Quota for CML workloads

An overview of quotas for CML workloads.

In this release, the quota is implemented at user level.
By default, 8 GB memory and 2 vCPU cores are configured for each user.
The above resources will be sufficient for running simple sessions but nor the spark workloads as the executors will not find additional resources.
CML administrators can configure custom quota for the user under Site Administration Page.
If the quota for a user is exhausted, the workload remains in the pending state till the required resources are available.
If the quota for users is modified, it will be reflected when the next workload is submitted.
note
Do not toggle off this default quota once you have enabled the quota management in CML.
note
Do not edit the GPU resources on the user quota tab under site administration when the workspace is provisioned without any GPU resources.
GPU resources can be edited if the workspace is provisioned with GPU resources.