Quota is defined in terms of Cloud TPU cores. A single Cloud TPU device comprises 4 TPU chips and 8 cores: 2 cores per TPU chip. A Cloud TPU v2 Pod (alpha) consists of 64 TPU devices containing 256 TPU chips (512 cores). The number of cores also specifies the quota for a particular Cloud TPU. For example, a quota of 8 enables the use of 8 cores. A quota of 16 enables use of up to 16 cores, and so forth.
The notation: version-cores, for example v2-8, indicates the Cloud TPU version and the number of cores. Since the number of cores are also used to specify quota, this notation also describes a Cloud TPU quota allocation. For example, v2-8 indicates a v2 configuration with a quota of 8 cores.
When you create a new Google Cloud Platform project, Cloud TPU allocates a default quota to the project.
There are separate quota types: On-demand (non-preemptible), and preemptible TPUs.
Quota for single Cloud TPU device configurations
- On-demand TPUs: Default quota is at least 2 TPUs (16 cores).
- Preemptible TPUs: Default quota is at least 6 TPUs (48 cores).
Quota for Cloud TPU v2 Pod (alpha)
- On-demand (non-preemptible) Cloud TPU Pods: Quota is available by contacting sales through the TPU quota page. You need to request Cloud TPU v2 Pod cores per project per region.
Requesting additional quota
From the Service pulldown, select the Cloud TPU API service. You must update the TPU cores per project per region. Available quota types are:
- TPU v2 cores per project per region
- Preemptible TPU v2 cores per project per region
- TPU v3 cores per project per region
- Preemptible TPU v3 cores per project per region
- TPU v2 Pod cores per project per region
Click Edit Quotas to start the quota increase process.
You will receive a notification when the additional quota is approved. Note: Due to high demand, it's possible that some requests for additional quota will not be approved.