About Cloud TPU reservations

This document explains how to reserve Cloud TPUs for your exclusive use by requesting a future reservation in calendar mode or a long-term reservation associated with a committed use discount.

Reservations are useful for the following cases:

  • Planned or unplanned usage spikes
  • Obtaining high-demand resources
  • Long-running training jobs and inference workloads
  • Workloads requiring a high assurance of capacity

A reservation is one consumption option for Cloud TPU. For more information, see Cloud TPU consumption options.

Choose a reservation type

Cloud TPU offers two types of reservations: future reservations in calendar mode (short-term) and long-term reservations, which are attached to a committed use discount (CUD). Both reservation types give you a high level of assurance that TPUs are available when you need them, for a specified time period. The following table shows the differences between future reservations in calendar mode and long-term reservations:

Future reservations in calendar mode (short-term) Long-term reservations
Duration 1 to 90 days 1 to 3 years
Supported TPU versions v5e, v5p, and v6e All TPU versions
Cost (for more information, see DWS pricing) Up to 30% less than on-demand 30% to 55% less than on-demand
How to request Self-service using the Compute Engine API or the Google Cloud console Manual process through Cloud Sales or your account manager
Committed use discount (CUD) Not supported CUD required

What's next