Cloud Run- the rise of serverless and containerization

What if you could get serverless benefits without the vendor lock-in? See how Cloud Run combines container flexibility with serverless operations to solve common pain points.

#1about 2 minutes

Understanding the serverless model and its common challenges

Serverless offers an operational model with no infrastructure management and a pay-for-usage cost model, but common pain points include cold starts and vendor lock-in.

#2about 1 minute

Bringing the benefits of serverless to containerized applications

Cloud Run combines the flexibility of containers with the operational benefits of serverless, allowing you to deploy applications to production in seconds.

#3about 1 minute

Comparing Cloud Run to Cloud Functions and App Engine

Unlike source-based options like Cloud Functions or App Engine, Cloud Run is container-based, offering greater flexibility for HTTP services and async jobs.

#4about 5 minutes

Deploying a simple Node.js application using gcloud CLI

A live demonstration shows how to deploy a containerized "Hello World" application from a local folder to a public HTTPS endpoint using a single gcloud command.

#5about 3 minutes

Breaking down the pay-per-use billing model

Cloud Run's pricing is based on CPU, memory, and request count, with billing calculated in 100-millisecond increments for the duration of active requests.

#6about 2 minutes

Using custom dependencies and binaries in your application

Cloud Run allows you to package any dependency, library, or binary within your container, providing more flexibility than traditional function-as-a-service platforms.

#7about 2 minutes

Managing deployments with traffic splitting and revisions

Cloud Run supports safe, gradual rollouts by allowing you to split traffic between different immutable revisions of your service.

#8about 1 minute

Achieving application portability and avoiding vendor lock-in

By using the standard container format, Cloud Run ensures your application is portable and can be deployed anywhere, from on-premise to other cloud vendors.

#9about 4 minutes

Practical strategies for minimizing and managing cold starts

Reduce cold start latency by using lean base images, warming up a global cache on the first request, and configuring minimum instances to keep containers on standby.

#10about 3 minutes

Understanding autoscaling behavior and its potential risks

Services can scale from zero to thousands of instances automatically, which is powerful but creates a risk of high costs from uncontrolled retries or unwanted traffic.

#11about 5 minutes

Configuring concurrency to optimize cost and performance

Adjusting the number of concurrent requests per container instance is a key lever for balancing performance and cost, as higher concurrency requires fewer running instances.

#12about 1 minute

Using a load balancer for multi-region deployments

Deploy the same service to multiple regions and use a global load balancer to automatically route users to the closest instance, reducing latency.

#13about 4 minutes

Deploying a WebSocket application using buildpacks

A live demo shows how to deploy a real-time whiteboard application using WebSockets, leveraging buildpacks to automatically create a container without a Dockerfile.

#14about 4 minutes

Automating deployments with a custom Cloud Build pipeline

A demonstration of a `cloudbuild.yaml` file shows how to define multi-step build, push, and deploy pipelines for more complex CI/CD automation.

#15about 2 minutes

A summary of Cloud Run's key features and benefits

Cloud Run offers a simple developer experience, auto-scaling, built-in redundancy, and support for modern protocols like HTTP/2, WebSockets, and gRPC.