Retrieving data stored in the STANDARD_IA, MAZ_STANDARD_IA, ARCHIVE, MAZ_ARCHIVE or DEEP_ARCHIVE storage class incurs data retrieval fees, which will be charged by the size of the retrieved data.
Note:
For more information on storage classes, see Overview. Data Retrieval Fee Billing Items
STANDARD_IA/MAZ_STANDARD_IA Data Retrieval Fees
|
Data stored in MAZ_STANDARD_IA or STANDARD_IA can only be read/downloaded after it is restored. The data retrieval fees are calculated based on the size of the object. | STANDARD_IA Storage MAZ_STANDARD_IA | Pay-as-you-go |
Attention:
If you configure a lifecycle policy for low-frequency storage data (such as setting low-frequency storage data to settle into archive storage or DEEP_ARCHIVE), retrieval fees and request fees will occur during the execution lifecycle policy.
ARCHIVE / MAZ_ARCHIVE / DEEP_ARCHIVE Data Retrieval Fees
|
For Archive Type data (such as ARCHIVE, DEEP_ARCHIVE), the data cannot be read or downloaded before it is restored (thawed). To access and download such data, you must first restore it as a STANDARD storage class copy. In this case, data retrieval can also be referred to as data thawing (the process of restoring archived data to standard data).
Fees are categorized based on different archive types (such as ARCHIVE, DEEP_ARCHIVE) and retrieval modes as follows: ARCHIVE Expedited Retrieval Fees ARCHIVE Standard Retrieval Fees ARCHIVE Bulk Retrieval Fees DEEP_ARCHIVE Standard Retrieval Fees DEEP_ARCHIVE Bulk Retrieval Fees | Archive storage MAZ_ARCHIVE Deep archive storage | Pay-as-you-go |
Billing Mode and Calculation Formula
|
Pay-as-you-go | STANDARD_IA Data Retrieval Fees MAZ_STANDARD_IA Data Retrieval Fees ARCHIVE Storage Data Retrieval Fees MAZ_ARCHIVE Storage Data Retrieval Fees DEEP_ARCHIVE Data Retrieval Fees | Daily billing cycle Data Retrieval Fees = Unit Price per GB x Daily Data Retrieval Volume |
Attention:
Standard storage: No need to retrieve data, so no data retrieval fees are generated.
Infrequent storage: Data needs to be retrieved with no waiting and fast reading. Data retrieval fees are billed based on usage.
ARCHIVE: Data needs to be retrieved, supporting three retrieval modes with varying duration and fees. Data retrieval fees are billed based on usage.
Expedited retrieval mode: Recovery tasks can be completed in 1-5 minutes.
Standard retrieval mode: Recovery tasks are completed within 3-5 hours.
Bulk retrieval mode: Recovery tasks are completed within 5-12 hours.
DEEP_ARCHIVE: Data needs to be retrieved, supporting two retrieval modes with varying duration and fees. Data retrieval fees are billed based on usage. Standard retrieval mode: Recovery time is 12-24 hours.
Bulk retrieval mode: Recovery time is 24-48 hours.
INTELLIGENT_TIERING: Only expedited retrieval for the archive tier of intelligent tiering storage is charged, consistent with archive storage expedited retrieval. Retrieving data from other storage types is free of charge. Data retrieval fees are charged based on the actual data amount retrieved, determined by the actual completion time of the retrieval task, not the request initiation time.
Example:
User A triggered a standard retrieval request for 10GB deep archive data on September 14, 2025. The task was completed on September 15, 2025. Later, User A initiated another standard retrieval request for 5GB deep archive data on September 15, 2025, and the task was done on September 16, 2025.
Since data retrieval fees are settled daily, the payments generated by the above two retrieval tasks were deducted on September 16, 2025 and September 17, 2025, respectively, corresponding to the actual data retrieval completed on September 15, 2025 and September 16, 2025.
Data Retrieval Pricing
Specific prices are subject to the real-time list prices published by the Tencent Cloud Pricing Center. Please go to the Product Pricing Center to view the latest unit prices and cost estimates. Details are as follows: Resource package Pricing: Not applicable (the data retrieval feature does not support resource package deduction).
Billing Examples
Note:
Prices in the example below are for reference only. For the actual prices, please see COS Pricing. The storage capacity is calculated in binary, for example, 1 TB = 1024 GB.
Example: STANDARD_IA storage usage fees + STANDARD_IA data retrieval fees + STANDARD_IA request fees + Public downstream traffic fees
Assume that on November 1, 2020, user B uploaded 5 GB of data to a bucket residing in Guangzhou region in the STANDARD_IA storage class, generating 100 requests. On November 2, user B read the data over the public network with CDN disabled, generating 100 requests. Apart from these operations, user B did not perform any other operations. As fees for a day were settled the next day, then:
STANDARD_IA storage usage fees: Settled daily starting from November 2, 2020.
STANDARD_IA Data retrieval fees: Settled on November 2, 2020.
STANDARD_IA request fees: Settled on November 2 and 3, 2020.
Public network downstream traffic fees: Settled on November 3, 2020.
The fee analysis is as follows:
STANDARD_IA storage usage fees = 0.018 USD/GB/month / 30 * 5 GB * 30 days = 0.09 USD
STANDARD_IA data retrieval fees = 0.002 USD/GB * 5 GB = 0.01 USD
STANDARD_IA request fees = 0.01 USD/10,000 requests x 100 requests / 10,000 x 2 = 0.0002 USD
Public network downstream traffic fees = 0.1 USD/GB x 5 GB = 0.5 USD
Based on the analysis above, the total bill for user B in November is calculated as follows: 0.09 + 0.01 + 0.0002 + 0.5 = 0.6002 USD.