Connecting to Cloud Storage Buckets
This page introduces Cloud Storage as an option for storing your machine learning data and training output, and describes how to give your Cloud TPU access to the data objects on Cloud Storage.
Before you begin
You need a Cloud TPU service account in order to access a Cloud Storage bucket.
Create a Cloud TPU service account for your project.
gcloud beta services identity create --service tpu.googleapis.com --project project-id
The command returns a Cloud TPU service account with following format:
service-project-id@cloud-tpu.iam.gserviceaccount.com
Follow the instructions found in either the Cloud TPU quickstart guides or Manage TPUs to configure your Google Cloud project and create your Cloud TPU VM and TPU resources.
Write data to Cloud Storage
Console
Go to the Cloud Storage page on the Google Cloud console.
Create a new bucket, specifying the following options:
- A unique name of your choosing.
- Default storage class:
Standard
- Location: The region where you created the Cloud TPU. For more information about regions and TPU availability, see TPU regions and zones.
CLI
Use the
gcloud storage buckets create
command to create a Cloud Storage bucket:gcloud storage buckets create gs://bucket-name --location region
where:
region is the region where you created the Cloud TPU. For more information about regions and TPU availability, see TPU regions and zones.
bucket-name is the name of the bucket you want to create.
Use the
gcloud storage cp
command to write files to the Cloud Storage bucket:gcloud storage cp -r local-data-dir gs://bucket-name
where local-data-dir is a local path to your data. For example:
$HOME/your-data
Give your Cloud TPU access to Cloud Storage
You need to give your Cloud TPU read and write access to your Cloud Storage objects. To do that, you must grant the required access to the Cloud TPU service account used by the Cloud TPU. The following sections show how to find the Cloud TPU service account and grant the necessary access.
Authorize the Cloud TPU service account
The recommended way to authorize the Cloud TPU service account is by using fine-grained access control lists (ACLs). You can also grant broader permissions using IAM permissions.
Using fine-grained ACLs for Cloud TPU (Recommended)
If you store training data on Cloud Storage, the Cloud TPU service account needs read and write permission on the bucket.
Console
Go to the Cloud Storage browser page to view the buckets you own.
Select the bucket whose ACL you want to modify.
Select the Permissions tab.
Select Grant access to add a new permission and type the complete service account name in the New principals edit box.
If you are reading from this bucket, you must authorize the TPU Service Account to read from the resource. Do this by granting the Service Account the
Storage Legacy > Storage Legacy Bucket Reader
role.If you are writing to this bucket, you must authorize the TPU Service Account to write to the resource. Do this by granting the Service Account the
Storage Legacy > Storage Legacy Bucket Writer
role.
CLI
If you are reading from this bucket, grant read permission for the Cloud TPU service account:
gcloud storage buckets add-iam-policy-binding gs://bucket-name --member=serviceAccount:service-account --role=roles/storage.objectViewer
If you are writing to this bucket, grant write permission for the Cloud TPU service account:
gcloud storage buckets add-iam-policy-binding gs://bucket-name --member=serviceAccount:service-account --role=roles/storage.objectCreator
Using IAM permissions for Cloud TPU (Alternative)
If you want to grant broader permissions instead of granting access to each bucket explicitly, you can grant the Identity Access Management (IAM) Storage Admin role to the Cloud TPU service account.
Click the Grant access button to add principals to the project.
Enter the names of the Cloud TPU service account in the Principals text box.
Click the Roles dropdown list.
Enable the following roles:
Project > Viewer
Storage > Storage Admin
Cloud Storage FUSE
Cloud Storage FUSE allows you to mount and access Cloud Storage buckets as local file systems. This allows applications to read and write objects in your bucket using standard file system semantics.
See the Cloud Storage FUSE documentation for details about how Cloud Storage FUSE works and a description of how Cloud Storage FUSE operations map to Cloud Storage operations. You can find additional information about how to use Cloud Storage FUSE, such as how to install the gcsfuse CLI and mounting buckets on GitHub.
Clean up
Disconnect from the Cloud TPU, if you have not already done so:
exit
In your Cloud Shell, delete the Cloud TPU:
gcloud compute tpus tpu-vm delete tpu-name --zone=zone
Verify the VM has been deleted by running
gcloud compute tpus tpu-vm list
. The deletion might take several minutes.gcloud compute tpus tpu-vm list --zone=zone
A response like the following indicates your instances have been successfully deleted.
Listed 0 items
Run the following command to delete the Cloud Storage bucket and its contents, replacing bucket-name with the name of the bucket you created:
gcloud storage rm --recursive gs://bucket-name
What's next
- For more information about creating Cloud Storage buckets and writing data to those buckets, see the Cloud Storage create buckets documentation.
- For more information about service accounts, see the authentication overview.