Progressive Delivery

Progressive Delivery

Actively monitor an app's behavior over time and take action when sub-optimal behavior is detected


Problem:
Today an enterprise healthcare software company uses canary-style deployments to rollout out updates to their applications. Lives depend on the software healthcare providers use during critical patient treatment. It is critical that a new application version behave as expected – with no availability or functionality issues at rollout.

Canary-style deployments typically rely on raw metrics such as HTTP error codes and latency to determine if the deployment should progress or be rolled back. However, monitoring HTTP metrics alone during rollout doesn’t give them a high level of confidence in the stability of their applications, or the ability to verify the new version is behaving as it should.

Goal:
A way to verify a new version is behaving as it should at rollout. Ideally, the app owner is alerted when app behavior is sub-optimal quickly — no matter the cause of the change in behavior. An app owner is alerted something within the production system adversely impacts app behavior.

We are shaping the Aspen Mesh Solution now: 

It’s necessary to know what optimal behavior looks like to know whether the application’s behavior has changed with a new rollout. App behavior cannot be accurately evaluated with low-level HTTP metrics.

Aspen Mesh actively monitors an application’s behavior over time to provide a clear understanding of what valid behavior looks like and takes action when sub-optimal behavior is detected. This applies across production systems, there are alerts for any change in behavior, whether it’s caused by end-user usage patterns or something technically systemic.

There is a lot of untapped telemetry and trace data available in Istio and Kubernetes. Today this data is largely ignored. When you use this data to learn the optimal operational state of an application you can now detect any anomalies in your application’s behavior and react accordingly. Detecting and remediating anomalies allows you to meet your key business goals by constantly assessing your application and reacting to changes in technical or business events.

Request Early Access now  – Get unlimited access for one year and all new features as they are released.

Request Early Access

Sign-up for early access to the Aspen App Intelligence Platform and receive instructions for how to get your account.

Start using 360° App Performance Insight features on Day One.

Learn More and Request Access

Become a Lighthouse User

We’re looking for modern app teams operating at scale who recognize they have to adapt how they work to excel. We want you to share your biggest problems and help us shape features of our SaaS platform.

Aspen Mesh is looking for teams ready to leverage the power of the technologies in their microservices environment to  optimize application performance.

Learn More

Get in Touch

We would like to hear from you.