Skip to content

Working with shared storage

Shared storage is a network file share available to all VMs in your workspace. It is the recommended location for storing data and files that need to be accessible across multiple VMs or that you want to preserve independently of any individual VM.

Why use shared storage?

Unlike your VM's local disk, shared storage is:

  • Accessible from all VMs in your workspace — any user can read and write to it
  • Backed up and archived — shared storage is backed up separately to your VM disk, giving your data an additional layer of protection
  • Persistent — files survive VM deletion. If you delete a VM, anything stored only on its local disk is lost; files on shared storage are not

The shared storage drive is the recommended location for your project data and any outputs you want to preserve between sessions.

Accessing shared storage

Windows

Shared storage is mounted automatically as the Z:\ drive when your VM starts. You can access it from File Explorer, RStudio, VS Code, or any application exactly as you would a local drive.

# R example
data <- read.csv("Z:/your-project/data/demographics.csv")

# Python example
import pandas as pd
df = pd.read_csv(r"Z:\your-project\data\demographics.csv")

If the Z:\ drive is not mounted, it should reconnect automatically on next login. If it does not appear, contact the TVS service desk.

Linux

On Linux VMs, shared storage is mounted under /fileshares/. You can access it from the terminal or any application as you would a normal directory.

ls /fileshares/

# Python example
import pandas as pd
df = pd.read_csv("/fileshares/your-project/data/demographics.csv")

# R example
data <- read.csv("/fileshares/your-project/data/demographics.csv")

Storage limits

Each workspace has a storage limit agreed at deployment or during contracting. If you are approaching your limit or need additional capacity, contact the TVS service desk to request an increase.

Maximum storage limit

The maximum available shared storage per workspace is 4 TB. Requests for increases above your current allocation should be submitted via the service desk — see Contact and support.

Best practices

  • Use shared storage for project data — not your VM's local disk. Data on local disk is lost if the VM is deleted.
  • Keep your workspace organised — use clear folder structures so other workspace members can find files easily.
  • Clean up regularly — remove files you no longer need to keep within your storage allocation.

How data reaches shared storage

Data is not delivered to shared storage by default. There are three routes by which data may be available in your workspace, and which route is used depends on your project setup:

  • Shared storage — your workspace manager may place data directly in shared storage, making it immediately accessible to all VMs in your workspace
  • Airlock — data may be delivered via an airlock import request, which you then download and can copy to shared storage if needed
  • Databricks — if pre-agreed and your workspace has Databricks, data may be available directly in the data mart rather than as files

TVS SDE will advise you on which route applies to your project. See Understanding your data for more on how data is structured and delivered.