Breaking Down Dev and Ops Silos with Communication, Collaboration and Trust
Organizations are now embracing DevOps as the most dependable path of continuously improving application development quality, velocity and efficiency. But it will remain only a theory unless you can successfully re-architect your organization’s Dev and Ops silos to enable better communication, collaboration and trust between teams.
But there’s a big challenge. Dev is all about implementing changes quickly, while Ops is focused on keeping the production environment stable. Unless and until the two groups can reconcile these opposing views and share a common way to think and work, nothing will improve.
This is a massive cultural change to drive. Analyzing your current siloed state of Dev and Ops while envisioning a synergistic DevOps future of continuous improvement can reveal a transformation landscape of seemingly insurmountable obstacles.
In reality, the challenges your Dev and Ops organizations will face are surmountable, though the journey isn’t always smooth. To help you get on your way towards better communication, collaboration and trust under DevOps, here are some steps your organization can take.
1. Generate New Transparency Between Dev and Ops
Start with a meeting of the minds where members of your Dev and Ops teams share their roles, concerns and realities with each other. This helps open conversations around new opportunities for shared responsibilities under DevOps.
Begin talking about using a set of common tools across Dev and Ops. This DevOps toolchain—which should involve all systems, including your mainframe—keeps everyone speaking a “common language.”
2. Shift Ops Left in the Development Lifecycle
Shift your Ops folks left into all facets of the development lifecycle. Early awareness about changes to business applications allows Ops to set alert values and create documentation on the new applications.
This helps build trust, communication and collaboration during the development cycle. You need everyone to be on the same page.
Ops should also be involved in creating the acceptance criteria. After all, it’s their environment in which the new software must run. Design reviews help them keep abreast of scope and design changes.
Sprint plans are also critical. Dev may want to implement 10 changes a day, which could impact production stability. Working together, Dev and Ops can find a compromise that moves projects along quickly while ensuring production is stable.
3. Help Dev Adopt an Ops Mindset
Dev has a role to play in the Ops world as well. Traditionally, performance and capacity were concerns left entirely to Ops, but that is a recipe for division—this is the essence of Dev throwing a project “over the wall” to Ops and moving on to the next thing.
Instead, performance and capacity people should work closely with Dev, sharing metrics and tools with developers so those people can learn coding efficiencies and understand tradeoffs. This will help Dev reduce the impact of poorly designed code on production systems as well as help them set realistic, achievable SLAs.
4. Continuously Improve with the Right Measurements
Finally, management has to look at the risk/reward incentives. If you only reward Dev on speed and lines of code, and Ops on a stable environment, DevOps can’t happen.
New key performance indicators (KPIs) that focus Dev and Ops on the right success metrics must be defined to ensure everyone is continuously improving toward the shared goal of a stable production environment that tolerates and embraces more rapid change. Together, the teams should use these metrics to create an ongoing way to talk and to jointly decide when a package is ready for production.
There’s much more to the process of improving communication, collaboration and trust between your Dev and Ops teams in an effort to achieve DevOps, but these four steps should be helpful. Take one step at a time, addressing issues and concerns as they arise. DevOps can be a game-changer, but only when silos come down and people step outside their comfort zones and expand their worldviews.
To learn more about making the transition to DevOps, read our eBook “How to Ditch Waterfall for DevOps on the Mainframe.”
Latest posts by Denise Kalm (see all)
- Where You Should Be Driving DevOps Automation on the Mainframe - December 3, 2018
- The DevOps Balancing Act of Measurement and Reward - October 9, 2018
- The Evolving Responsibility of Mainframe Application Testing in DevOps - September 6, 2018