Help Center> ModelArts> Billing> Billing Items> ExeML
Updated on 2023-11-30 GMT+08:00
View PDF

ExeML

Description

Model training and inference in ModelArts ExeML jobs use compute and storage resources, which are billed. For details, see Table 1.

  • Compute resource fee:
    • If dedicated resource pools are used for model training and inference in ExeML jobs, compute resources are not billed.
    • If public resource pools are used for model training and inference in ExeML jobs, compute resources are billed.
  • Storage resource fee: fee for storing data in OBS
Table 1 Billing items

Billing Item

Description

Billing Mode

Billing Formula

Compute resource

Public resource pools

Usage of compute resources (vCPUs and GPUs)

For details, see ModelArts Pricing Details.

Pay-per-use

Specification unit price x Number of compute nodes x Usage duration

Dedicated resource pools

Fees for dedicated resource pools are paid upfront upon purchase. There are no additional charges for running ExeML jobs.

For details about fees for dedicated resource pools, see Dedicated Resource Pool.

N/A

N/A

Storage resource

Object Storage Service (OBS)

OBS is used to store the input and output data of training and inference.

For details, see OBS Pricing Details.

CAUTION:

OBS resources for storing data are continuously billed. To stop billing, delete the data stored in OBS.

Pay-per-use

Yearly/Monthly

Creating an OBS bucket is free of charge. You pay only for the storage capacity and duration you actually use.

Billing Example

The following prices are for reference only. For the actual prices, see pricing details for each service.

Example: Running an ExeML job using a public resource pool; billing items: compute resources and standard storage

Assume that you create an ExeML image classification project on April 1, 2023. Data validation is done from 10:00:00 to 10:06:00, image classification is done from 10:06:00 to 11:12, service deployment is done at 11:30:00, and the real-time service is stopped at 12:00:00. A public resource pool is used to run the instance. The compute flavor is Compute-intensive ECS 1-ExeML (GPU) (unit price: $6.93 USD/hour) for model training and Compute-intensive 3 instance (CPU) for service deployment (unit price: $0.23 USD/hour). The billing items are compute and storage resources. The fee for running this ExeML job is calculated as follows:
  • Compute resource fee = Specification unit price x Training job running duration + Specification unit price x Service running duration

    Compute resource fee = $6.93 USD/hour x 0.1 hour + $0.23 USD/hour x 0.5 hour = $0.808 USD

  • Storage fee: ExeML job data is uploaded to or exported from OBS. The storage fee is based on the OBS billing rules.

Fee for running the ExeML job = Compute resource fee ($0.808 USD) + Storage fee

Parent topic: Billing Items