
The beauty of automation is that workflows can be thought of as micro-services. Don’t create one massive workflow. Instead, create several small workflows that can be used and reused to create chains of actions targeted at a specific task – provision VLAN, create default gateway, create port channel. Each of these does a specific function and can be used standalone or, chained together, can create a workflow to be used after a server is deployed.
Automation Will Take Your Job
One of the top fears that permeate many organizations regarding automation is the thought of losing jobs to a piece of software. Perhaps it stems from watching robots take over assembly lines. In that case, people were no longer doing the assembly, but new jobs were created to program, run and maintain the robots. Similarly, when it comes to automation of tasks, think about it in terms of assisting IT and not replacing IT. To become more productive, we must embrace new tools that can make life easier and turn the focus from the tactical mundane tasks to more interesting strategic work.
Automation Only Applies to Certain Parts of the Life Cycle
This journey of automation must consider the full life cycle of the data center which includes, provisioning, validating, troubleshooting and remediating; all these functions can reap the benefits of automation. This may sound like we are trying to automate everything, but, we are going to automate those tasks in each step of the life cycle that will be of most use. Any task that is repeated on a somewhat regular basis is a prime candidate. Those tasks that are done once or very infrequently will be left as is.
When it comes to networking, these are exciting times. We all have an opportunity to shape the future of the data center. The way we design, implement and operate is changing, and automation plays a major role in long term success. The use of automation needs to be at your pace, when you need it, where you need it.
Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Informa.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating.