Continuous delivery provider Harness launched the 24×7 Service Guard for its continuous delivery-as-a-service platform today, a machine learning-powered tool for health checking of APM, monitoring and logging tools. According to the company, the tool provides a “safety net” for application services by automating code rollbacks and providing “total operational visibility” into how services run.

”For example, if a microservice is deployed at 9 p.m. but doesn’t indicate a performance problem until 9 a.m. the next day, the Harness Service Guard protects the microservice —  and the developer who created it — by automatically restoring the service back to its previous working version,” Harness wrote in its announcement. “In doing so, developers gain the power of not only deploying their own code into production but also verifying health checks and automating rollbacks on a 24×7 basis.”

Harness explained the platform is based on unsupervised machine learning algorithms designed to process operational data from major APM and log analytics tools, including “KMeans clustering, entropy, Symbolic Aggregate Representation (SAX), and Hidden Markov Models, as well as applying several neural nets to reduce false positives,” the company wrote.

Harness CEO Jyoti Bansal said that the release is based on feedback from their users on how the company’s engineering operations are structured. “They loved our Continuous Verification, and the ability to instantly understand the quality of an artifact released into production, but they wanted us to push our machine learning farther than that,” Bansal said. “They wanted to give their developers complete power over production applications, including the ability to operationalize those applications and automate rollbacks if needed. With 24×7 Service Guard, we’ve combined Continuous Delivery and AIOps into a single, easy-to-use platform that democratizes the power of deployments for the entire engineering team.”