Docs

Torch Object Storage

Task processing

Multi-file compression

TOS allows you to create a task that compresses one or more files into an archive. After compression, you can store the archive in a specified bucket for management and transfer. This topic covers notes, billing details, and steps for multi-file compression.

tip

Multi-file compression is currently in closed beta. Please contact your account manager for access.

Prerequisites

Before you begin, you must first grant TOS permission to read from and write to the relevant buckets. If you have enabled server-side encryption for the destination bucket, you must also grant TOS permission to access Key Management Service (KMS). TOS will use KMS keys to encrypt files.

The first time you use the multi-file compression feature, navigate to the Multi-file package compression tab on the Task processing page. Then, complete the following steps:

Click Click to Authorize under Grant read/write permissions to give TOS read and write permissions to the bucket.

Click Click to Authorize under Grant KMS permissions to give TOS permission to access KMS.

Notes

Category	Description
QPS	The default multi-file compression quota is 5 QPS per account in each region. Requests exceeding this quota are throttled, which may result in failed task creation. Reduce the request rate and try again. To increase the QPS quota, submit a ticket to technical support.
Compression format	TOS only supports the ZIP format and the Deflate compression algorithm. It does not support other compression formats or algorithms.
File restrictions	Number of files: The total number of files you want to compress must be less than 1,000. File size: The total size of the files you want to compress must be less than 200 GB. Storage type: Files in Archive, Cold Archive, or Deep Cold Archive cannot be directly compressed. Restore these files before compressing them.
Others	TOS cannot generate encrypted archives.

Billing notes

Using this feature incurs fees for file compression, requests, and data retrieval. Specifically:

Billable item	Description
Multi-file compression	You are charged based on the size of the files before compression. For more information, see Data processing.
Requests	TOS sends GetObject requests when reading source files. When writing an archive, TOS sends PutObject requests. You are charged for these requests. The cost of a request depends on the storage type of the files being requested. For more information, see Requests.
Data retrievals	You are charged based on the amount of data retrieved when you compress files in Infrequent Access Storage or Flashback Archive. For more information, see Data retrievals.

Steps

In the left navigation pane, click Buckets. In the Buckets list, click the bucket you want to manage.

In the left navigation pane, select Data processing > Task processing.

On the task processing page, select the Multi-file package compression tab. Then, click Create task.

In the Create task dialog box, configure the following parameters.

Parameter	Description
Main task type	File processing is selected by default.
Subtask type	File compression is selected by default.
Source file type	Specify which source files you want to compress. Options: File: Select individual source files. Folder: Select an entire folder as the compression source. You can only select one folder per compression task.
File path	If you select File as Source file type, select the files you want to compress. You can only select files in the current bucket.
Folder	If you select Folder as Source file type, select the folder you want to compress. You can only select a folder in the current bucket.
Directory flattening	Specify the directory structure of the files in the archive. Options: Preserve directory structure: TOS preserves the original directory structure of the source files. For example, if a source file is stored as "test/aaa.jpg", its path in the archive will be the same. Flatten all directory structures: TOS removes the original directory structure and places all files in the archive's root directory. For example, assume that the "test" folder has the following structure: test/aaa/music.test test/bbb/data.text If you set the archive name to "aaa.zip", select Flatten all directory structures, and compress the folder, the generated "aaa.zip" archive will have the following structure: music.test data.text
Compression type	zip is selected by default.
Target bucket	Select the bucket where the archive will be stored. tip Currently, you can only store the archive in the same region as the source bucket. Therefore, select a destination bucket in the same region.
Target path	Specify the path in the destination bucket where the archive will be stored.
Target file name	Enter the name of the archive. The name must end with ".zip".

Once you have finished configuring, click OK.

After creating the task, verify that the archive appear in the destination bucket. If the archive is present, then the compression task was successful.

Last updated: 2026.10.19 11:17:30