It’s widely recognized that adopting DevOps principles can lead to many business benefits, including more agile and higher quality releases, and better alignment with customer needs. Unfortunately, one of the biggest challenges in DevOps adoption is the cultural divide between teams.
Also, container technology is gaining the attention of many in the dev community, due to the acceleration of continuous deployment and automation, and this doesn’t make the life of operations teams that much easier. I recently attended Container World Santa Clara, and the consensus is that in the next 5-10 years, containers will still be running in parallel or on top of virtualized infrastructures. So, the current or old systems are not retiring at the same pace as the new technologies are added, further increasing complexity.
As a result, as your operations teams are adapting and looking at new technologies, they’re still responsible for running and maintaining the existing systems, and often with many disjointed monitoring systems at their disposal. They’re making sure that with all the automation and the continuous updates, your critical services are still running smoothly.
Don’t re-invent the wheel, innovate instead
In addition to the variety of tools used in DevOps practice, from the popular code repositories such as Git, build, automation and configuration systems such as Jenkins, Puppet or Chef, you often have homegrown monitoring tools written in different languages. These tools may be good enough to report accurately on test results from a particular app delivery component, but they lack comprehensive visibility across the entire app delivery lifecycle, from requirements definition to a production environment.
When we talk about visibility, frequently asked questions are:
- Do you have information about developer code check-ins?
- Which tests were performed?
- Where in your code bugs occur more frequently?
- Is there a particular bug or issue that is directly correlated to more change?
- Who is checking the code most commonly?
- Which builds are breaking?
- Is the new build causing a production problem?
- What is the business impact of a particular code release?
Analytics platforms address these questions and much more. They provide a continuous feedback loop across the entire application delivery lifecycle. Both developers and operations teams can standardize on a single platform, unite different viewpoints and share insights. Now, you can start understanding how much time your developers in code review for a particular issue, or which bugs or builds affected your customers. You can start tracing all elements, from code definition, check-ins, tests, builds, code review and impact on production, as well as application performance in production.
Only when your dev teams start feeling the pain of IT colleagues and your IT team starts getting involved earlier in the development process, cultural gaps between the teams begin to disappear. And finally, your dev teams can focus on business innovation and not on building tools. As a result, your organization can start reaping all the benefits of DevOps adoption and closely aligning with customer needs. Check out Splunk DevOps to learn more about how analytics can help.
What are the typical obstacles you are facing in your DevOps journey and how are you gaining visibility across the application delivery lifecycle?
Stela Udovicic
Sr. Product Marketing Manager
Splunk Inc.