This document explains the different methods that you can use to request capacity in Capacity Planner. To learn more about the features and use cases for Capacity Planner, see instead Capacity Planner overview .
To help ensure that your Google Cloud project, folder, or organization has sufficient capacity for expected or unexpected growth, you can request capacity for a future date and time. Based on your requirements for cost, assurance level, and resource type, you can use either capacity requests (for best-effort assurance) or future reservation requests (for high-level assurance).
Comparison of methods to request capacity
The following table summarizes the key differences between the methods that you can use to request capacity in Capacity Planner. To help you choose the method that best fits your workload's needs, you can also use the flowchart in this document.
Ideal for indicating potential needs of a large number of resources, such as in the following cases:
- Event-based planning
- Migrations to Google Cloud
- Product launches and expansions
- Workloads with changing requirements
Ideal for obtaining capacity with a high-level assurance, such as in the following cases:
- Business-critical workloads
- Peak scale events
- Large-scale migrations
- Compliance requirements
By using capacity requests, you can request resources such as the following:
- BigQuery
- Cloud Storage
- Compute Engine instances
- Disks
- Google Cloud networking products
- Google Cloud VMware Engine
- GPUs
- TPUs
- Vertex AI
For a full list of supported resources, contact your assigned technical account manager (TAM).
- Compute Engine instances
- GPUs
For capacity requests, you incur charges as follows:
- You pay as you go (PAYG).
- Commitments are available based on the type or resources that you request. For more information, contact your TAM.
For future reservation requests, you incur charges as follows:
- You pay for the reservation period. For details, see reservations billing .
- 24 hours before the start of your reservation period, you can optionally attach your reserved capacity to a commitment .
- Create a capacity request by contacting your TAM .
- If Google Cloud approves your request, then, at your chosen start date and time, you can use the requested capacity to create resources.
- Create a future reservation request .
- If Google Cloud approves your request, then, at your chosen start date and time, you can use the reserved capacity to create instances. For more information, see Consume an automatically consumed reservation .
Choose a method to request capacity
Use the following flowchart to choose the method to request capacity that best fits your workload needs:
The questions in the preceding diagram are the following:
-
Do you need resources for creating compute instances or GPUs?
-
Yes: Go to the next question.
-
No: Use capacity requests .
-
-
Do you need a high-level of assurance for your requested resources?
-
No: Use capacity requests .
Use capacity requests
Use capacity requests to inform Google Cloud that you need a large number of resources across multiple regions or zones. After you create a capacity request, the time that Google Cloud takes to review it varies based on the type and number of resources that are specified in your request. If approved, then Google Cloud schedules the provisioning of your requested resources on your chosen date and time. However, because the resources obtained through capacity requests are subject to real-time availability, Google Cloud might provision only part, or none, of your requested resources. You pay only for the provisioned resources that you use.
For more information about the requirements and limitations that you apply when you create capacity requests, see About capacity requests .
Use future reservation requests
Use future reservation requests to reserve resources in a single zone. After you create a request, Google Cloud typically reviews your request within five days. If approved, then Compute Engine creates on-demand reservations with your requested capacity on your chosen date and time. You have exclusive access to your reserved resources during your reservation period, preventing others from using them. You also commit to pay for the reserved resources for the entire reservation period, whether you use the resources or not.
For more information about the requirements and limitations that you apply when you create future reservation requests, see About future reservation requests .
