Local Data Disk

aifare platform provides a local data disk for each instance, offering high-speed storage for training data, models, and temporary files. This document introduces the usage, directory structure, and best practices for the local data disk.

What is the Local Data Disk

The local data disk is a high-performance storage space mounted to each instance, suitable for storing large files, datasets, and files that require frequent read/write operations. Data on the local data disk is retained after the instance is shut down but is not included in custom images.

Main Directory

Directory	Description
/data	Local data disk, high-speed storage

How to Use the Local Data Disk

1. Enter the instance and open the terminal or JupyterLab file manager.
2. Navigate to the /data directory to view or manage your data.
3. You can copy, move, or edit files in this directory as needed.

Example: Save a Dataset to the Local Data Disk

cp dataset.zip /data/

Best Practices

• Store large datasets, training data, and temporary files on the local data disk for optimal performance.
• Do not store important backup files or data that needs to be shared across instances on the local data disk; use the public network drive instead.
• Regularly clean up unnecessary files to free up space.

Notes

• The local data disk is persistent storage; data is retained after instance shutdown.
• Data on the local data disk is not included in custom images or shared across instances.
• Please do not store sensitive or confidential information on the local data disk unless necessary.

For more information, please refer to the aifare platform documentation or contact customer support.