Set up TPU internal IP addresses

Overview

When you set up a Cloud TPU using the gcloud compute tpus create command or the Cloud Console, each Cloud TPU instance must have an internal IP address range (--range) associated with it. This document provides the information you need to set up the internal IP address range.

IP address range

  • The IP address range must be from within the internal IP address ranges:
     10.0.0.0        -   10.255.255.255  (10/8 prefix)
     172.16.0.0      -   172.31.255.255  (172.16/12 prefix)
     192.168.0.0     -   192.168.255.255 (192.168/16 prefix)
  • Your range must be one of the following formats:

     W.X.Y.Z/N
     W.X.Y.Z 
    where: W,X,Y,Z are integers in the range 0-255 with no leading zeros. If you specify the block size, /N, you must pick a size that best fits the accelerator type that you are requesting. The best practice is to not specify /N and allow Cloud TPU to compute the suffix for you.

    Examples of valid Cloud TPU instance IP address ranges are 10.0.3.0 and 172.31.0.0.

  • You must select a range that does not conflict with another network resource on the network that you are using. For example, if you deploy your Cloud TPU on the "default" network that is created on standard Compute Engine projects, then this network will already have a subnetwork for us-central1 using the 10.128.0.0/20 block and you cannot place any of your subnetworks inside this block. This is true for all other regions.

Was this page helpful? Let us know how we did:

Send feedback about...