If most of your files are large static files, enabling Range GETs can help increase the file response speed during origin-pull and improve the large file delivery efficiency.
Range GETs refers to origin-pull based on range requests.
Range is one of the HTTP request headers, which is used to get files in the specified range. You can use a range request to request only partial file content from the server. For example, if a request carries the HTTP header
range: bytes=0-999, the first 1,000 bytes of the file will be returned to the user.
In CDN, after Range GETs is enabled, origin-pull requests will carry the
Range header by default. If the partial file requested by a user is not cached on the node or its cache has expired, CDN will perform Range GETs to pull and cache the requested partial file to the node and return it to the user. After Range GETs is disabled, if the user request doesn't carry the
Range header, CDN will still pull the entire file during origin-pull.
For large files such as APK, audio, and video files, you can use range requests to effectively improve the delivery efficiency of large files, shorten the response time, and reduce the pressure on the origin.
Log in to the CDN console.
Click Domain Management on the left sidebar to enter the domain name management list.
Select the target domain name and click Manage to enter the domain name configuration page.
Click the Origin-pull Configuration tab to view the Range GETs configuration items.
In Range GETs configuration, Range GETs is disabled for all files by default. You can also add multiple custom rules for files as needed. Range GETs rules can be matched by file extension, file directory, and full path.
|You can select All Files, File Extension, File Directory, or Full Path:
All Files: This Range GETs rule applies to all files. It is the default rule and cannot be deleted.
File Extension: This Range GETs rule applies to the specified file extensions.
File Directory: This Range GETs rule applies to the specified file directories.
Full Path: This Range GETs rule applies to the specified file paths.
|Enter the content based on the selected file type:
If Type is File Extension, you can enter one or multiple file extensions separated by ";".
If Type is File Directory, you can enter one or multiple file directories separated by ";", and the entered content cannot end with "/", such as
If Type is Full Path, you can enter one or multiple full file paths separated by ";", such as
/index.html;/test/\*.jpg. The file path supports the * wildcard.
|Range GETs can be enabled or disabled.
Enable: If Range GETs is enabled, range requests are used for origin-pull requests. After Range GETs is enabled, if the user request does not carry the
Range header and the requested files are larger than 4 MB in size, the CDN node splits the origin-pull request into several sub-requests for origin-pull based on a shard size of 1 MB. If the requested files are smaller than 4 MB in size, the CDN node pulls complete files from the origin server. If the user request carries the
Range header, the CDN node uses the
Range header for origin-pull.
Disable: If Range GETs is disabled, range requests are not used for origin-pull requests.
If your files are larger than 4 MB in size, we recommend you enable Range GETs for such files. If only part of your files are large ones, we recommend you enable Range GETs for them through match by file type, file directory, or full path and disable Range GETs for other files.
You can configure up to 20 Range GETs rules. The lower the rule, the higher the priority. When a user requests a file, the file will be matched with rules in sequence by priority, and the rule with the highest priority will be executed preferentially after a successful match.
If Range GETs needs to be enabled for all files, configure Range GETs for the domain name
cloud.tencent.com as follows:
User A requests for the resource
http://cloud.tencent.com/test.apk. After the node receives the request and finds that the cached file
test.apk has already expired, the node initiates an origin-pull request. Since Range GETs is enabled for all files in the current rule, the node uses a range request to obtain and cache the resource in shards. If user B also makes a range request for the same file to the same node and the shards that are stored on the node match the specified byte segments in the range request, the resource is directly returned to user B even though the shards are not completely obtained.
If Range GETs needs to be enabled for only part of your files, configure Range GETs for the domain name
cloud.tencent.com as follows:
When user A makes a request for the
http://cloud.tencent.com/test.apk resource, as the bottom rule has a higher priority than the top rule, Range GETs will be used for the request if the node resource is not hit or the cached resource has expired. If user B makes a request for the
http://cloud.tencent.com/test.jpg resource, as it only matches the rule for all files, Range GETs won't be used when origin-pull is performed for the request.