tencent cloud

CLS is pay-as-you-go by default for all regions. Users are charged based on their actual storage usage, requests, traffic, and other billable items on a daily basis. For more information on the regions, see Regions and Access Endpoints.

Pricing

For details about pay-as-you-go pricing, see Product Pricing.

Billable Item Description

Traffic fees

Storage fees

1. Standard/IA logs and indexes have different storage prices and are billed separately.
2. Storage is billed based on average usage. Storage usage values are collected on each clock-hour, and averaged per day.
3. Log data and index data are stored based on the storage cycle set by users. After the data expires, the storage fees will not be calculated and the expired data will be deleted at the next clock-hour.
   For example, if the storage cycle is set to 3 days and logs are uploaded on June 15 12:15, the effective storage cycle is from June 15 12:15 to June 18 12:15, and the storage usage of expired data will be deducted on June 18 13:00. (It is not recommended that users modify the storage cycle frequently. The storage cycle should be modified at most once a day.)

• Extending the storage cycle
  When you extend the storage cycle, CLS will store the data for a longer period. For example, if you extend the storage cycle from 7 days to 15 days on June 15 12:30, the data uploaded on June 8 12:00 will be stored until June 23 13:00, instead of being deleted on June 15 13:00.
• Shortening the storage cycle
  When you shorten the storage cycle, CLS will store the data for a reduced period. For example, if you shorten the storage cycle from 15 days to 7 days on June 15 12:30, the data uploaded between June 1 12:00 and June 8 13:00 will be deleted on June 15 13:00.

Calculation fees

Other fees

The service request and topic partition fees have low unit prices and only account for a very small part of the total fees if you use CLS normally. Such fees are mainly used to prevent abuse such as high numbers of meaningless concurrent API calls.

Billing Example

NGINX log query and analysis

Company A's website receives 100 million requests per day, and each request produces an average of a 100-byte log. This results in 100 million logs per day for a total of approximately 9.31 GB. Company A decides to upload the NGINX access logs to the CLS in Beijing region, and store them in the standard storage for 15 days, with the log search, analysis, and alarm features enabled but the data processing, download, consuming, and shipping features disabled.

As you can see from the billable items, the index traffic and storage fees are the major fees for the following reasons:

• Index is the core of log search and analysis. Log data cannot be searched for if there are no indexes. Therefore, the unit price of the index traffic (0.062 USD/GB/day) is higher than that of the write traffic (0.032 USD/GB/day).
• If logs are compressed when uploaded, the write traffic and log storage capacity will be calculated based on the compressed log data volume, while the index traffic and storage capacity will be calculated based on the data volume of uncompressed raw logs. As the log compression ratio generally ranges from 1:4 to 1:10, the index traffic and storage capacity will be about 4–10 times the write traffic and log storage capacity respectively.

Therefore, to reduce the product costs, you can appropriately adjust the index configuration to reduce the index traffic and storage capacity.

For example, the raw log is as follows:

10.20.20.10 ::: [Tue Jan 22 14:49:45 CST 2019 +0800] ::: GET /online/sample HTTP/1.1 ::: 127.0.0.1 ::: 200 ::: 647 ::: 35 ::: http://127.0.0.1/

The log will be captured as the following fields during log collection:

IP: 10.20.20.10
bytes: 35
host: 127.0.0.1
length: 647
referer: http://127.0.0.1/
request: GET /online/sample HTTP/1.1
status: 200
time: [Tue Jan 22 14:49:45 CST 2019 +0800]

• If full-text index is enabled, the index generated by the log will be 172 bytes in size (i.e., size of all field keys and values). The index traffic will be generated, and the storage capacity will be used accordingly.
• If you don't enable full-text index and only enable key-value index for the request and status fields, the indexes generated by the log will be 48 bytes in size (i.e., size of the key and value of request and status, which are 48 bytes / 172 bytes = 27.9% of all fields). In this way, the index traffic and storage capacity will be reduced to 27.9% of those for full-text index. However, you can only use the request and status fields to search for and analyze logs and view other fields in this case.

For more methods to reduce product costs, see Saving Product Use Costs.