Is DevOps dead?

In certain circles, DevOps has become a dirty word – an outdated, ‘of the minute’ trend that was banded about in the tech world without anyone having a solid idea of what it means. It’s very easy to say you’re doing DevOps, but often, everyone is on a different page.

For many, DevOps has stopped being something to shout about – it’s what we expect as a minimum and it’s business as usual. In a fast-paced digital world, taking months to deploy your code or respond to customer feedback is no longer good enough. With the likes of Amazon deploying code an average of every 11.7 seconds, the expectations of customers are shifting.

Having a successful DevOps strategy isn’t just about having the right tools available, or having everyone sit in the same room – it’s about a complete cultural and procedural overhaul. Done right, DevOps can make life much easier for everyone and attract the best talent. Done wrong, DevOps is just another failed experiment that will get teams frustrated and falling back into time-consuming habits.

I’m working with who?!

One of the major obstacles to successful implementation of DevOps is cultural change. Siloes between operational, development, design and security teams should be broken and replaced by a product team, requiring a redefinition of roles and responsibilities. These teams should have cross-functional skillsets, and be small and self-organising. Thus, training and assessing the skills of the workforce is essential.

Whilst a pilot project can be a low-risk way of starting to implement DevOps, scaling this strategy can be difficult and slow. This is a major hurdle preventing large organisations from achieving the agility and speed that allows them to compete with the tech giants.

Create a flow

Improving workflows requires coordination between application delivery and backend infrastructure. Standards for each phase of the project – building, testing, delivery and monitoring – must be defined and agreed by the whole team. With the right approach to governance, a balance can be struck between flexibility and quality assurance.

Automation is the key to unlocking efficiency through DevOps. Using technologies such as microservice architectures will help to form a deployment pipeline for each service and lower the risk of code changes, whilst using Infrastructure as Code will increase the efficiency and repeatability of the build process.

Tic tac tool

Finally, whilst tooling isn’t the only thing taking you from monthly to weekly deployment, agreeing and using the right toolkit will ensure an efficient workflow. Whilst there are innumerable available tools, agreeing a toolkit that covers release, configuration management, orchestration, monitoring, testing and containerisation will ensure the team is able to provide robust service delivery and adapt to users’ needs in real-time.

The end goal

Of course, the real game-changer is the speed of delivery. Creating an efficient DevOps workflow is pointless if we are not considering the outcomes. But given the speed of change in technology, DevOps is about more than efficient operations – it’s about keeping up with your customers.

Call it what you will – continuous delivery, DevOps, or tech in the modern era – DevOps practices will be the line between those who survive the pace of digital, and those who don’t.

Published by

0sophietaylor

I am a Digital Graduate on the Sopra Steria Digital Consultancy scheme. I'm currently working on DevOps and Infrastructure Management, as well as being involved in projects with our Digital Innovation Team.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.