Reviewing some of my latest DevOps posts. I’ve covered why DevOps is important, asked the question Who Owns DevOps and answered that the CIO needs to own transformation initiatives. In my last post, I stated that agile development teams should stick to owning the practices tied to Development like automating testing and standardizing deployments so that they stay aligned and focused on initiatives that improve customer experiences, automate a workflow or generate revenue

So, what is the role of the Ops team in a DevOps transformation? Can the team responsible for keeping the lights on, resolving incidents under expected SLAs, handling requests to deploy application upgrades, ensuring that business systems are secure, keeping up with the latest patches and upgrades transform their practice with DevOps?

DevOps Practices is really About Agile Ops

If the CIO owns the transformation and Dev is largely providing a supporting role, then the opportunity to transform is really with Ops. Here’s my extended rationale:

• • The real benefit in cloud environments is with enterprises that manage thousands of server instances and petabytes of data and where standardization and automation provide significant cost advantages. New configuration tools like Docker, Puppet, and Chef are designed to automate configuration management to these extremes – but it is Ops that should learn the technology and take on the responsibility of configuring them. Same responsibility (configuring systems) new tools.

• • Agile development teams have already aligned their effort to business-driven priorities, but their frustration is when Ops or the infrastructure can’t support frequent changes. Perhaps Dev needs to spin up new environments to test an upgrade or evaluate a new platform and it takes too long to configure? Maybe releases need to be scheduled more frequently but the deployment steps are too complicated? These are Operational accountabilities that might require contributions from the Dev team. Ops should collaborate with Dev to define the application, configuration, and other changes that will drive the capability to release more frequently.

• • The increase in cloud instances and applications – along with higher expected service levels from customers and stakeholders require an updated approach to monitoring applications and collecting performance data. Applications need many more monitors, real-time alerts, and longer time frame performance trending. Ops teams need to invest in their skills to respond to this greater need.

• • Security is only going to get more business-critical and more difficult to perform. Because Ops is in the front lines protecting security, they are in a better position to drive infrastructure and configuration standards that help simplify the disparity of assets that need secure configurations and support.

• Server loads are increasingly more dynamic especially when many applications must handle variable loads driven by global, big data, or mobile computing needs. Ops are already versed in handling load balancers, application clusters, SANs, SDNs, and other tools to dynamically scale system resources. The additional automation tools might require more “coding” but should not be outside of Ops’ ability to learn.