Help Center/ Scalable File Service/ Best Practices/ Using OBS+SFS Turbo for Storage Acceleration in AI Scenarios/ Resource Planning and Costs
Updated on 2025-06-05 GMT+08:00
View PDF
Share

Resource Planning and Costs

The following table describes the resource planning and costs in this best practice.

Table 1 Resource planning and costs

Dimension

Description

Resource planning
  • OBS: provides the bucket for storing data from training datasets and pre-trained models. Set the bucket storage class to Standard and bucket policy to Private.
  • SFS Turbo: Select the HPC file system type. Select the type based on the storage capacity and performance required. You are advised to select 250 MB/s/TiB or higher in AI scenarios.
  • ModelArts: An AI development platform that uses multi-node multi-card distributed training.
  • VPC: provides the VPC and subnet.
  • Algorithm and data: prepares the algorithms and datasets required for AI training, such as the Swin-Transformer algorithm and ImageNet21K datasets.
NOTE:

To achieve optimal acceleration performance, you are advised to select the same region and AZ for the SFS Turbo HPC file system and ModelArts resource pool.

Costs
NOTICE:

Prices provided are for reference only. Refer to the prices displayed on the Huawei Cloud console.

To know more about this solution or if you have any questions when using this solution, seek support through Solution Consultation.