Google Cloud Run is Google's fully managed serverless container platform that lets you deploy any containerized workload — HTTP services, batch jobs, or pull-based workers — on Google's global infrastructure without managing servers or clusters. It sits at the intersection of serverless convenience and container portability: if your code runs in a container, it runs on Cloud Run, regardless of language or framework. The key mental model is that Cloud Run manages instance lifecycle for you — scaling from zero to thousands of instances based on traffic, CPU, or events — but every design decision around concurrency, billing, and networking flows from understanding how those instances start, serve, and stop.
What This Cheat Sheet Covers
This topic spans 22 focused tables and 139 indexed concepts. Below is a complete table-by-table outline of this topic, spanning foundational concepts through advanced details.
Table 1: Cloud Run Resource Types
Cloud Run offers three distinct resource primitives, each designed for a different workload shape: HTTP-driven services, task-to-completion batch work, and continuous pull-based processing. Choosing the right resource type is the foundation of any Cloud Run architecture.
| Type | Example | Description |
|---|---|---|
gcloud run deploy my-api --image gcr.io/proj/api | HTTP-driven; receives requests at a stable HTTPS endpoint; autoscales based on traffic and CPU; supports scale-to-zero. | |
gcloud run jobs create my-job --image gcr.io/proj/job --tasks 100 | Runs to completion with no HTTP endpoint; ideal for batch processing, DB migrations, model training; tasks up to 168 hours. |