This post is from the Collabnet VersionOne blog and has not been updated since the original publish date.
How to Overcome the Challenges of Optimizing Continuous Delivery Pipelines
As organizations look to optimize their delivery pipelines through continuous delivery practices they run into many challenges. Let’s take a clear look at how these challenges affect your DevOps journey and how you can overcome them. Top 2 Challenges to Optimizing Delivery Pipelines Here are the top two challenges I run into when I visit organizations adopting or maturing DevOps:
- Manual Processes Dependent on Outside Stakeholders The dominating challenge I see are the manual processes that involve stakeholders outside of the core delivery team. Manual processes are of course less efficient than automated processes but when you throw in dependencies on actions from outside stakeholders, you really slow things down.
- Over/Engineered Automated Processes The next most complex interaction tends to be an automated process that is over/engineered or poorly planned and constructed. You can probably think of one in your organization. Often times simplifying the over/engineered process requires the same level of thought, care, analysis, and execution as it took to build.
How to Overcome the Challenges to Optimizing Delivery Pipelines What’s are the steps to successfully overcoming these challenges?
- Document It! Although Agile and DevOps teach us to focus on working software over comprehensive documentation, you can’t improve a system that you don’t understand. We aren’t talking about creating a planning document about a nebulous prediction of what we think we are going to build; we are talking about mapping a system that has already been built.In order to effectively manage and reduce the complexity of systems that tend to be set and static, we have to get something written down so we can create a plan to improve it.
- Make a Plan Identify the steps in your value stream that have the longest wait times. How you ask? Well I am glad you did. After all, this is a VersionOne blog and I would be remised to not mention that our solution, Continuum, cannot only help with the orchestration and automation of your delivery pipelines, but also provide the valuable instrumentation you need to find wait times and performance metrics that you will need to first see and then decide how to best optimize your delivery pipelines.When you are ready to begin breaking down the complexity of your DevOps machine, start by measuring it instead of sticking your finger to the wind to decide where to go. After all, the right tools to navigate are a compass and map, if you had access to them, why wouldn’t you use them to plan, navigate, and find your end point?When you start with VersionOne Continuum you get the instrumentation you need navigate the journey of breaking down complexity in your devops machine, and delivering value faster and more efficiently. You will also get the added benefit of data driven measurements that enable you, your team, or your management to make objective investment decisions that support delivery efficiency.
Conclusion I hope this has ignited a burning desire to revisit your DevOps machine documentation and revise a plan to optimize it. Learn more about measuring DevOps to provide objective continuous improvement in this video of a recent lightening talk I gave at DevOpsDays Charlotte. [embed]https://youtu.be/KvGBgJHTnWE[/embed]