Compute
With flexible compute, you can create specific compute resources from your Kubernetes cluster that you can then assign to different GenAI workloads. You can use compute resources to add GPU-capable compute that can significantly speed up the time to train new generators and generate synthetic datasets.
Here is what you can define when you add a new compute.
Compute property | Description |
---|---|
Name | A unique name for the compute. |
CPUs | The number of CPU cores. |
Memory | The amount of memory. |
GPUs | The number of GPUs. |
GPU memory | The amount of GPU memory. |
Toleration | A toleration that matches the taint defined for your Kubernetes cluster |
Order Index | The order in which the compute appears in the Compute drop-down. The default compute has order_index=1 |
Add a new compute
You need to be a super admin to add a new compute.
Prerequisites
Before you define a new compute, check the available compute resources in your Kubernetes cluster. Steps
- From the profile menu, select Compute.
- Click + New compute.
- Define the compute configuration.
- Click Save.
Result
The next time users configure a new generator training or synthetic dataset, they can select the new compute from the Compute drop-down list.
In addition, the new compute remains available in the Computes list.
Edit a compute
You can make changes to exising computes.
Steps
- Open the Compute kebab menu and select Edit.
- Change the compute configuration.
- Click Save.
Result
The compute configuration is now updated.
Delete a compute
You can delete an existing compute from its kebab menu.
Steps
- Open the Compute kebab menu and select Delete.
- Click Yes, delete compute in the confirmation pop-up.