DevOps Foresight, From Prediction to Recommendation

DZone 's Guide to

DevOps Foresight, From Prediction to Recommendation

Deep learning identifies patterns in release pipelines, gauges the likelihood of software release success, and offers recommendations to improve pipelines.

· DevOps Zone ·
Free Resource

Great speaking with Tim Johnson, Director of Product Marketing, and Sam Fell, Vice President of Marketing at Electric Cloud about their new predictive analytics solution, ElectricFlow DevOps Foresight.

DevOps Foresight applies machine learning (ML) algorithms to the massive amounts of data generated by toolchains to develop a risk score that predicts the outcome of releases before they go to production. Taking predictive analytics even further, it also provides key recommendations on how to improve pipelines based on developer influence, code complexity influence etc.

Image title

ElectricFlow DevOps Foresight provides DevOps teams with the steps to eliminate sources of “release anxiety” by having more insight into existing development practices and release patterns. Bottlenecks and inefficiencies in the software delivery process are easier to find and correct. The ability to understand resource allocation for new and complex application and environment requirements also helps ensure the best path toward successful software releases.

Image title

Based on insights into past patterns of success and failures, DevOps Foresight predicts the likelihood of a releases’ success. Much like a credit score, the creation of a release’s risk score numerical value is based on developer, code, and environment profiles and gives stakeholders an easy, visual way to interpret the likelihood of success for a particular build or pipeline. If the score is high, DevOps teams can look at those profiles to determine what, specifically within those profiles, is driving up the risk.

Image title

In order to make recommendations for improving the pipeline, DevOps Foresight looks at contributing factors and what has helped to improve them in the past and suggests appropriate changes in teams, code or environments.  Managers will be able to proactively answer questions like:

  • Are we going to finish our Release on time?

  • Can we move faster or can we do more?

  • Will this release cause more or fewer quality issues?

  • What’s the likelihood of a production deployment failure?

“Electric Cloud long established great ways for its customers to gain insights into releases that are moving through the pipeline, but it’s been a challenge for organizations to gain insights into releases before they’re finalized or even started,” said Carmine Napolitano, CEO of Electric Cloud. “Improving the pipeline is often based on trial and error or best guesses. What we aim to do with ElectricFlow DevOps Foresight is provide data-driven insights much earlier in the process by looking at past successes, build complexity, author profiles and more, and then recommend ways to improve the pipeline based on facts. I’m proud to say the team at Electric Cloud has built a tool that can save our customers hundreds of hours of work and relieve unnecessary release anxiety for each and every product they deliver.”

You can try ElectricFlow DevOps Insight here.

devops ,predictive analytics ,deep learning

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}