This epic is to track the requirements and work needed to support consumption-based pricing for GA.

Requirements:
1. P0: The system must enable RH and customers to view real-time metrics for total usage across all pods in the cluster and at the account level per billing period (eg. monthly). Capture metrics to support consumption-based pricing for vCPU-hours based on requested CPUs. Examples:

• User runs Small NB session for 2 hours. vCPU-hours = 1 requested vCPU * 2 hours = 2 vCPU hours.
• User runs Large NB session for 3 hours. vCPU-hours = 7 requested vCPUs * 3 hours = 21 vCPU hours.

2. P0: Capture metrics to support consumption-based pricing for vCPU-hours based on vCPU limits. Examples:

• User runs Small NB session for 2 hours. vCPU-hours = 2 vCPUs * 2 hours = 4 vCPU-hours.
• User run Large NB session for 3 hours. vCPU-hours = 14 vCPUs * 3 hours = 42 vCPU-hours.

3. P0: Capture metrics to support consumption-based pricing based on notebook pod hours. In this case, the usage time is independent of resources consumed (eg. usage time is the same if user runs a small NB session or large NB session).

4. P0: Capture metrics to support consumption-based for GPU-hours based on requested GPUs. Examples:

• User requests 2 GPUs for NB session that lasts 2 hours. GPU-hours = 2 * 2 = 4.