Cloud Run is a developer-focused, fully managed application platform for running code on Google's scalable infrastructure.
Designed to make developers productive and allow them to focus on solving business problems. As a serverless offering, it abstracts away all infrastructure management, allowing developers to focus on Code.
Cloud Run increases innovation velocity and reduces cost and risk. It makes it an ideal solution for quickly and efficiently deploying web applications, APIs, and event-driven microservices.
It leverages Cloud Build to build container images, and works well together with other services on Google Cloud, so you can build full-featured applications.
Cloud Run has use cases across:
π Public and Private Services: Websites, REST, or GraphQL API.
Streaming with WebSockets, AI agents, private HTTP, or gRPC microservices.
π Data Processing: Process queue messages, event-driven architecture, scheduled scripts, Background processing, Batch data processing.
You can deploy code to Cloud Run Via:
β Container: Deploy a container image.
Supported Repositories: Artifact Registry, Docker Hub
β Repository: Continuously deploy code from a connected source git repository.
Supported Repository: Github
β Source Code: Deploy your code to Cloud Run with a single command.
Supported languages: Python, Node.js, Go, Java, Ruby, PHP, and .NET
β Function: Deploy single-purpose functions.
Supported languages: Python, Node.js, Go, Java, Ruby, PHP, and .NET
There are two main resources for executing workloads in Cloud Run, which are Cloud Run Services and Cloud Run Jobs:
Cloud Run Services is used to execute code that responds to web requests or events.
The services include:
β Serving HTTP traffic under a dedicated endpoint
β Pay when code is processing requests.
β Use out-of-the-box URL with TLS
β Leverage automatic Scaling with minimum and maximum instances.
β Access a dedicated endpoint that is invoked by HTTPS request events, websockets, HTTP/2, and gRPC.
β Empty built-in traffic splitting for gradual rollouts and rollbacks.
β Utilize revision history.
Cloud Run Job is used to execute code that performs work on jobs and exits when finished.
Jobs execute tasks to completion, and these features:
β Do not have an HTTP endpoint
β Have a maximum task timeout
β Are executed manually or on a schedule.
β Run a specified number of tasks up to seven days
β Are paid only while the job is executing
β Have execution history.
A developer team might choose Cloud Run for their application because:
π They want to deploy containerized applications without the burden of managing infrastructure or services.
π Their application experiences variable traffic, and they want automatic scaling (including scaling down to zero) for cost-efficiency.
π They need to run stateless web services, API, or event-driven functions packaged as containers.
Top comments (0)