Start the Virtual Machine

Virtual machines (VM), which GCP calls Compute Engines, are like computers hosted elsewhere that a user logs into remotely to use. Like any computer, it needs to be booted up before it can be used.

Step 1. Navigate to the VIAME Cloud Dashboard in a web browser

This VIAME Cloud project contains both a development (dev) and a production (prod) environment. As the names imply, dev is for developing, testing, and experimenting with new models or methods, while prod will contain the most recent version in operational use. Each environment has its own dashboard page:

Note

Only those with valid permissions will be able to view these pages.

Step 2. Boot the VM

Under the Compute Engine panel, click “ Go to Compute Engine”

On the page that opens, click “View instances” in the upper left.

The page that opens will show two VM instances:

  • sefsc-viame-web-…: This is a standard VM used for common data processing tasks or other non-resource-intensive processes like running machine learning (ML) models that have already been trained.
  • sefsc-viame-worker-…: This is a high-performance VM with a GPU and is only used for training machine learning (ML) models. Do not use it for anything else. It is too expensive to use for common tasks.

Select the check box next to the VM you wish to start. Only use the worker instance for resource-intensive processes. You can select both if needed, but again, only use the worker VM if needed.

Click START / RESUME

Click “START” on the verification popup. You will see a black “Starting <instance-name>…” banner at the bottom of the browser window while the VM is booting, where <instance-name> is the name of the VM instance. When it is ready, the banner will change to “VM instance started.”

3. Verify that the VM is running

There should now be a green check mark next to the running VM. If needed, refresh the dashboard using the blue REFRESH botton on the dashboard itself, as shown in the screenshot above (as opposed to the reloading the page in the browser.)