Simulation
Toggle minimum instances and incoming traffic to see how Cloud Run balances idle cost against request latency.
Change the inputs and watch how the outcome changes.

Toggle minimum instances and incoming traffic to see how Cloud Run balances idle cost against request latency.
With minimum instances at its default of 0, Cloud Run removes the last container instance when the revision receives no traffic. You are not charged for CPU or memory while the service is unused, which removes idle cost during quiet periods.
Starting scenarioEvery input at its default
With minimum instances at 0 and cold starts acceptable, Cloud Run instantaneously scales out container instances to absorb the spike, and you pay only for the time code runs. The first requests absorb a brief startup delay, which keeps cost minimal.
What changed
With minimum instances set to 1, the pre-warmed instance is already running when the spike arrives, so initial requests are served with no cold-start delay. This trades the continuous idle cost of the warm instance for meeting strict latency targets.
What changed
Because cold-start delay is now unacceptable but minimum instances is still 0, the first requests during the spike must wait while Cloud Run provisions a container from zero. That startup delay violates strict latency requirements even though cost stays at pay-per-use.
What changed
Because you changed minimum instances from the default of 0 to 1, Cloud Run keeps one container instance running even though the revision receives no traffic. The warm instance is billed continuously, trading idle cost for a faster first response.
What changed
Requests are served instantly because the warm instance is running, but since cold starts were acceptable this performance was unnecessary. Keeping minimum instances at 1 means paying continuous idle charges that scale-to-zero would have avoided.
What changed
Scaling to Zero and Idle Resource Costs
When a team wants to avoid the delay of starting from zero, the minimum instances setting on Cloud Run keeps a specified instance idle or warm. Keeping a warm instance trades some idle cost for faster response and more consistent performance. This is the central cost tradeoff in serverless planning: scale-to-zero minimizes cost, while warm instances trade cost for consistent performance.
Domain 1 of 3
27 practice questions
Practice this DomainLast updated