A growing number of organizations have bought into the philosophy of DevOps as a way to deliver business value faster and more frequently and also experienced pockets of excellence in limited business or technical areas. However, considerable gaps remain when it comes to implementing DevOps at scale, ensuring that development and operations teams can be as collaborative and efficient as possible. As we move into 2019 with predictions that DevOps is at the crossroads of massive changes and facelifts, and with the focus on integration at the edge between services, it is essential to look at some of the missing links that are hindering enterprise-wide implementation.
In this blog post, we summarize the key focus areas and approaches to meet the challenges of modern DevOps environments and center the focus of DevOps teams for large-scale success.
Visibility of work
DevOps is essentially about continuously facilitating the flow of work through the value stream in order to deliver business value to end-users faster and more frequently. However, one of the biggest challenges of DevOps today is lack of visibility — especially shared visibility — into application delivery process and outcomes. It stems from the fact that DevOps outfits, in general, still work with fragmented tools for automation. So, a typical DevOps toolchain, for instance, might have 10 to 15 disparate tools that automate different isolated points in the pipeline. The issue is none of these tools can track how value is moving from one step to another in the software delivery process — causing friction within teams and impacting velocity of feature delivery. This becomes especially troublesome when attempting to scale DevOps in large and distributed enterprises — when there are thousands of developers with duplicate tools across distributed geographies, varying processes, and multiple business units.
To regain control, organizations must focus on enhancing visibility throughout the entire delivery pipeline across all projects and releases. “The fact is, you can’t manage what you can’t see. If you don’t focus on proper visibility, unfortunately a lot of time will be wasted on unfocused efforts,” remarked Lance Knight, SVP and GM, ConnectALL in an interview with David Rubenstein, Editor-in-Chief and COO and David Lyman, CEO and Publisher, SD Times at the recent DevOps Enterprise Summit 2018 in Las Vegas. He attested to the fact that while the DevOps community has done a great job of creating best-of-breed point solutions for different kinds of automation, the ecosystem is still fragmented in general. “A platform that integrates these disparate tools and enables legacy information gathering efforts to be replaced with collaboration and governance will provide the much-needed visibility and align individual teams with the business objectives,” believes Lance.
The move to a shared, system-wide parameter for value
When organizations are split into silos, it’s common for each silo to have its own parameter or currency for value, with the differences between these currencies being akin to the cracks in the floor for things to disappear into. For instance, while business teams may speak the language of portfolio items or features, Dev and Ops may use terms like releases, versions, and packages when describing value-add and progress of work. Without a common currency that spans across teams and stakeholders, it can become very difficult to answer questions such as, “I need to test this user story, where can I find it?” or “Has this user story been deployed to production yet?” and measure how value is progressively enriched as it flows from one step to another, all the way through to the end user.
In DevOps implementations, artifacts can be the common parameter or currency for collaboration between teams. Some of the common artifacts used by DevOps practitioners at different stages of the DevOps process can be user story, epic, defect, requirement, test case, feature, ticket, etc. Focusing on how artifacts flow between tools is critical for empowering teams to inform the progress and movement of value and correlate code driven information to business outcomes.
Measurement of success
DevOps is widely understood to be a journey that is based on continual improvement. But if you’re not measuring results, you can’t understand how to expand DevOps throughout your organization. Metrics such as business enablement, cycle times, escaped defects etc. are essential to gauge the success of your initiative, tell you what’s working and what’s not and guide you towards the insight that explains the reason for success or failure of a given approach. It’s important to look at metrics at every stage of a value stream map to really get precise information of how fast value is moving at every phase of delivery, where the bottlenecks exist and what processes need improvement.
“Good measurement of your DevOps efforts is based on the ability to integrate the various tools across the lifecycle, analyze and correlate myriad events, and report on the findings to give managers actionable insights that align with the business needs and outcomes,” according to Lance. Once integrated, you can capture end-to-end flow time metrics to uncover bottlenecks and identify efficiency opportunities to optimize the process.
And this is the core where a Value Stream Integration platform like ConnectALL brings value to the DevOps lifecycle. ConnectALL can tap into the tools and technology that are already in place and then track value as it moves through those tools and processes, without interrupting the DevOps tool stack in any manner, providing enhanced visibility to all stakeholders all the way through to the production deployment. “Why people want to talk to us is because we are able to integrate disparate systems. But more than that, we are a provider of framework, which is a platform for improving visibility, capturing metrics at the CIO level and improving predictability,” remarked Lance when he caught up with Vance McCarthy, Editor and Online Program Director for Integration Developer News at DOES18. “Once your tools are integrated, you can improve the flow of information and drive continuous improvement and feedback for successful DevOps initiatives.”
Truth be told, DevOps has been a bit of a gamble for many organizations so far. While the benefits of putting DevOps into practice in terms of improved speed of delivery and better-quality releases are there for everyone to see, simply moving faster is no longer good enough. How do you ascertain if you are moving ahead in the right direction? Are your teams and processes functioning at the optimum level? Are you confident that the business is focusing on the right projects and initiatives?
If you are committed to scaling DevOps across the entire organization, you need to continuously analyze and assess the value you are receiving from your DevOps tools and practices, enact change as required and foster a culture of continual refinement.
Content Marketer with over six years of experience in planning, creation, distribution and analysis of content initiatives and assets, across various channels and formats. My core content philosophy is to gauge audience sentiment and deliver impactful takeaways in the unique ways that resonate with them.