AI-powered, automated observability and deep visibility into serverless apps running on Google Cloud Run

Cloud Run brings serverless to containers. You can run containers on Cloud Run for Anthos for consistency between on-prem and cloud environments, or in fully managed Cloud Run environments. A managed platform built from Knative, Cloud Run automatically scales your stateless applications, abstracts away all infrastructure management (K8s), and lets you focus on what matters most: building fantastic applications.

Read on to see how Dynatrace monitors the dynamics in your Cloud Run environment and understands and raises the performance of serverless applications through AI and automation.

Run custom-built images as containers at scale with Google Cloud Run

Cloud Run allows you to run custom-built images as containers and enables you to leverage the technology that best fits your needs based on specific problems. However, the broad variety of technologies that can run in Linux containers (such as Java, .NET core, Golang, Node.js) makes it challenging to easily monitor polyglot microservices stacks. The complexity of monitoring Cloud Run environments with heterogeneous stacks increases further when a platform auto-scales deployments based on demand.

Automate white-box monitoring of Cloud Run applications with Dynatrace

Dynatrace AI-powered, automated monitoring supports Cloud Run for Anthos. It provides full observability and visibility into your applications, proactively alerts on issues that affect performance, and helps you confidently manage operations and costs across the environment.

• Easy, out-of-the-box, auto-monitoring for supported application technologies like Java, .NET Core, Golang, Node.js, and PHP
• Auto-monitoring of processes in containers
• Automated code-level visibility and distributed request and service tracing of applications in containers
• Auto-detection of Knative meta-information like serving.knative.dev/service, serving.knative.dev/revision, and serving.knative.dev/configuration.

With the out-of-the-box capabilities of Dynatrace, you can nail down performance issues in your microservices communication in real-time. In the screenshot above, the response time of the Golang-based Frontend slowed down due to an issue in the Python-based RecommendationService.

How to raise the performance of your serverless apps

Dynatrace supports Cloud Run for Anthos by monitoring the underlying Kubernetes environment with the OneAgent Operator. The OneAgent deployed on the nodes automatically monitors the pods and containers that are deployed through Cloud Run (please note that Dynatrace yet only supports Cloud Run for Anthos).

What’s next

We’re continuously extending our capabilities for monitoring Kubernetes-based environments and the workloads that run on them. We’re working on ingesting Kubernetes events and improved deployment/pod level visibility, so stay tuned.

Let us know what you think

What’s your experience with Cloud Run? Do you have any insights you’d like to share with us? Would you like to put Dynatrace to the test to gain insights into your applications? Please share your feedback in the Dynatrace Open Q&A forum or reach out with your questions.