Skip to content

Get started with TIR Block Storage

View as Markdown

Create a Block Storage disk, mount it on a TIR Instance, and write a file to it.

You need a TIR project in Delhi or Chennai, a role that can create datasets and update instances, and an instance to mount the disk on. If you need one, create an instance first.

  1. In TIR, open your project. In the side menu, under Storage, select Block Storage (Disk Dataset).
  2. Select Create Dataset. Before the project has its first disk, select Create Storage on the first screen instead.
  3. Keep the generated Dataset Name or type your own: up to 50 characters, using lowercase letters, numbers, and hyphens, starting and ending with a letter or number. The name becomes the folder the disk appears in on the instance, so choose one you want to read in a path.
  4. Set Disk Size. It starts at 100 GB. A disk can only grow later, and you pay for every GB you set, so start close to what you need.
  5. Select Create.

TIR confirms “Dataset <name> created successfully” and opens Next Steps: Using Disk Storage. Close the panel to see the disk in the list. Its Status reads Pending and then Ready. Wait for Ready before you mount it.

A running instance restarts to apply the mount, so save your work first.

  1. Open your instance’s page and select the Associated Datasets tab.
  2. Tick your disk and select Update.
  3. Tick Are you sure you want to continue updating datasets? and select Update again.

When the instance is Running again, the disk shows Mounted on the Associated Datasets tab.

Open a terminal on the instance, in JupyterLab or over SSH, and write a file to the disk’s folder:

echo "hello" > /datasets/<name>/hello.txt
ls /datasets/<name>

Replace <name> with your disk’s Dataset Name. The folder is read-write, and the file stays on the disk when you unmount it or delete the instance.

For help with a message, see Troubleshooting TIR Block Storage.