WO2024174694A1

WO2024174694A1 - Cache node scheduling method and apparatus, and device and medium

Info

Publication number: WO2024174694A1
Application number: PCT/CN2023/139692
Authority: WO
Inventors: 毕天枢; 齐铁鹏; 许磊
Original assignee: 天翼云科技有限公司
Priority date: 2023-02-24
Filing date: 2023-12-19
Publication date: 2024-08-29
Also published as: CN115842863B; CN115842863A

Abstract

Provided in the present application are a cache node scheduling method and apparatus, and a device and a medium, which aim to avoid the influence of a back-to-source sudden increase on a normal service. The method comprises: receiving a request for a target resource; performing DNS resolution on the target resource to obtain a first resolution result, and acquiring a second resolution result from a cache, wherein the second resolution result is a DNS resolution result that is previously obtained by means of performing DNS resolution on the target resource; when the first resolution result is different from the second resolution result, determining a first parent-layer cache node according to the first resolution result, and determining a second parent-layer cache node according to the second resolution result; and when the first parent-layer cache node is different from the second parent-layer cache node, sending to the first parent-layer cache node a resource request which carries information of the second parent-layer cache node, such that the first parent-layer cache node acquires the target resource from the second parent-layer cache node, and returns the target resource to an edge-layer cache node.

Description

A cache node scheduling method, device, equipment, and medium

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to a Chinese patent application filed with the China Patent Office on February 24, 2023, with application number 202310163245.5 and application name “A Cache Node Scheduling Method, Apparatus, Equipment, and Medium”, all contents of which are incorporated by reference in this application.

Technical Field

The present application relates to the technical field of content distribution networks, and in particular to a cache node scheduling method, device, equipment, and medium.

Background Art

By utilizing cache nodes at different levels in the content delivery network (CDN), source site resources that are farther away from users can be cached in cache nodes that are closer to users. This allows users to obtain resources nearby when requesting a resource, thereby reducing source site pressure and increasing access speed.

However, when the parent cache node of the edge layer cache node changes due to machine cutover and resource adjustment (that is, the parent node of the edge layer cache node switches), a sudden increase in back-to-source traffic will occur, affecting the execution efficiency of normal business requests and reducing the service performance of the cache node.

Summary of the invention

In view of the above problems, embodiments of the present application provide a cache node scheduling method, apparatus and device, medium and product to overcome the above problems or at least partially solve the above problems.

In a first aspect of an embodiment of the present application, a cache node scheduling method is provided, which is applied to an edge layer cache node, and the method includes:

receiving a request for a target resource;

Performing a DNS resolution on the target resource to obtain a first resolution result, and obtaining a second resolution result from the cache, where the second resolution result is the DNS resolution result obtained by performing a DNS resolution on the target resource previously;

When the first parsing result and the second parsing result are different, determining the first parent cache node according to the first parsing result, and determining the second parent cache node according to the second parsing result;

When the first parent cache node is different from the second parent cache node, a resource request carrying information of the second parent cache node is sent to the first parent cache node, so that the first parent cache node obtains the target resource from the second parent cache node and returns the target resource to the edge cache node.

A second aspect of an embodiment of the present application provides a cache node scheduling method, which is applied to a parent cache node after an edge cache node performs a parent node switch, and the method includes:

Receive a resource request from an edge layer cache node, where the resource request carries information of a first node, where the first node is a parent layer cache node before the edge layer cache node performs a parent layer node switch;

Generate and send a prefetch request to the first node according to the configured prefetch header and information of the first node;

Receive the target resource returned by the first node in response to the pre-fetch request, and return the target resource to the edge layer cache node.

A third aspect of an embodiment of the present application provides a cache node scheduling method, which is applied to a parent layer cache node before an edge layer cache node switches a parent layer node, and the method includes:

Receive a pre-fetch request from a third node, where the third node is the parent node of the edge layer cache node after the parent node is switched. Cache nodes;

In the case where the target resource corresponding to the pre-fetch request is cached, returning the target resource to the third node, so that the third node returns the target resource to the edge layer cache node;

In the case where the target resource corresponding to the pre-fetch request is not cached, an abnormal status code is returned to the third node, so that the third node returns to the source or returns to the parent.

In a fourth aspect of an embodiment of the present application, a cache node scheduling device is provided, which is applied to an edge layer cache node, and the device includes:

A request receiving module, used for receiving a request for a target resource;

A DNS resolution module is used to perform DNS resolution on the target resource to obtain a first resolution result, and obtain a second resolution result from a cache, where the second resolution result is the DNS resolution result obtained by previously performing DNS resolution on the target resource;

A node determination module, used to determine a first parent cache node according to the first analysis result and a second parent cache node according to the second analysis result when the first analysis result and the second analysis result are different;

A resource request module is used to send a resource request carrying information of the second parent cache node to the first parent cache node when the first parent cache node is different from the second parent cache node, so that the first parent cache node obtains the target resource from the second parent cache node and returns the target resource to the edge cache node.

In a fifth aspect of the embodiments of the present application, a cache node scheduling device is provided, which is applied to a parent cache node after an edge cache node performs parent node switching, and the device includes:

A first receiving module is used to receive a resource request from an edge layer cache node, where the resource request carries information of a first node, where the first node is a parent layer cache node before the edge layer cache node performs a parent layer node switch;

A first generating module, configured to generate and send a prefetch request to the first node according to the configured prefetch header and information of the first node;

The first transmission module is used to receive the target resource returned by the first node in response to the pre-fetch request, and return the target resource to the edge layer cache node.

In a sixth aspect of an embodiment of the present application, a cache node scheduling device is provided, which is applied to a parent layer cache node before an edge layer cache node switches a parent layer node, and the device includes:

A pre-fetch request receiving module, used for receiving a pre-fetch request from a third node, where the third node is a parent layer cache node after the edge layer cache node performs parent layer node switching;

A first response module is used to return the target resource to the third node when the target resource corresponding to the pre-fetch request is cached, so that the third node returns the target resource to the edge layer cache node;

The second response module is used to return an abnormal status code to the third node when there is no target resource corresponding to the cache pre-fetch request, so that the third node returns to the source or returns to the parent.

In the seventh aspect of the embodiments of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the cache node scheduling method of the first aspect are implemented, or the steps of the cache node scheduling method of the second aspect are implemented, or the steps of the cache node scheduling method of the third aspect are implemented.

In an eighth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program/instruction is stored, and when the computer program/instruction is executed by a processor, the steps of the cache node scheduling method of the first aspect are implemented. Or implement the steps of the cache node scheduling method of the second aspect or implement the steps of the cache node scheduling method of the third aspect.

In a ninth aspect of an embodiment of the present application, a computer program product is provided, including a computer program/instruction, which, when executed by a processor, implements the steps of the cache node scheduling method of the first aspect, or implements the steps of the cache node scheduling method of the second aspect, or implements the steps of the cache node scheduling method of the third aspect.

The embodiments of the present application include the following advantages:

In this embodiment, the edge layer cache node can find the second parent layer cache node that may cache the target resource in a timely manner after switching the parent layer node based on the first analysis result and the second analysis result, and send a resource request carrying the information of the second parent layer cache node to the first parent layer cache node, so that the first parent layer cache node can pre-fetch resources from the second parent layer cache node, thereby reducing the number of back-to-source times of the edge layer cache node after switching the parent layer node, avoiding the impact on the execution efficiency of normal business requests due to a sudden increase in back-to-source, and thereby improving the service performance of the cache node.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

Figure 1 is a flowchart of the implementation of the existing consistent hashing scheme

FIG2 is a flowchart of an implementation method of a cache node scheduling method in an embodiment of the present application;

FIG3 is a schematic diagram of an implementation process of a cache node scheduling method according to an embodiment of the present application;

FIG4 is a flowchart of another cache node scheduling method in an embodiment of the present application;

FIG5 is a flowchart of another cache node scheduling method in an embodiment of the present application;

FIG6 is a schematic diagram of the structure of a cache node scheduling device according to an embodiment of the present application;

FIG7 is a schematic diagram of the structure of another cache node scheduling device according to an embodiment of the present application;

FIG8 is a schematic diagram of the structure of another cache node scheduling device according to an embodiment of the present application;

FIG. 9 is a schematic diagram of an electronic device in an embodiment of the present application.

DETAILED DESCRIPTION

In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

In the CDN system, the multi-level structural design enables the edge layer cache node and its parent layer cache node to cache resources. When users obtain resources, if they can hit the resources of the edge layer cache node or the parent layer cache node, they can directly obtain the corresponding resources. Therefore, a high cache hit rate is a requirement for the CDN system to maintain normal services. If the resource hit rate of the CDN system is low, it will cause greater pressure on the source station, affecting the normal execution of the user's business request, and in severe cases, it will cause service failures of the CDN system.

The existing CDN system mainly adopts the consistent hashing scheme to ensure a high resource hit rate. As shown in Figure 1, when the user's business request (i.e., a request for a certain resource) reaches the edge layer cache node, the edge layer cache node can request the target resource from its parent layer cache node if it does not cache the target resource or the cache time of the target resource exceeds the survival time. If the parent layer cache node is used in the configuration file associated with the edge layer cache node, If the domain name is configured, the edge layer cache node can construct a consistent hash ring (referred to as a hash ring) containing the domain names and IP addresses of each parent layer cache node associated with the target resource according to the DNS resolution result obtained from the Domain Name System (DNS), and obtain the corresponding IP address according to the hash value of the domain name corresponding to the target resource in the consistent hash ring and go to the corresponding parent layer cache node to obtain the target resource.

It can be understood that when the parent cache node of the edge layer cache node does not change, the DNS resolution result obtained by the edge layer cache node usually remains unchanged. Therefore, the consistent hash ring constructed through the DNS resolution result can enable the edge layer cache node to obtain specific resources from a specific parent layer cache node (that is, a specified parent layer machine), thereby improving the resource hit rate.

Each time the edge layer cache node returns to the parent layer cache node to obtain resources (such as the cache time of the resource exceeds the configured survival time), it needs to delete the temporary information corresponding to the previously established consistent hash ring and reinitialize it (such as re-obtaining the DNS resolution result and building a consistent hash ring). This allows the edge layer cache node to build a new consistent hash ring based on the new DNS resolution result when its parent layer cache node switches, thereby obtaining resources from the new parent layer cache node.

Based on the above consistent hashing scheme, each resource is usually cached in only one parent cache node of the edge cache node. However, as the business volume increases, the parent cache nodes of the edge cache nodes will continue to expand. When the parent cache nodes change due to machine cutover and resource adjustment, the DNS resolution results containing the parent cache node information will also be updated. When the edge cache node obtains resources based on the new DNS resolution results, it may obtain resources from the parent cache node that does not cache the resource, resulting in a sudden increase in back-to-source, which affects the efficiency of normal business request execution and reduces the service performance of the cache devices in the CDN system.

In response to the problems existing in the above-mentioned related technologies, the present application proposes a cache node scheduling method, which adds automated cache node scheduling on the basis of the original consistent hashing scheme, so that when the edge layer cache node obtains resources from the new parent layer cache node due to changes in the DNS resolution results, the new parent layer cache node can first pre-fetch resources from the parent layer cache node that may cache the resource, thereby reducing the number of back-to-source times of the edge layer cache node after switching the parent layer node, avoiding the impact on the execution efficiency of normal business requests due to a sudden increase in back-to-source, and thereby improving the service performance of the cache node.

The cache node scheduling method provided by the embodiment of the present application is described in detail below through some embodiments and their application scenarios in combination with the accompanying drawings.

In the first aspect, as shown in FIG. 2 , which is a flowchart of an implementation of a cache node scheduling method provided in an embodiment of the present application, the method is applied to an edge layer cache node. The method may include the following steps:

Step S11: Receive a request for a target resource.

In a specific implementation, the edge layer cache node can be directly connected to the user's client. At this time, the request for the target resource is the business request issued by the client. The edge layer cache node can also be the upper cache node (i.e., the parent cache node) of other edge layer cache nodes. At this time, the request for the target resource can be a return-to-parent related request issued by its lower-layer cache node, or a prefetch request issued by its same-level cache node for prefetching resources.

Step S12: Perform DNS resolution on the target resource to obtain a first resolution result, and obtain a second resolution result from the cache, where the second resolution result is the DNS resolution result obtained by previously performing DNS resolution on the target resource.

It can be understood that, for a request for a target resource currently received by the edge layer cache node, the first resolution result is: the DNS resolution result obtained by the edge layer cache node after receiving the request. DNS resolution result; the second resolution result is: the DNS resolution result obtained by the edge layer cache node when performing DNS resolution on the target resource before receiving the request, and the DNS resolution result obtained by performing DNS resolution on the target resource before is cached in the memory by the edge layer cache node.

In the specific implementation, after receiving a request for a target resource, the edge layer cache node will query the DNS for information about each parent layer cache node (i.e., parent layer node) for the target resource (such as the domain name and IP address of the parent layer cache node) to obtain a first resolution result, and obtain the information about each parent layer cache node previously obtained by querying the DNS for the target resource from the cache of the edge layer cache node or the cache of the DNS to obtain a second resolution result.

Step S13: when the first parsing result and the second parsing result are different, determining the first parent cache node according to the first parsing result, and determining the second parent cache node according to the second parsing result.

It is understandable that the DNS resolution results (such as the first resolution result and the second resolution result) may contain multiple IPs. Therefore, the process of the edge layer cache node determining the parent layer cache node based on the DNS resolution results mainly includes the IP screening process, that is, the edge layer cache node will screen the IP according to the obtained DNS resolution results to find the appropriate parent layer cache node IP (such as the parent layer cache node IP with a higher resource hit rate) to obtain the target resource.

In the specific implementation, considering that the parent layer cache node corresponding to the target resource before the parent layer node switches (i.e., the second parent layer cache node or the first node described below) is more likely to have cached the target resource, the edge layer cache node can perceive whether its own parent layer node has been switched by comparing the first analysis result and the second analysis result, so that when the first analysis result and the second analysis result are different (i.e., the parent layer node has been switched), the parent layer cache node corresponding to the target resource before the parent layer node is switched is determined in time according to the second analysis result, and the parent layer cache node corresponding to the target resource after the parent layer node is switched is determined according to the first analysis result (i.e., the first parent layer cache node or the third node described below).

Step S14: When the first parent cache node and the second parent cache node are different, a resource request carrying information of the second parent cache node is sent to the first parent cache node, so that the first parent cache node obtains the target resource from the second parent cache node and returns the target resource to the edge cache node.

It is understandable that before and after the parent layer node is switched, the parent layer cache node corresponding to some resources determined by the edge layer cache node based on the DNS resolution results (i.e., the first resolution result and the second resolution result) may not have changed (i.e., the first parent layer cache node is the same as the second parent layer cache node), which means that after the parent layer node is switched, the edge layer cache node can still hit these resources, so there is no need to pre-fetch these resources.

Considering that the first parent cache node and the second parent cache node are different, if the edge cache node directly returns to the first parent cache node to obtain the target resource, since the first parent cache node and the parent cache node to which the first parent cache node returns to the parent may not have cached the target resource, the edge cache node eventually needs to obtain the target resource from the source station, resulting in a sudden increase in the number of returns to the source after the parent node switches.

Therefore, when the parent layer cache node switches, the edge layer cache node in the present application will first send a resource request carrying the information of the old parent layer cache node (i.e., the second parent layer cache node) corresponding to the target resource, so as to schedule the new parent layer cache node (i.e., the first parent layer cache node) corresponding to the target resource to go to the old parent layer cache node to obtain the target resource, thereby reducing the number of return to the source after the parent layer cache node switches.

By adopting the technical solution of the embodiment of the present application, the edge layer cache node can find the second parent layer cache node that may cache the target resource in time after the parent layer node is switched according to the first parsing result and the second parsing result, and send a resource request carrying the information of the second parent layer cache node to the first parent layer cache node, so that the first The parent layer cache node can pre-fetch resources from the second parent layer cache node, thereby reducing the number of back-to-source times after the edge layer cache node switches the parent layer node, avoiding the impact on the execution efficiency of normal business requests caused by a sudden increase in back-to-source, and thus improving the service performance of the cache node.

The above technical solution is further described below in conjunction with Figure 3. As shown in Figure 3, the embodiment of the present application provides an implementation process of a cache node scheduling method, including:

(1) The edge layer cache node determines whether the DNS resolution results are different or whether there is a pre-fetch request already in progress.

In the specific implementation, when the edge layer cache node switches the parent layer node, the DNS will cache the information of the relevant parent layer cache nodes after the parent layer node is switched. Whenever the edge layer cache node receives a request for a certain resource (ie, the target resource), it will request the DNS for the information of each parent layer cache node currently associated with the target resource, obtain the DNS resolution result, and thus complete the DNS resolution of the target resource.

After obtaining the DNS resolution result currently corresponding to the target resource (i.e., the first resolution result), the edge layer cache node compares the first resolution result with the DNS resolution result previously obtained from the DNS request (i.e., the second resolution result). If the first resolution result and the second resolution result are different, it means that the parent layer cache node of the edge layer cache node has changed (i.e., the edge layer cache node has switched the parent layer node). The DNS resolution result previously obtained from the DNS request can be saved in a database associated with the edge layer cache node (such as HostDB).

As a possible implementation method, considering that the edge layer cache node may frequently switch the parent layer node in a short period of time, at this time, in order to find the parent layer cache node that is more likely to cache the target resource, the edge layer cache node needs to obtain the information of the relevant parent layer cache node before the first parent layer node switch in the time period. Therefore, the minimum switching time (i.e., the first target time) can be added to the configuration file of the edge layer cache node to limit the edge layer cache node from updating the DNS resolution result previously obtained from the DNS request.

Exemplarily, each time the edge layer cache node switches the parent layer node, the DNS will cache the information of the relevant parent layer cache nodes before and after the parent layer node switch. Within a certain time period, after the edge layer cache node switches the parent layer node for the first time, the edge layer cache node will cache the first historical resolution result corresponding to the target resource obtained from the DNS (including the information of the relevant parent layer cache node before the parent layer node switch), and obtain the configured first target time from the configuration file.

In the case where the cache time of the first historical resolution result does not exceed the first target time, the edge layer cache node can directly determine the first historical resolution result as the second resolution result. It is understandable that within the first target time, no matter how many times the edge layer cache node has switched the parent layer node, the information of the relevant parent layer cache node before the first parent layer node switch will be used as the second resolution result (i.e., the update of the old hash ring, etc. is restricted) to determine the old parent layer cache node that is more likely to cache the target resource.

When the cache time of the first historical resolution result exceeds the first target time and the second historical resolution result corresponding to the target resource reacquired from the DNS is different from the first historical resolution result, the second historical resolution result is determined as the second resolution result.

It can be understood that in order to reduce the number of old parent cache nodes that are determined to be unavailable nodes (such as nodes that have completed machine cutover and have been deleted), after the first target time, if the edge cache node switches the parent node again, and the information of the relevant parent cache node before the secondary parent node switch (i.e., the second historical analysis result) is different from the first historical analysis result, the edge cache node can use the second historical analysis result to update the cache The first historical parsing result is convenient for the edge layer cache node to use the second historical parsing result as the second parsing result to determine the available old parent layer cache node.

It should be noted that the cache time of the edge layer cache node for the historical resolution results (such as the first historical resolution results and the second historical resolution results mentioned above) obtained after each parent layer node switch may not be limited by the survival time. For example, the edge layer cache node may only obtain and cache the historical resolution results once after each parent layer node switch, and use the historical resolution results from the cache at any subsequent time. The corresponding survival time may also be added to the configuration file of the edge layer cache node so that the edge layer cache node obtains the historical resolution results from the DNS multiple times according to the survival time.

For example, after the edge layer cache node performs a parent layer node switch, the edge layer cache node caches the third historical resolution result corresponding to the target resource obtained from the DNS (i.e., the information of the relevant parent layer cache node before the parent layer node switch in the current DNS cache), and determines whether the third historical resolution result is configured with a survival time according to the configuration file. In the case where the third historical resolution result is configured with a survival time, the edge layer cache node periodically obtains the third historical resolution result corresponding to the target resource from the DNS according to the survival time, and uses each newly obtained third historical resolution result to overwrite the third historical resolution result corresponding to the target resource cached last time. Each time the second resolution result is obtained, the edge layer cache node will determine the currently cached third historical resolution result as the second resolution result.

It can be understood that, depending on the different configured conditions for triggering the acquisition of historical resolution results (such as the edge layer cache node can be configured to: when the cache time of the third historical resolution result exceeds the survival time and a request for the target resource is received, then the third historical resolution result corresponding to the target resource will be re-acquired, or the edge layer cache node can be configured to: when the cache time of the third historical resolution result exceeds the survival time, then the third historical resolution result corresponding to the target resource will be re-acquired), the time interval between two consecutive acquisitions of the third historical resolution result by the edge layer cache node can be the same or different.

(2) When the DNS resolution results are different or there is an ongoing pre-fetch request, the edge layer cache node sends a resource request carrying the information of the corresponding old parent layer cache node to the corresponding new parent layer cache node; when the DNS resolution results are the same or there is no ongoing pre-fetch request, the edge layer cache node still obtains the target resource from the corresponding old parent layer cache node.

In the specific implementation, if the DNS resolution result is the same or there is no pre-fetch request in progress, it means that the specific node of the edge layer cache node back to the parent has not changed. At this time, the edge layer cache node can normally obtain the target resource directly from the corresponding parent layer cache node (equivalent to the corresponding old parent layer after the subsequent parent layer node is switched).

In the case where the DNS resolution results are different or there is an ongoing pre-fetch request, the edge layer cache node can build the first hash ring and the second hash ring according to the corresponding DNS resolution results before and after the parent layer node is switched (i.e., the first resolution result and the second resolution result), and obtain the information (such as IP and port) of the new parent layer cache node (i.e., the first parent layer cache node) and the old parent layer cache node (i.e., the second parent layer cache node) according to the hash value of the domain name corresponding to the target resource in the first hash ring and the second hash ring. Among them, the above-mentioned second hash ring can be saved in the cache of DNS and/or edge cache node after establishment, so that it can be accessed at any time later.

If the hash values of the domain names corresponding to the target resources obtained from the first hash ring and the second hash ring are different, and/or the information of the new parent cache node is different from that of the old parent cache node, it means that the change of the parent cache node affects the specific node of the edge cache node back to the parent. Then, the edge cache node sends a resource request carrying the information of the corresponding old parent cache node to the corresponding new parent cache node to schedule the new parent cache node for resource pre-fetching.

As a possible implementation method, the parent's effective time (i.e., the second target time) can be added back to the configuration file of the edge layer cache node to limit the duration of the edge layer cache node scheduling the new parent layer cache node for resource prefetching through a resource request carrying a request header (i.e., the information of the second parent layer cache node) after sensing the switching of the parent layer node.

Exemplarily, the edge layer cache node can obtain the configured second target time, and within the second target time after the edge layer cache node switches the parent layer node, send a resource request carrying information of the second parent layer cache node to the first parent layer cache node, wherein the information of the second parent layer cache node may include the IP and port of the second parent layer cache node.

(3) The new parent cache node receives a resource request carrying information of the corresponding old parent cache node, generates and sends a pre-fetch request to the old parent cache node.

In the specific implementation, when the new parent layer cache node processes the resource request issued by the edge layer machine, it will detect whether the resource request carries a request header. If it carries a request header and the new parent layer cache node does not cache the target resource corresponding to the resource request, it will first pre-fetch resources from the cache node corresponding to the IP in the request header.

Exemplarily, after the edge layer cache node performs a parent layer node switch, the parent layer cache node (i.e., the new parent layer cache node) receives a resource request from the edge layer cache node, which carries information of the first node (i.e., the old parent layer cache node); the new parent layer cache node generates and sends a prefetch request to the first node based on the configured prefetch header and information of the first node, wherein the configured prefetch header may be a "Ctr-Pre-Request: yes" field, indicating that the request is a prefetch request, and other requests of the new parent layer cache node (such as related requests back to the parent) may add a "Ctr-Pre-Request: no" field, indicating that the request is a non-prefetch request; the new parent layer cache node receives the target resource returned by the first node for the prefetch request, and returns the target resource to the edge layer cache node.

As a possible implementation method, considering that after the edge layer cache node switches the parent layer node, the old parent layer cache node may stop serving, at this time, the new parent layer cache node will get an abnormal status code when prefetching any resource from the old parent layer cache node. If the new parent layer cache node keeps retrying the request to the old parent layer cache node, the resource request efficiency will be greatly reduced. Therefore, when the new parent layer cache node fails to obtain resources from the first node (such as receiving an abnormal status code for a prefetch request for any resource), the information of the first node (such as IP) is marked as unavailable information, that is, the unavailable information represents the information of the first node marked when the edge layer node failed to obtain resources from the first node before; and before each time the first node needs to prefetch resources, first determine whether the information of the first node is unavailable information. If the information of the first node is not unavailable information, generate and send a prefetch request to the first node based on the configured prefetch header and the information of the first node.

Exemplarily, the new parent cache node can determine whether the IP address of the old parent cache node carried in the received resource request is available. If it is not available, it will not pre-fetch resources from the old parent cache node, but will instead return to the source and the parent according to the configuration file associated with the new parent cache node.

(4) The old parent cache node receives the pre-fetch request and returns the target resource corresponding to the pre-fetch request to the new parent cache node. The new parent cache node then returns the target resource to the corresponding edge cache node.

In specific implementation, the old parent cache node receives the prefetch request from the third node (i.e., the new parent cache node), and queries whether it has cached the target resource corresponding to the prefetch request. If the target resource corresponding to the prefetch request is cached, the old parent cache node returns a normal status code and the target resource to the third node, so that the third node returns the target resource to the edge cache node; if the target resource corresponding to the prefetch request is not cached, the old parent cache node returns a normal status code and the target resource to the edge cache node. The storage node returns an abnormal status code to the third node, so that the third node returns to the source or returns to the parent. Returning to the parent means that the third node directly obtains the target resource from its own parent cache node; returning to the source means that the parent cache node of the third node does not cache the target resource (or does not have a valid cache of the target resource), and the parent cache node of the third node needs to continue to obtain the target resource from the upper cache node (if the cache nodes at all levels of the upper layer do not have the target resource, the target resource will eventually be obtained from the source station), and then the third node obtains the target resource from the parent cache node.

After receiving the exception status code, the third node will not retry pre-fetching resources from the old parent cache node multiple times to avoid the link becoming longer due to multiple retries, and will delete the pre-fetch header and the information of the old parent cache node carried in the pre-fetch request to generate a return-to-parent related request and send it to the return-to-parent related node (i.e., the second-layer parent cache node) to avoid the return-to-parent related node performing additional resource pre-fetching due to receiving the pre-fetch header and the information of the old parent cache node.

As a possible implementation method, when the received resource request is a shard request, the new parent cache node obtains the cache uniform resource locator (URL) of the shard corresponding to the resource request, and generates and sends a prefetch request for the corresponding shard to the old parent cache node based on the configured prefetch header, cache URL, and information of the old parent cache node.

It is understandable that, considering that the edge layer cache node may generate multiple shard requests according to the received request for the target resource, each shard request is used to request a shard of the target resource. In the existing CDN system, the cache node will respond to any shard request corresponding to the target resource using the query result for the index shard (such as the first shard of the target resource), that is, after the new parent layer cache node generates a pre-fetch request according to any received shard request and sends the pre-fetch request to the old parent layer cache node, the old parent layer cache node will query whether it has cached the index shard of the target resource corresponding to the pre-fetch request. In the case that the old parent layer cache node does not cache the shard corresponding to the pre-fetch request but caches the index shard, the old parent layer cache node will still return a normal status code to the new parent layer cache node, but at this time the new parent layer cache node cannot obtain the corresponding shard from the old parent layer cache node, nor will it return to the source or return to the parent, resulting in the new parent layer cache node and the edge layer cache node caching the wrong resource.

In order to avoid the above-mentioned situation of caching wrong resources, the new parent cache node in the present application will add the cache url (i.e., the Ctl-Chash-Key header referred to in Figure 3) in the received shard request to the prefetch request. The old parent cache node can directly query whether it has cached the shard corresponding to the prefetch request based on the cache url, thereby skipping the operation of the shard plug-in (i.e., the old parent cache node will not use the query result for the index slice to respond to the prefetch request for the shard), so that the new parent cache node can know the actual caching status of the shard in the old parent cache node through the received status code, and can return to the source or parent in time, thereby avoiding the problem of cache errors.

Based on the above embodiments, the present application compares the DNS resolution results before and after the parent node switch under the scheme based on the existing consistent hashing back to the parent, and realizes the scheduling of each cache node in the CDN system through the information interaction between the edge layer node, the new parent layer cache node and the old parent layer cache node. By pre-fetching resources between groups through the same-layer cache nodes, the cache of the old parent layer cache node before the parent layer node switch can be obtained, and it can adapt to complete requests and shard requests without adding other configurations. It solves the problem of a sudden increase in the source return after the parent layer node is switched in the logic of the original scheme, and can adapt to the situation of switching the parent layer node multiple times in a short period of time. There is no need to manually increase the cache of the parent layer cache node to spread the resources, and the automatic cache node scheduling is completed.

In the second aspect, as shown in FIG. 4, another cache node scheduling method provided by an embodiment of the present application is implemented The flowchart is applied to the parent layer cache node after the edge layer cache node performs parent layer node switching. The method may include the following steps:

Step S21: receiving a resource request from an edge layer cache node, where the resource request carries information of a first node, where the first node is a parent layer cache node before the edge layer cache node performs parent layer node switching;

Step S22: Generate and send a prefetch request to the first node according to the configured prefetch header and information of the first node;

Step S23: receiving the target resource returned by the first node in response to the pre-fetch request, and returning the target resource to the edge layer cache node.

By adopting the technical solution of the embodiment of the present application, the edge layer cache node can, based on the first parsing result and the second parsing result, promptly find the second parent layer cache node that may cache the target resource after the parent layer node is switched, and send a resource request carrying the information of the second parent layer cache node to the first parent layer cache node, so that the first parent layer cache node can pre-fetch resources from the second parent layer cache node, thereby reducing the number of back-to-source times of the edge layer cache node after the parent layer node is switched, avoiding the impact on the execution efficiency of normal business requests due to the sudden increase in back-to-source, and thereby improving the service performance of the cache node.

As a possible implementation,

Generate and send a prefetch request to the first node according to the configured prefetch header and information of the first node, including:

Determine whether the information of the first node is unavailable information, where the unavailable information is used to represent: information of the first node marked when the edge layer node previously failed to obtain resources from the first node;

When the information of the first node does not belong to unavailable information, a prefetch request is generated and sent to the first node according to the configured prefetch header and the information of the first node.

As a possible implementation, generating and sending a prefetch request to the first node according to the configured prefetch header and information of the first node includes:

When the resource request is a shard request, obtain the cache uniform resource locator URL of the shard corresponding to the resource request;

According to the configured prefetch header, cache URL, and information of the first node, a prefetch request for the corresponding shard is generated and sent to the first node.

In a third aspect, as shown in FIG. 5 , a flowchart of another cache node scheduling method provided in an embodiment of the present application is provided, which is applied to a parent cache node before an edge cache node switches a parent node. The method may include the following steps:

Step S31: receiving a pre-fetch request from a third node, where the third node is a parent cache node after the edge cache node has undergone parent node switching;

Step S32: if the target resource corresponding to the pre-fetch request is cached, return the target resource to the third node, so that the third node returns the target resource to the edge layer cache node;

Step S33: when there is no target resource corresponding to the pre-fetch request in the cache, an abnormal status code is returned to the third node, so that the third node returns to the source or returns to the parent.

By adopting the technical solution of the embodiment of the present application, the edge layer cache node can find the second parent layer cache node that may cache the target resource in time after switching the parent layer node according to the first parsing result and the second parsing result, and By sending a resource request carrying the information of the second parent cache node to the first parent cache node, the first parent cache node can pre-fetch resources from the second parent cache node, thereby reducing the number of back-to-source times of the edge cache node after the parent node switches, avoiding the impact on the execution efficiency of normal business requests due to a sudden increase in back-to-source, and thus improving the service performance of the cache node.

For the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the order of the actions described, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all partial embodiments of the present application, and the actions involved are not necessarily required by the embodiments of the present application.

In a fourth aspect, FIG6 is a schematic diagram of the structure of a cache node scheduling device according to an embodiment of the present application, the device is applied to an edge layer cache node, and the device includes:

A request receiving module, used for receiving a request for a target resource;

In some embodiments of the present application, before obtaining the second parsing result from the cache, the device further includes:

A first acquisition module is used to cache a first historical resolution result corresponding to a target resource obtained from a DNS after the edge layer cache node switches the parent layer node, and obtain a configured first target time;

The DNS resolution module includes:

A first parsing submodule, configured to determine the first historical parsing result as the second parsing result when the cache time of the first historical parsing result does not exceed the first target time;

The second resolution submodule is used to determine the second historical resolution result as the second resolution result when the cache time of the first historical resolution result exceeds the first target time and the second historical resolution result corresponding to the target resource reacquired from the DNS is different from the first historical resolution result.

A second acquisition module is used to cache a third historical resolution result corresponding to the target resource obtained from the DNS after the edge layer cache node performs a parent layer node switch;

A third acquisition module is used to periodically update the cached third historical analysis result corresponding to the target resource according to the survival time when the third historical analysis result is configured with a survival time;

The DNS resolution module includes:

The third parsing submodule is used to determine the third historical parsing result currently cached by the edge layer cache node as the second parsing result.

In some embodiments of the present application, the resource request module includes:

A first resource request submodule, used to obtain a configured second target time;

The second resource request submodule is used to send a resource request carrying information of the second parent cache node to the first parent cache node within a second target time after the edge cache node switches the parent node.

In some embodiments of the present application, the node determination module includes:

A first determination module is used to construct a first hash ring according to the first parsing result, and obtain a second hash ring corresponding to the second parsing result from the cache of the edge layer cache node;

The second determination module is used to determine the first parent layer cache node and the second parent layer cache node respectively according to the first hash ring and the second hash ring.

In a fifth aspect, FIG. 7 is a schematic diagram of the structure of another cache node scheduling device according to an embodiment of the present application, the device is applied to the parent layer cache node after the edge layer cache node performs parent layer node switching, the device comprising:

In some embodiments of the present application,

The first generation module includes:

The first generating submodule is used to determine whether the information of the first node is unavailable information, where the unavailable information is used to represent: the information of the first node marked when the edge layer node previously failed to obtain resources from the first node;

The second generating submodule is used to generate and send a pre-fetch request to the first node according to the configured pre-fetch header and the information of the first node when the information of the first node does not belong to unavailable information.

In some embodiments of the present application, the first generation module includes:

The third generation submodule is used to obtain the cache uniform resource locator URL of the slice corresponding to the resource request when the resource request is a slice request;

The fourth generation submodule is used to generate a pre-fetch header, a cache URL, and information of the first node according to the configuration. And send a pre-fetch request for the corresponding shard to the first node.

In a sixth aspect, FIG8 is a schematic diagram of the structure of another cache node scheduling device according to an embodiment of the present application, the device is applied to the parent layer cache node before the edge layer cache node switches the parent layer node, and the device includes:

It should be noted that the device embodiment is similar to the method embodiment, so the description is relatively simple, and the relevant parts can be referred to the method embodiment.

The present application also provides an electronic device, with reference to FIG9 , which is a schematic diagram of the electronic device proposed in the present application. As shown in FIG9 , the electronic device 100 includes: a memory 110 and a processor 120 , the memory 110 and the processor 120 are connected via a bus communication, the memory 110 stores a computer program, and the computer program can be run on the processor 120 to implement the steps in the cache node scheduling method disclosed in the present application.

The embodiment of the present application also provides a computer-readable storage medium on which a computer program/instruction is stored. When the computer program/instruction is executed by a processor, the cache node scheduling method disclosed in the embodiment of the present application is implemented.

The embodiment of the present application also provides a computer program product, including a computer program/instruction, which, when executed by a processor, implements the cache node scheduling method disclosed in the embodiment of the present application.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

The embodiments of the present application are described with reference to the flowcharts and/or block diagrams of the methods, systems, devices, storage media, and program products according to the embodiments of the present application. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the processes and/or boxes in the flowchart and/or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing one process or multiple processes in the flowchart and/or one box in the block diagram. or a device that performs the functions specified in more than one box.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce computer-implemented processing, so that the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

Although some embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiment and all changes and modifications that fall within the scope of the present application.

Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

The above is a detailed introduction to a cache node scheduling method, device and equipment, medium and product provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

A cache node scheduling method, characterized in that it is applied to edge layer cache nodes, the method comprising:

receiving a request for a target resource;

Performing a domain name system DNS resolution on the target resource to obtain a first resolution result, and acquiring a second resolution result from a cache, where the second resolution result is a DNS resolution result obtained by previously performing a DNS resolution on the target resource;

When the first parsing result and the second parsing result are different, determining a first parent cache node according to the first parsing result, and determining a second parent cache node according to the second parsing result;

In the case that the first parent cache node is different from the second parent cache node, a resource request carrying information of the second parent cache node is sent to the first parent cache node, so that the first parent cache node obtains the target resource from the second parent cache node and returns the target resource to the edge cache node.
The method according to claim 1, characterized in that before obtaining the second parsing result from the cache, the method further comprises:

After the edge layer cache node performs parent layer node switching, caches the first historical resolution result corresponding to the target resource obtained from the DNS, and obtains the configured first target time;

The obtaining the second parsing result from the cache includes:

If the cache time of the first historical analysis result does not exceed the first target time, determining the first historical analysis result as the second analysis result;

When the cache time of the first historical resolution result exceeds the first target time and the second historical resolution result corresponding to the target resource reacquired from the DNS is different from the first historical resolution result, the second historical resolution result is determined as the second resolution result.
The method according to claim 1, characterized in that before obtaining the second parsing result from the cache, the method further comprises:

After the edge layer cache node performs parent layer node switching, caches the third historical resolution result corresponding to the target resource obtained from the DNS;

In the case where the third historical analysis result is configured with a survival time, periodically updating the cached third historical analysis result corresponding to the target resource according to the survival time;

The obtaining the second parsing result from the cache includes:

The third historical parsing result currently cached by the edge layer cache node is determined as the second parsing result.
The method according to any one of claims 1 to 3, characterized in that the sending of the resource request carrying the information of the second parent cache node to the first parent cache node comprises:

Get the configured second target time;

Within the second target time after the edge layer cache node performs parent layer node switching, a resource request carrying information of the second parent layer cache node is sent to the first parent layer cache node.
The method according to any one of claims 1 to 3, characterized in that, when the first parsing result and the second parsing result are different, determining a first parent cache node according to the first parsing result, and determining a second parent cache node according to the second parsing result, comprises:

Constructing a first hash ring according to the first parsing result, and obtaining a second hash ring corresponding to the second parsing result from the cache of the edge layer cache node;

The first parent cache node and the second parent cache node are determined respectively according to the first hash ring and the second hash ring.
A cache node scheduling method, characterized in that it is applied to a parent cache node after an edge layer cache node performs a parent layer node switch, and the method comprises:

Receive a resource request from the edge layer cache node, where the resource request carries information of a first node, where the first node is a parent layer cache node before the edge layer cache node performs a parent layer node switch;

Generate and send a prefetch request to the first node according to the configured prefetch header and information of the first node;

Receive the target resource returned by the first node in response to the pre-fetch request, and return the target resource to the edge layer cache node.
The method according to claim 6, characterized in that

The generating and sending a prefetch request to the first node according to the configured prefetch header and the information of the first node comprises:

Determine whether the information of the first node is unavailable information, where the unavailable information is used to represent: information of the first node marked when the edge layer node previously failed to obtain resources from the first node;

When the information of the first node does not belong to unavailable information, a prefetch request is generated and sent to the first node according to the configured prefetch header and the information of the first node.
The method according to claim 6 or 7, characterized in that the generating and sending a prefetch request to the first node according to the configured prefetch header and the information of the first node comprises:

In the case where the resource request is a slice request, obtaining a cache uniform resource locator URL of the slice corresponding to the resource request;

According to the configured prefetch header, the cache URL, and the information of the first node, a prefetch request for the corresponding segment is generated and sent to the first node.
A cache node scheduling method, characterized in that it is applied to a parent layer cache node before an edge layer cache node switches a parent layer node, and the method comprises:

receiving a pre-fetch request from a third node, where the third node is a parent cache node after the edge layer cache node performs parent layer node switching;

In the case where the target resource corresponding to the pre-fetch request is cached, returning the target resource to the third node, so that the third node returns the target resource to the edge layer cache node;

In the case where the target resource corresponding to the pre-fetch request is not cached, an abnormal status code is returned to the third node, so that the third node returns to the source or returns to the parent.
A cache node scheduling device, characterized in that it is applied to an edge layer cache node, and the device comprises:

A request receiving module, used for receiving a request for a target resource;

A DNS resolution module is used to perform DNS resolution on the target resource to obtain a first resolution result, and obtain a second resolution result from a cache, where the second resolution result is a DNS resolution result obtained by previously performing DNS resolution on the target resource;

a node determination module, configured to determine a first parent cache node according to the first analysis result and a second parent cache node according to the second analysis result when the first analysis result and the second analysis result are different;

A resource request module is used to send a resource request carrying information of the second parent cache node to the first parent cache node when the first parent cache node is different from the second parent cache node, so that the first parent cache node obtains the target resource from the second parent cache node and returns the target resource to the edge cache node.
A cache node scheduling device, characterized in that it is applied to a parent cache node after an edge layer cache node performs a parent layer node switch, and the device comprises:

A first receiving module is used to receive a resource request of the edge layer cache node, wherein the resource request carries information of a first node, where the first node is a parent layer cache node of the edge layer cache node before the parent layer node is switched;

A first generating module, configured to generate and send a prefetch request to the first node according to the configured prefetch header and information of the first node;

The first transmission module is configured to receive the target resource returned by the first node in response to the pre-fetch request, and return the target resource to the edge layer cache node.
A cache node scheduling device, characterized in that it is applied to a parent layer cache node before an edge layer cache node switches a parent layer node, and the device comprises:

A pre-fetch request receiving module, used for receiving a pre-fetch request from a third node, wherein the third node is a parent layer cache node after the edge layer cache node performs parent layer node switching;

A first response module, configured to return the target resource to the third node if the target resource corresponding to the pre-fetch request is cached, so that the third node returns the target resource to the edge layer cache node;

The second response module is used to return an abnormal status code to the third node when the target resource corresponding to the pre-fetch request is not cached, so that the third node returns to the source or returns to the parent.
An electronic device comprises a memory, a processor and a computer program stored on the memory, wherein the processor executes the computer program to implement the cache node scheduling method as described in any one of claims 1 to 5 or the cache node scheduling method as described in any one of claims 6 to 8 or the cache node scheduling method as described in claim 9.
A computer-readable storage medium having a computer program/instruction stored thereon, characterized in that when the computer program/instruction is executed by a processor, the cache node scheduling method as described in any one of claims 1 to 5 is implemented, or the cache node scheduling method as described in any one of claims 6 to 8 is implemented, or the cache node scheduling method as described in claim 9 is implemented.
A computer program product, comprising a computer program/instruction, characterized in that when the computer program/instruction is executed by a processor, it implements the cache node scheduling method as described in any one of claims 1 to 5, or implements the cache node scheduling method as described in any one of claims 6 to 8, or implements the cache node scheduling method as described in claim 9.