1. Login Log into your your Rapid Access Account at https://cloud.cybera.ca/auth/login/ (Please make sure you logged in using Edmonton Location).
2. Login will give you access to RAC dashboard (Fig-1), where you can see the overview of resource allocation and usage.
3. Select Project > Compute > Instances and Click Launch Instance.
43. Configure an Instance.
Below are mandatory values required to launch an instance:
...
Flavour: Select from below GPU types and flavours
GPU Type | Flavour series | Flavours |
NVIDIA K80 | g1 | g1.large.auto-destruct, g1.medium.auto-destruct, g1.xlarge.auto-destruct |
NVIDIA TITAN Xp | g2 | g2.large.auto-destruct, g2.medium.auto-destruct, g2.xlarge.auto-destruct |
NVIDIA TITAN RTX | g3 | g3.large.auto-destruct, g3.medium.auto-destruct, g3.xlarge.auto-destruct |
NOTE: By default all GPU-enabled instances are self-destruct after 24 hours otherwise extended.
...
You can also configure additional settings like Access and Security, Networking, network ports and Advanced options.
54. After configuring, Click on Launch
Once Launched, RAC will start creating a new instance and Instance status will change from build to active.
...
Log-in to the Rapid Access Cloud dashboard at https://cloud.cybera.ca.
In the left-hand panel under “Compute”, click “Instances”.
Click the name of a GPU-enabled instance you would like to extend the lease on. Note the “Lease Date”; this will initially be set to the time and date the instance was created plus twenty-four hours.
Click the “Submit” button; the Lease Date will be set to twenty-four hours from the time you click Submit.
This tutorial assumes that you have the following already in place:
...
A Rapid Access Cloud Account.
A stable internet connection.
...
Relaunching a Shelved Instance
To relaunch an instance of yours that has been shelved:
Log into your Rapid Access Account at https://cloud.cybera.ca/ (Please make sure you logged in using Edmonton Location).
Select Project > Compute > Instance
- Click on the Disclosure triangle on the far right and choose "Unshelve Instance"
- Wait for the instance to restart
NOTE: If a GPU is unavailable the instance will be unable to unshelve and will return to it's "Shelved Offloaded" state.