Workload Cost and Efficiency Analysis
The Workload Cost & Efficiency Analysis metrics dashboard provide detailed insights into the cost and efficiency of your individual AI Workloads and AI Workbench including usage percentages and potential savings.
- Workload Cost: Displays the total cost of a specific workloads.
- CPU: Displays the percentage of CPU resources utilized by the workload execution. Potential savings are displayed by multiplying the CPU cost defined at chargeback setup * unused CPU core hours.
- Memory: Indicates the total memory consumed by the workload, represented as a percentage. This is calculated using the peak memory used during execution, measured in gigabytes multiplied by milliseconds. Potential savings are displayed by multiplying the memory cost defined at chargeback setup * unused Memory GB hours.
- GPU: Displays the percentage of GPU resources utilized by the workload execution. Potential savings are displayed by multiplying the GPU cost defined at chargeback setup * unused GPU core hours.
- GPU Memory: Indicates the total GPU memory consumed by the workload, represented as a percentage. This is calculated using the peak memory used during execution, measured in gigabytes multiplied by milliseconds. Potential savings are displayed by multiplying the GPU memory cost defined at chargeback setup * unused GPU Memory GB hours.
- Overall: Displays the average usage for both CPU and memory. The percentage is calculated as (CPU percentage + Memory percentage) / 2. Potential savings are calculated by adding potential savings from all resources: (CPU Potential Savings + Memory Potential Savings).
These metrics help you identify the under-utilization of resources. High CPU or memory or GPU wastage may suggest the need to reallocate resources, optimize usage, or adjust configurations to allocate fewer resources.
Dashboard color indicators:
- Green: Usage is between 75% and 100%—indicating efficient utilization.
- Orange: Usage is between 25% and 74%—indicating moderate utilization.
- Blue: Usage is between 0% and 24%—indicating low utilization.
