Compute options for model training and scoring
When you train or score a model or function, you choose the type, size, and power of the hardware configuration that matches your computing needs.
Default hardware configurations
Choose the hardware configuration for your Watson Machine Learning asset when you train the asset or when you deploy it.
|Capacity type||Capacity units per hour|
|Extra small: 1x4 = 1 vCPU and 4 GB RAM||0.5|
|Small: 2x8 = 2 vCPU and 8 GB RAM||1|
|Medium: 4x16 = 4 vCPU and 16 GB RAM||2|
|Large: 8x32 = 8 vCPU and 32 GB RAM||4|
|Extra large: 16x64 = 16 vCPU and 64 GB RAM||8|
Compute usage for Watson Machine Learning assets
Deployments and scoring consume compute resources as capacity unit hours (CUH) from the Watson Machine Learning service.
You can monitor the total monthly amount of CUH consumption for the Watson Machine Learning service on the Environments page.
Compute usage details
The rate of consumed CUHs is determined by the computing requirements of your deployments. It is based on such variables as:
- type of deployment
- type of framework
- complexity of scoring Scaling a deployment to support more concurrent users and requests also increases CUH consumption. As many variables affect resource consumption for a deployment, it is recommended that you run tests on your models and deployments to analyze CUH consumption.
The way that online deployments consume capacity units is based on framework. For some frameworks, CUHs are charged for the number of hours that the deployment asset is active in a deployment space. For example, SPSS models in online deployment mode that run for 24 hours a day, seven days a week, consume CUHs and are charged for that period. An active online deployment has no idle time. For other frameworks, CUHs are charged according to scoring duration. Refer to the CUH consumption table for details on how CUH usage is calculated.
Compute time is calculated to the millisecond, with a 1-minute minimum for each distinct operation. For example:
- A training run that takes 12 seconds is billed as 1 minute
- A training run that takes 83.555 seconds is billed exactly as calculated
CUH consumption by deployment and framework type
CUH consumption is calculated by using these formulas:
|Deployment type||Framework||CUH calculation|
|Online||AutoAI, AI function, SPSS, Scikit-Learn custom libraries, Tensorflow, RShiny||deployment_active_duration no_of_nodes CUH_rate_for_capacity_type_framework|
|Online||Spark, PMML, Scikit-Learn, Pytorch, XGBoost||score_duration_in_seconds no_of_nodes CUH_rate_for_capacity_type_framework|
|Batch||all frameworks||job_duration_in_seconds no_of_nodes CUH_rate_for_capacity_type_framework|
Parent topic: Managing frameworks and software specifications