Platform For AI:Upload and download files

File transfer station

The file transfer station is available when you upload or download files in Terminal and JupyterLab. It quickly uploads large local files to a DSW instance at no extra cost. A single upload allows you to share the file across multiple DSW instances under the same RAM user.

In the upper-right corner of the page, click File Staging to view the transfer list or manage files in the file transfer station.

1. Click to view the transfer list in the file transfer station.

2. Click to the right of a file to perform the following actions:

   • After a file transfer completes, the system automatically saves the file to the instance. To use the file in another instance under the same RAM user, go to the destination instance and click Save to instance.

   • Files in the file transfer station expire after 7 days by default. To extend a file's validity period, click Extend before it expires.

   • Click Delete to remove the file from the file transfer station. This action does not delete the file from the instance.

     

In addition, the file transfer station has the following limitations:

• Upload quantity: You can upload a maximum of 5 files at a time. Additional files are queued for upload.

• Upload method: Files that are 10 MB or smaller are uploaded directly to the instance through your browser. Files that are larger than 10 MB and no more than 5 GB are automatically uploaded to the file transfer station and then saved to the instance.

• Folder uploads: You cannot upload folders directly. Compress them into an archive file first. For more information, see How to upload and download folders.

• File validity: By default, files are stored for 7 days. You can extend the validity period before a file expires, with each extension adding another 7 days.

• Storage capacity: The total storage capacity is 10 GB for a maximum of 1,000 files. You can manually delete files in the file transfer station.

• Download path: Files to be downloaded must be in the /mnt/workspace/ directory. If a file is not in this directory, copy or move it there first.

Q: How do I upload and download folders?

DSW does not support uploading or downloading folders directly. Instead, compress them into an archive file. The DSW Terminal provides a Linux environment where you can use commands such as tar, gzip, and unzip to compress and decompress files. The following example uses tar.

1. Run tar --version to check if tar is installed. If it is not installed, use the following commands to install it.

   # Installation command for Debian-based systems (such as Ubuntu)
   sudo apt install tar
   
   # Installation command for Red Hat-based systems (such as CentOS and Fedora)
   sudo yum install tar

2. Compress or decompress a folder.

   # Compress a folder. /path/to/directory is the folder to be compressed.
   tar -cvf archive_name.tar /path/to/directory
   
   # Decompress a folder.
   tar -xvf archive_name.tar

Q: What should I do if clicking "Download" does not respond or the download fails?

This problem is usually caused by network congestion or browser-related issues. Try the following steps:

1. Use the Terminal method first.

2. Wait a few moments. Large files may take longer to begin downloading.

3. Switch to a different browser or retry the download in your browser's incognito mode.

4. For large files (such as those over 200 MB) or if your network connection is unstable, we recommend downloading the file via a mounted OSS bucket.

Q: What should I do if I get a "file transfer station is full" error?

The total storage capacity of the file transfer station is 10 GB. To free up space, go to the file transfer station management page and clear files. If the page does not update immediately, refresh your browser.

Q: Why am I always redirected to the file transfer station when uploading a file?

This is expected. To ensure stable and fast uploads, DSW automatically routes files larger than 10 MB through the file transfer station, saving them to your instance after the transfer completes.

Q: How do I upload large local files (such as models larger than 5 GB) or large amounts of data to DSW?

A DSW instance's system disk has limited space and is intended for temporary storage. Avoid uploading large files or datasets directly to the instance. Instead, upload the data to Alibaba Cloud OSS first, and then mount the OSS bucket to your DSW instance.