GCP Basics · Module 07
Containers & Serverless – Cloud Run, Cloud Run Functions, GKE and Cloud Build
Run containers on Cloud Run with revisions and traffic splitting, react to events with Cloud Run functions and Eventarc, operate GKE Autopilot and Standard clusters, build and store images with Cloud Build and Artifact Registry, and choose the right compute option for each workload.
What you will learn
- Deploy Cloud Run services and jobs, split traffic between revisions and secure access
- Trigger Cloud Run functions from events with Eventarc and schedule work with Cloud Scheduler
- Choose between GKE Autopilot and Standard and connect with kubectl
- Build images with Cloud Build and store them in Artifact Registry
- Choose between Compute Engine, GKE, Cloud Run, functions and App Engine
Lessons
- Cloud Run: Services, Jobs, Revisions and Traffic
Cloud Run runs stateless containers that scale from zero; every deploy creates an immutable revision, traffic can be split between revisions, and jobs run containers to completion.
- Cloud Run Functions, Eventarc and Cloud Scheduler
Cloud Run functions run small pieces of code in response to HTTP requests or events delivered by Eventarc from Cloud Storage, Pub/Sub, audit logs and more; Cloud Scheduler triggers work on a cron schedule.
- GKE: Standard vs Autopilot
GKE runs managed Kubernetes; Autopilot manages nodes for you and bills per Pod, Standard gives full node control; use regional clusters, private nodes, release channels and Workload Identity Federation for GKE.
- Artifact Registry and Cloud Build
Cloud Build runs build steps in containers from a cloudbuild.yaml or a trigger; Artifact Registry stores Docker images and language packages with IAM, vulnerability scanning and cleanup policies.
- Choosing Compute: VMs, GKE, Cloud Run, Functions or App Engine
Pick compute by how much control you need and how the workload behaves: Compute Engine for full control, GKE for Kubernetes platforms, Cloud Run for stateless containers and jobs, functions for event glue; App Engine is a legacy option for existing apps.