Sela. | Cloud Better.

Auto-scaling Azure Pipelines agents with KEDA on K8S (HE)

Gal Hutman, Chief Cloud &  Architect & Haim Dagan, DevOps & Cloud Infrastructure Consultant.

Dive with them into the fantastic tutorial and do it yourself.

 

Did you know you can run Azure DevOps Agents as containers on Kubernetes with the same experience as the Microsoft-hosted Agents?

 

Agents can run on Linux, macOS, or Windows machines and be packaged into containers.

 

When running self-hosted agents on Kubernetes, there is no out-of-the-box support for autoscaling. However, with KEDA you can now auto-scale your self-hosted agents on Kubernetes based on the number of pending jobs in your agent pool.

 

With the Azure Pipelines Scaler, we can now achieve that experience using KEDA (Kubernetes Event-Driven Autoscaler).

 

With the addition of Azure Pipelines support in KEDA, it is now possible to autoscale your Azure Pipelines agents based on the agent pool queue length.

 

Self-hosted Azure Pipelines agents are the perfect workload for this scaler. You can create a scalable CI/CD environment by autoscaling the agents.