Now let’s look at how much the solution will cost.
Cloud Run is charged based on three factors:
CPU and memory per 100 ms of execution
For example, 1 vCPU and 512 MB of memory for 1 second could be an execution unit. If you assigned 2 vCPUs and 1 GB of memory to a service, it would be charged four times the amount per 100 ms; but if the extra resources mean it completed in 500 ms instead of 1 second, you would only be billed twice the amount.
A cost per request
The first one million requests are free per month; then they are $0.40 per one million requests. Although this is a very small amount, it can add up, and in Chapter 14, we will look at scaling options when you have many million requests.
Network ingress and egress
As with most services in Google Cloud, there is also a charge for network ingress and egress. The requests and responses for this service are small, but again they can add up, and later in the book we will look at options for this too.
Summary
In this introduction to Cloud Run, you have learned how to deploy a service to Cloud Run and how to monitor the performance of the service. You have also seen how to tweak the performance of the service by changing the number of CPUs and the concurrency of the service, but this is not always straightforward.
For this project, you used the following Google Cloud services directly:
- Cloud Run is used as the container runtime to run the containerized service.
- Cloud Logging is used to provide logging for the service.
- Cloud Storage is used to retrieve the previously stored list on tags.
Behind the scenes, you have also been using Cloud Build and Artifact Registry to build and store the container images, respectively, and we will discuss these later.
In Chapter 7, you will step up the complexity again and look at another Cloud Run service that is used to provide a REST API and is backed by a relational database.