CI/CD / DevOps / Monitoring

CircleCI Insights Helps DevOps Teams Quantify High-Performance

8 Apr 2020 11:22am, by

All engineering teams want to be high-performing, but how do they really know? CircleCI, a provider of DevOps-focused continuous integration and continuous delivery (CI/CD) services, is releasing a new feature to help teams quantify how high performing they really are — or aren’t.

“The thought was we can provide you with visibility and more information about what you are doing on CircleCI,” said Kunal Jain, CircleCI senior product manager. “Imagine you are running a workflow or a pipeline on CircleCI. You have a lot of info on different information in the stack. Insights API gives info to the user about the status of all the divergent tooling running on CircleCI and credits usage info.”

In the world of increasingly distributed systems, Jain says not only can you monitor your times and your status, but you can watch what you’re getting back from internal and external endpoints. Once CircleCI switched to this credit-based system, its customers continued to crave more information to better make decisions and optimize their spend.

“To give you visibility, to give you more insight. Do you want to buy more compute so your things run faster or do you want to optimize them?” He said, adding that “It just goes back from our platform perspective to give all that data back to the users.”

The Circle CI Insights API gives customers all they need to know per job.

Jain gave the example: “I have to run a test suite, and that takes 15 minutes in a Docker medium machine and if I switch it to a Docker small it takes 16 minutes and I could optimize my credits spent by reducing it by 10%.”

The data sent back via the four endpoints of the Insights API should help users optimize decision making.

These metrics are based on Google’s Jez Humble’s four key DevOps success metrics, for each workflow and every job that you are running within the CircleCI platform:

  • Success rate
  • Failure rate
  • Throughput
  • MTTR (mean time to restore)

All with the CircleCI credit spend overlaying it to make sense of both the efficiency and expense of their CI/CD pipelines. Jain calls it a “birds-eye view” that allows better decision-making.

One of the company’s clients, e-commerce provider Bolt has been finding value in the new feature. “We’ve built a dashboard using CircleCI’s Insights endpoints to get a full view of our integration tests. We currently have it configured to show the health of our main workflows and repos, which is helpful to pinpoint potential problems. I look forward to building more use cases with these endpoints over time,” said Bolt software engineer Roopak Venkatakrishnan, in a statement.

Jain said that they are thinking about more sophisticated ways to make changes when anomalies are detected, including future automated fixes on the horizon.

The roadmap looks like it’ll be well-tended to since CircleCI just announced this week another $100 million in Series E funding.

CircleCI is a sponsor of The New Stack.

A newsletter digest of the week’s most important stories & analyses.