CN110012106A - A method, device and system for coordinated caching based on edge computing - Google Patents

Abstract

The embodiment of the invention provides a kind of coordination caching method, apparatus and system based on edge calculations, after coordinator node receives response message, calculate the caching value of utility of object content, and when the caching value of utility of object content is less than preset threshold, the first identifier for not needing to cache in edge network node for indicating object content is added in object content；When the caching value of utility of object content is greater than or equal to preset threshold, the second identifier for needing to cache in edge network node for indicating object content is added in object content, then, for the object content after addition first or second mark, the response message comprising the object content is sent to edge network node.The embodiment of the present invention can reduce the degree of redundancy of content caching, while reduce spatial cache occupancy, to save Internet resources.

Description

A method, device and system for coordinated caching based on edge computing

technical field

The present invention relates to the technical field of computer communication, in particular to a method, device and system for coordinated caching based on edge computing.

Background technique

With the rapid development of mobile Internet technology and the wide application of Internet of Things technology, the number of edge devices in the network has increased rapidly, resulting in a rapid increase in network traffic, which has brought huge challenges to the network. At present, the end-to-end transmission mechanism mainly used in the network causes the repeated transmission of a large number of popular content in the network, which wastes network resources. Moreover, the rapid growth of network traffic makes it difficult for the delay of content transmission in the network to meet the needs of the network. The content may be, for example, video files, news, newsletters, and the like. Therefore, how to improve the utilization rate of network resources, reduce network delay, and improve user experience has become the focus of people's attention.

ICN (Information-Centric Network) network, as a new hotspot in the research of next-generation network architecture, adopts the information-centric communication method to replace the existing end-centric communication method, and directly identifies the content uniformly. And based on content identification for positioning, routing and transmission, the network is built into an integrated platform for information transmission, storage and services. One of the most important features in ICN is in-network caching, which utilizes the storage space in network nodes to cache copies of content and respond to user requests to reduce redundant traffic and relieve network bandwidth pressure. The existing methods for caching content in the ICN network include: TERC (Transparent En-Route Caching) technology, which is the default caching decision technology of most ICNs. In this technology, content is distributed on each distribution path. Cached in all nodes of the node, when the user needs to request the content, the content can be obtained directly from the requested node, which can shorten the time for the user to obtain the content.

However, the existing caching method for content in the ICN network caches the content on all nodes on each distribution path, so that the cached content is redundant and occupies a lot of space, so there is a waste of network resources. question.

SUMMARY OF THE INVENTION

The purpose of the embodiments of the present invention is to provide a coordinated caching method, device and system based on edge computing, so as to reduce the redundancy of content caching in the information center network. The specific technical solutions are as follows:

In a first aspect, an embodiment of the present invention provides an edge computing-based coordination caching method, which is applied to a coordination node in an information center network, where the information center network includes an origin server, a cloud network layer, a coordination layer, and an edge network layer, the cloud network layer includes a number of cloud network nodes, the coordination layer includes a number of coordination nodes, the edge network layer includes a number of edge network nodes, and the coordination layer is respectively connected with the cloud network layer and the edge network layer. communication connection, the method includes:

Receive a first user request, where the first user request is a user request forwarded by the edge network node, and the first user request is used to obtain the target content; query whether the edge content placement information table ECT is recorded with the target content The mark corresponding to the content; if the mark corresponding to the target content is recorded in the ECT, the first user request is forwarded to the edge network that caches the target content by using the routing information recorded in the ECT node; if the mark corresponding to the target content is not recorded in the ECT, forward the first user request to the cloud network node closest to the coordinating node, so that the cloud network node processes the a first user request; receive a response message forwarded by the cloud network node, the response message includes the target content; calculate the cache utility value of the target content by using a first preset formula; if the cache If the utility value is less than the preset threshold, a first identifier is added to the target content; the first identifier is used to indicate that the target content does not need to be cached in the edge network node; for the target content after adding the first identifier, Send a response message containing the target content to the edge network node; if the cache utility value is greater than or equal to a preset threshold, add a second identifier to the target content; the second identifier is used to indicate The target content needs to be cached in the edge network node; for the target content after adding the second identifier, a response message containing the target content is sent to the edge network node.

Optionally, the first preset formula is expressed as:

In the formula, CU(i) represents the cache utility value of the target content i, P(i) represents the popularity of the target content i, Size(i) represents the storage space occupied by the target content i, and CR(i) represents the cache of the target content i. The network revenue ratio of the target content; wherein, the popularity of the target content is used to represent the frequency of the target content being requested; the network revenue ratio of the target content cache is used to represent the target content cached at edge network nodes saves time percentage.

In a second aspect, an embodiment of the present invention provides a coordinated caching method based on edge computing, which is applied to a cloud network node in an information center network, and the method includes:

Receive a second user request, where the second user request is a user request forwarded by the coordination node, the second user request is used to obtain target content, and the second user request carries the identifier of the target content information; calculate the hash HASH value corresponding to the identification information of the target content, and search the cloud network node corresponding to the HASH value in the distributed hash table DHT; when the cloud network node corresponding to the HASH value is found node, determine whether the target content has been cached in the cloud network node; if the target content is cached in the cloud network node, send a response message for the second user request to the coordination node ; if the target content is not cached in the cloud network node, forward the second user request to the source server, so that the source server returns a response message according to the second user request; receive the response message; calculate the HASH value of the target content in the response message; based on the obtained HASH value of the target content, cache the target content to the cloud network node corresponding to the HASH value, and store the response The message is forwarded to the coordinating node.

In a third aspect, an embodiment of the present invention provides a coordinated caching method based on edge computing, which is applied to an edge network node in an information center network, and the method includes:

Receive a third user request, where the third user request is a user request sent by the requesting node, and the third user request is used to obtain the target content; determine whether the target content is cached in each edge network node; if the edge network node has cached the target content; If the target content is cached in the network node, the target content is sent to the requesting node; if the target content is not cached in the edge network nodes, the third user request is forwarded to the requesting node. coordinating node, so that the coordinating node inquires whether the identification corresponding to the target content is recorded in the edge content placement information table ECT; receiving a response message sent by the coordinating node and containing the target content after adding the second identification document; calculate a cache probability; based on the cache probability, cache the target content, and send the target content to the requesting node.

Optionally, the step of calculating the cache probability includes: calculating the cache probability by using a second preset formula, and the second preset formula is expressed as:

In the formula, p _p (r) represents the cache probability of edge network node r, R represents the number of intermediate nodes between the requesting node and the coordinating node, and D _rS represents the distance between the edge network node r and the coordinating node.

In a fourth aspect, an embodiment of the present invention provides an edge computing-based coordination cache device, which is applied to a coordination layer in an information center network, where the information center network includes an origin server, a cloud network layer, a coordination layer, and an edge network layer, the cloud network layer includes a number of cloud network nodes, the coordination layer includes a number of coordination nodes, the edge network layer includes a number of edge network nodes, and the coordination layer is respectively connected with the cloud network layer and the edge network layer. A communication link, the apparatus includes:

a first receiving module, configured to receive a first user request, where the first user request is a user request forwarded by the edge network node, and the first user request is used to obtain target content; a query module, used to query edge content Whether the mark corresponding to the target content is recorded in the placement information table ECT; if the mark corresponding to the target content is recorded in the ECT, then the routing information recorded in the ECT is used to send the first user The request is forwarded to the edge network node that caches the target content; if the mark corresponding to the target content is not recorded in the ECT, the first user request is forwarded to the cloud network closest to the coordinating node node, so that the cloud network node processes the first user request; a second receiving module is configured to receive a response message forwarded by the cloud network node, where the response message includes the target content; the first a calculation module, configured to calculate the cache utility value of the target content through a first preset formula; a first judgment module, used to judge whether the cache utility value of the target content is less than a preset threshold; a first sending module, using If the cache utility value is less than a preset threshold, add a first identifier to the target content; the first identifier is used to indicate that the target content does not need to be cached in the edge network node; for adding the first identifier After the target content, send a response message containing the target content to the edge network node; the second sending module is configured to add a cache value to the target content if the cache utility value is greater than or equal to a preset threshold a second identifier; the second identifier is used to indicate that the target content needs to be cached in the edge network node; for the target content after adding the second identifier, a response message containing the target content is sent to the edge network node .

In a fifth aspect, an embodiment of the present invention provides an edge computing-based coordination cache device, which is applied to a cloud network layer in an information center network, and the device includes:

A third receiving module, configured to receive a second user request, where the second user request is a user request forwarded by the coordinating node, the second user request is used to obtain target content, and the second user request carries There is the identification information of the target content; the second calculation module is used to calculate the hash HASH value corresponding to the identification information of the target content, and searches for the cloud corresponding to the HASH value in the distributed hash table DHT a network node; a second judging module, used for judging whether the target content has been cached in the cloud network node when the cloud network node corresponding to the HASH value is found; a third sending module, used for if the cloud network node is in the cloud network If the target content is cached in the node, the response message for the second user request is sent to the coordinating node; the fourth sending module is configured to, if the target content is not cached in the cloud network node, then forward the second user request to the source server, so that the source server returns a response message according to the second user request; a fourth receiving module is configured to receive the response message; a third computing module, For calculating the HASH value of the target content in the response message; the first cache module is used to cache the target content to the cloud network node corresponding to the HASH value based on the obtained HASH value of the target content , and forward the response message to the coordinating node.

In a sixth aspect, an embodiment of the present invention provides an edge computing-based coordination cache device, which is applied to an edge network layer in an information center network, and the device includes:

The fifth receiving module is used to obtain a third user request, the third user request is a user request sent by the requesting node, and the third user request is used to obtain the target content; the third judgment module is used to judge each edge network Whether the target content is cached in the node; the fifth sending module is configured to send the target content to the requesting node if the target content is cached in the edge network node; the sixth sending module is configured to use If the target content is not cached in the edge network node, the third user request is forwarded to the coordinating node, so that the coordinating node inquires whether the edge content placement information table ECT records whether there is any content related to the target content. the identifier corresponding to the target content; the sixth receiving module is used to receive the response message sent by the coordinating node and contains the target content after adding the second identifier; the fourth calculation module is used to calculate the cache probability; the second A caching module, configured to cache the target content based on the caching probability, and send the target content to the requesting node.

In a seventh aspect, an embodiment of the present invention provides a coordinated caching system based on edge computing, including: an origin server, a cloud network node, a coordination node and an edge network node, the origin server communicates directly with the cloud network node, so The coordinating node is in communication connection with the cloud network node and the edge network node, respectively.

In an edge computing-based coordinated caching method, device, and system provided by the embodiments of the present invention, after the coordinating node receives a response message, it can calculate the caching utility value of the target content, and when the caching utility value of the target content is less than a preset threshold , adding a first identifier to the target content to indicate that the target content does not need to be cached in the edge network node; when the cache utility value of the target content is greater than or equal to the preset threshold, add a first identifier to the target content to indicate that the target content needs to be cached The second identifier cached in the edge network node, and then, for the target content after adding the first or second identifier, a response message containing the target content is sent to the edge network node, because the embodiment of the present invention can pass the coordinating node The edge network node is controlled to cache the target content that needs to be cached instead of all the target content. Therefore, the embodiment of the present invention can reduce the redundancy degree of content caching and reduce the occupation of cache space, thereby saving network resources.

In the method, device and system for coordinated caching based on edge computing provided by the embodiments of the present invention, after receiving the response message, the cloud network node calculates the HASH value of the target content in the response message, and based on the obtained HASH value of the target content, Cache the target content to the cloud network node corresponding to the HASH value, and forward the response message to the coordinating node. Since the embodiment of the present invention can cache the target content to the cloud network node corresponding to the HASH value based on the obtained HASH value of the target content, the target content is cached corresponding to a cloud network node, and does not need to be cached in other cloud network nodes. Repeated caching reduces the redundancy of content caching, reduces space occupation, and saves network resources.

In an edge computing-based coordinated caching method, device, and system provided by an embodiment of the present invention, an edge network node calculates a caching probability after receiving a response message, caches target content based on the caching probability, and sends the target content to request node. In the embodiment of the present invention, the target content is cached based on the calculated caching probability, instead of caching the target content in all edge network nodes on the transmission path after receiving the response message, thereby reducing the redundancy of content caching , while reducing the occupation of the cache space, thereby saving network resources, further enabling the user to quickly obtain the content when the user requests the content, and improving the cache hit rate. Of course, it is not necessary for any product or method of the present invention to achieve all of the advantages described above at the same time.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a schematic diagram of a network structure according to an embodiment of the present invention;

FIG. 2 is a flowchart of a first method for coordinated caching based on edge computing provided by an embodiment of the present invention;

3 is a flowchart of a second method for coordinated caching based on edge computing provided by an embodiment of the present invention;

4 is a flowchart of a third method for coordinated caching based on edge computing provided by an embodiment of the present invention;

5 is a flowchart of a fourth method for coordinated caching based on edge computing provided by an embodiment of the present invention;

6 is a flowchart of a fifth edge computing-based coordinated caching method provided by an embodiment of the present invention;

7 is a flowchart of a sixth method for coordinated caching based on edge computing provided by an embodiment of the present invention;

FIG. 8 is a schematic diagram of signaling interaction of an edge computing-based coordinated cache system according to an embodiment of the present invention;

9a is a simulation diagram of a relationship between a server load reduction rate and a popularity factor change provided by an embodiment of the present invention;

Fig. 9b is a simulation diagram of a relationship between a reduction rate of routing hops and a change in popularity factor provided by an embodiment of the present invention;

FIG. 9c is a simulation diagram of the relationship between cache redundancy and popularity factor variation provided by an embodiment of the present invention;

9d is a simulation diagram of a server load reduction rate provided by an embodiment of the present invention;

9e is a simulation diagram of a reduction rate of routing hops provided by an embodiment of the present invention;

FIG. 9f is a simulation diagram of the change of cache redundancy with the increase of cache capacity according to an embodiment of the present invention;

FIG. 10 is a schematic structural diagram of a first edge computing-based coordination cache device provided by an embodiment of the present invention;

FIG. 11 is a schematic structural diagram of a second type of edge computing-based coordination cache device provided by an embodiment of the present invention;

FIG. 12 is a schematic structural diagram of a third edge computing-based coordination cache device according to an embodiment of the present invention;

FIG. 13 is a schematic structural diagram of a fourth edge computing-based coordination cache device provided by an embodiment of the present invention;

FIG. 14 is a schematic structural diagram of a fifth edge computing-based coordination cache device provided by an embodiment of the present invention;

15 is a schematic structural diagram of a sixth kind of edge computing-based coordination cache device provided by an embodiment of the present invention;

16 is a schematic structural diagram of an electronic device according to an embodiment of the present invention;

17 is a schematic structural diagram of another electronic device provided by an embodiment of the present invention;

FIG. 18 is a schematic structural diagram of another electronic device provided by an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In order to solve the problem that the cache redundancy of the content in the existing ICN network is relatively large, occupies more space, and thus wastes network resources, the embodiments of the present invention provide a coordinated caching method, device and system based on edge computing.

In the embodiment of the present invention, the ICN network includes: a source server, a cloud network layer, a coordination layer, and an edge network layer. The cloud network layer includes a number of cloud network nodes, the coordination layer includes a number of coordination nodes, and the edge network layer includes a number of edge network nodes. The layers are respectively connected to the cloud network layer and the edge network layer, and the source server is connected to the cloud network layer. Referring to FIG. 1, FIG. 1 is a schematic diagram of a network structure according to an embodiment of the present invention. The coordination node of the coordination layer and the cloud network node of the cloud network layer can be directly communicated and connected, and one coordination node can manage multiple edge network nodes. Referring to Fig. 1, an edge network node can be divided into multiple neighborhoods, each neighborhood is controlled by a coordinating node, and there are multiple edge network nodes in one neighborhood. An edge computing-based coordinated caching method, device, and system provided by the embodiments of the present invention should be in the ICN network.

In the embodiment of the present invention, the coordination nodes included in the coordination layer may be selected from edge network nodes that can be directly communicated with the cloud network nodes before the coordination layer is divided in the embodiment of the present invention. The edge network node that is directly communicatively connected to the cloud network node can communicate with multiple user terminals to manage multiple user terminals, and can also directly communicate with the cloud network node. Therefore, the coordination layer formed by the coordination nodes selected from the edge network nodes that can be directly communicatively connected to the cloud network node can be respectively connected to the cloud network layer and the edge network layer. Specifically, a person skilled in the art can select a coordination node from edge network nodes according to actual requirements, which is not limited in this embodiment of the present invention.

Referring to FIG. 2, FIG. 2 is a first edge computing-based coordination caching method provided by an embodiment of the present invention, applied to a coordination node, and the method may include:

S201. Receive a first user request. The first user request is a user request forwarded by the edge network node, and the first user request is used to acquire the target content. In practical applications, the first user requests the target content to be acquired, which may be, for example, video files, news, and express messages.

S202 , query whether a mark corresponding to the target content is recorded in the ECT (Edge Content Table, edge content placement information table). In this embodiment of the present invention, when the coordinating node receives the first user request, the first user request is the user request forwarded by the edge network node, which may be called the third user request, and it is queried whether there is a record corresponding to the target content in the ECT mark. The ECT records markers corresponding to different contents and routing information corresponding to the markers. The routing information records the information of the edge network node corresponding to the storage target contents, and the marker corresponding to each content is unique.

S203 , if a mark corresponding to the target content is recorded in the ECT, forward the first user request to the edge network node where the target content is cached by using the routing information recorded in the ECT. If a mark corresponding to the target content is recorded in the ECT, the routing information recorded in the ECT can be directly used to forward the first user request to the edge network node that caches the target content, so that the edge network node that caches the target content sends the target content to the edge network node. The content is returned to the requesting node.

S204, if the mark corresponding to the target content is not recorded in the ECT, forward the first user request to the cloud network node closest to the coordinating node, so that the cloud network node processes the first user request. If no mark corresponding to the target content is recorded in the ECT, indicating that the target content is not cached in the edge network nodes controlled by the coordinator node, the first user request can be forwarded to the cloud network node closest to the coordinator node, so that the cloud The network node processes the first user request. In practical applications, forwarding the first user request to the cloud network node closest to the coordinating node may be forwarding the first user request to the cloud network node with the least number of hops from the coordinating node. In the embodiment of the present invention, the minimum number of hops refers to the minimum number of intermediate nodes passed on the transmission path from the coordinator node to the edge network node.

In the embodiment of the present invention, the coordinating node queries whether a mark corresponding to the target content is recorded in the ECT for the received first user request, and if so, it can forward the first user request to the edge network node that caches the target content . To prevent the requesting node from not requesting the target content in the edge network nodes on its request path, and further requesting the target content from the cloud network node, the requesting node can obtain the requested target content more quickly.

Referring to FIG. 3, FIG. 3 is a second edge computing-based coordination caching method provided by an embodiment of the present invention, applied to a coordination node, and the method may include:

S301: Receive a response message forwarded by the cloud network node, where the response message includes target content. In the embodiment of the present invention, a possible implementation manner is: the response message received by the coordinating node and forwarded by the cloud network node is: after the edge network node receives the user request sent by the requesting node for acquiring the target content, The target content is not obtained in the edge network node, the user request is forwarded by the edge network node to the coordinator node, and the coordinator node does not query and record the mark corresponding to the target content in the ECT, and the user request is further forwarded to the cloud network node. . When the cloud network node caches the target content, it returns a response message containing the target content.

Another possible implementation manner of the embodiment of the present invention is: the response message received by the coordinating node and forwarded by the cloud network node is: after the edge network node receives the user request sent by the requesting node for acquiring the target content, If the target content is not obtained in the network node, the user request is forwarded by the edge network node to the coordinator node, and the coordinator node does not query and record the mark corresponding to the target content in the ECT, and the user request is further forwarded to the cloud network node. The cloud network node does not cache the target content and forwards the user request to the origin server. After the origin server receives the user request, it returns a response message containing the target content for the user request.

S302: Calculate the cache utility value of the target content by using the first preset formula.

In the embodiment of the present invention, after receiving the response message, the coordinating node may calculate the cache utility value of the target content included in the response message by using a first preset formula, and the first preset formula may be expressed as:

In the formula, CU(i) represents the cache utility value of the target content i, P(i) represents the popularity of the target content i, Size(i) represents the storage space occupied by the target content i, and CR(i) represents the cache of the target content i. The network revenue ratio of the target content is used to indicate the frequency of the target content being requested; the network revenue ratio of the target content cache is used to indicate the percentage of time saved by the target content caching at the edge network nodes.

In the embodiment of the present invention, the step of obtaining the popularity of the target content may include: calculating the popularity of the target content by using a second preset formula, and the second preset formula may be expressed as:

In the formula, counter(i) represents the number of times the target content i is requested in a statistical period, and counter represents the total number of times that all contents are requested in a statistical period.

Denote P _i =(p _i1 , p _i2 ,...p _ik ) ^T , then p _ik represents the popularity of the target content i calculated in the first-order cycle, and p _i is regarded as a quadratic function of the statistical cycle to obtain a quadratic regression model :

p _i =b _i0 +b _i1 t+b _i2 t ² +ε, where t=1, 2, 3, ... k, ε～N(0,σ ² )

In the formula, b _i0 , b _i1 , and b _i2 are the coefficients of the quadratic regression model, t is the variable of the quadratic regression model, and t ₁ =t, t ₂ =t ² , the quadratic regression model is converted into two Metalinear regression model:

p _i =b _i0 +b _i1 t ₁ +b _i2 t ₂ +ε, where t ₁ =1, 2, 3,...k, t ₂ =1 ² , 2 ² , 3 ² ,...k ²

In the formula, the coefficient matrix is:

Denote B _i =(b _i0 , b _i1 , b _i2 ) ^T , then the estimated value of B _i is available Denote M=(T ^T T) ^-1 T ^T , then the matrix M can be determined when k is determined, that is, M can be calculated offline, and we can get:

In the formula, represents the estimated value of _pi , which will be as the popularity of the target content i.

In the embodiment of the present invention, the step of obtaining the network revenue ratio of the target content cache may include: calculating, by using a third preset formula, when the target content requested by the node n (that is, the requesting node n) requesting the target content is cached in the edge network The average network delay of , the third preset formula can be expressed as:

In the formula, D ₁ represents the average network delay when the target content requested by the requesting node n is cached in the edge network, Represents the probability that the target content requested by the requesting node n gets a response from the edge network, represents the propagation delay from the n nodes of the edge network layer I to the m nodes of the coordination layer, represents the propagation delay from the m node of the coordination layer to the n node of the edge network layer I, represents the sum of the transmission delays from the n nodes of the edge network layer I to the m nodes of the coordination layer, represents the sum of the transmission delays from the m nodes of the coordination layer to the n nodes of the edge network layer I, Represents the probability that the target content requested by the requesting node n gets a response at the cloud network layer, represents the propagation delay from the m node of the coordination layer to the g node of the cloud network layer C, represents the propagation delay from the g node of the cloud network layer C to the m node of the coordination layer, represents the sum of the sending delays from the m node of the coordination layer to the g node of the cloud network layer C, It represents the sum of the sending delays from the g node of the cloud network layer C to the m node of the coordination layer.

The fourth preset formula is used to calculate the average network delay when the target content requested by the requesting node n is not cached in the edge network. The fourth preset formula can be expressed as:

In the formula, D ₂ represents the average network delay when the target content requested by the requesting node n is not cached in the edge network.

The fifth preset formula is used to calculate the total delay of processing the target content requested by the requesting node n in the edge network cache. The fifth preset formula can be expressed as:

In the formula, H _i represents the total delay of the target content i requested by the requesting node n in the edge network cache processing, J represents the node set of the edge network cached content, A _i,j represents the target content i cached on the node j. processing delay.

Through the sixth preset formula, the network revenue ratio of the target content i cache is calculated, and the sixth preset formula can be expressed as:

In the formula, CR(i) represents the network revenue ratio of the target content i cached.

S303, if the cache utility value is less than the preset threshold, add a first identifier to the target content. The first identifier is used to indicate that the target content does not need to be cached in the edge network node.

S304, for the target content added with the first identifier, send a response message including the target content to the edge network node. In the embodiment of the present invention, when the cache utility value of the calculated target content is less than the preset threshold, a first identifier for indicating that the target content does not need to be cached in the edge network node is added to the target content, and the target content will be included The response message is sent to the edge network node. Specifically, those skilled in the art can set the preset threshold according to actual requirements, which is not limited in this embodiment of the present invention.

S305, if the cache utility value is greater than or equal to the preset threshold, add a second identifier to the target content. The second identifier is used to indicate that the target content needs to be cached in the edge network node.

S306, for the target content added with the second identifier, send a response message including the target content to the edge network node. In the embodiment of the present invention, when the obtained cache utility value is greater than or equal to the preset threshold, a second identifier for indicating that the target content needs to be cached in the edge network node is added to the target content, and a response containing the target content is added. The message is sent to the edge network node to control the edge network node to cache the target content that needs to be cached.

In the embodiment of the present invention, after the coordinating node receives the response message, it can calculate the cache utility value of the target content, and when the cache utility value of the target content is less than the preset threshold, add a message to the target content to indicate that the target content does not need to be stored in the target content. The first identifier cached in the edge network node; when the cache utility value of the target content is greater than or equal to the preset threshold, a second identifier for indicating that the target content needs to be cached in the edge network node is added to the target content, and then, for After adding the target content with the first or second identifier, the response message containing the target content is sent to the edge network node. Because the embodiment of the present invention can control the edge network node to cache the target content that needs to be cached through the coordination node, and Not all target contents are cached, therefore, the embodiment of the present invention can reduce the redundancy degree of content cache and reduce the occupation of cache space, thereby saving network resources.

Referring to FIG. 4, FIG. 4 is a third edge computing-based coordinated caching method provided by an embodiment of the present invention, applied to a cloud network node, and the method may include:

S401. Receive a second user request. The second user request may be a user request forwarded by the coordinating node, the second user request is used to obtain the target content, and the second user request carries the identification information of the target content. In the embodiment of the present invention, when the coordinating node receives the first user request and finds that the mark corresponding to the target content is not recorded in the ECT, the first user request is forwarded to the cloud network node. The cloud network node receives a second user request, where the second user request is the first user request forwarded by the coordinating node, the second user request is used to obtain the target content, and the second user request carries the identification information of the target content. Optionally, the identification information of the target content may be information used to identify the target content, such as a title of the target content, a relevant mark of the target content, etc., which is not limited in this embodiment of the present invention.

S402: Calculate a HASH value corresponding to the identification information of the target content, and search for a cloud network node corresponding to the HASH value in a distributed hash table DHT (Distributed Hash Table, distributed hash table). The DHT is a distributed storage method, and reference may be made to the implementation manner of the prior art for a specific storage process thereof, and details are not described herein again in this embodiment of the present invention. When the cloud network node receives the second user request forwarded by the coordinating node, it can calculate the hash HASH value corresponding to the identification information of the target content carried in the second user request, and then search the DHT for the hash value corresponding to the HASH value. Cloud network node.

S403, when the cloud network node corresponding to the HASH value is found, determine whether the target content has been cached in the cloud network node.

S404, if the target content is cached in the cloud network node, send a response message for the second user request to the coordinating node. If the target content is cached in the cloud network node, that is, the second user request can be responded to at the cloud network node, the cloud network node will generate a response message for the second user request, and then send the generated response message to the The message is sent to the coordinating node.

S405, if the target content is not cached in the cloud network node, forward the second user request to the origin server, so that the origin server returns a response message according to the second user request. If the target content is not cached in the cloud network node, that is, the cloud network node cannot respond to the second user request, the cloud network node will forward the second user request to the origin server, so that the origin server can respond to the second user request according to the second user request. The request returns a response message. In practical applications, the cloud network node may forward the second user request to the source server according to a preset FIB (Forwarding Information Table, forwarding information table). Specifically, the cloud network node forwards the second user request to the source server. manner, which is not limited in this embodiment of the present invention.

In the embodiment of the present invention, the cloud network node calculates the HASH value corresponding to the identification information of the target content for the received second user request, and searches the DHT for the cloud network node corresponding to the HASH value. When the target content is cached, a response message is generated for the second user request, and the response message is sent to the coordinating node, so that the response message can be quickly returned to the requesting node requesting the target content.

Referring to FIG. 5, FIG. 5 is a fourth edge computing-based coordinated caching method provided by an embodiment of the present invention, applied to a cloud network node, and the method may include:

S501, a response message is received. In this embodiment of the present invention, the response packet received by the cloud network node may be: when the source server receives the second user request forwarded by the cloud network node, the source server returns a response packet according to the second user request, the response The message contains the target content requested by the second user request.

S502: Calculate the HASH value of the target content in the response message. In the embodiment of the present invention, the HASH value of the identification information corresponding to the target content may be calculated. Specifically, for the process of calculating the HASH value of the identification information corresponding to the target content, reference may be made to the calculation process of the HASH value in the prior art, and the embodiment of the present invention is described here. No longer.

S503, based on the obtained HASH value of the target content, cache the target content in the cloud network node corresponding to the HASH value, and forward the response message to the coordinating node. For the process of caching the target content to the cloud network node corresponding to the HASH value, reference may be made to the HASH caching process in the prior art, and details are not described herein again in this embodiment of the present invention.

Since the embodiment of the present invention can cache the target content to the cloud network node corresponding to the HASH value based on the obtained HASH value of the target content, the target content is cached corresponding to a cloud network node, and does not need to be cached in other cloud network nodes. Repeated caching reduces the redundancy of content caching, reduces space occupation, and saves network resources.

Referring to FIG. 6, FIG. 6 is a fifth edge computing-based coordinated caching method provided by an embodiment of the present invention, applied to an edge network node, and the method may include:

S601. Receive a third user request. The third user request is a user request sent by the requesting node, and the third user request is used to acquire the target content.

S602: Determine whether the target content is cached in each edge network node.

S603, if the target content is cached in the edge network node, send the target content to the requesting node.

S604, if the target content is not cached in the edge network nodes, forward the third user request to the coordinator node, so that the coordinator node queries whether the edge content placement information table ECT records an identifier corresponding to the target content.

In this embodiment of the present invention, after receiving the third user request, the edge network node queries each edge network node whether the target content is cached. target content, the third user request is forwarded to the coordinating node, so that the coordinating node inquires whether an identifier corresponding to the target content is recorded in the edge content placement information table ECT. So that the third user requests to further request the target content from the coordinating node when the edge network node does not receive a response.

Referring to FIG. 7, FIG. 7 is a sixth edge computing-based coordinated caching method provided by an embodiment of the present invention, applied to an edge network node, and the method may include:

S701: Receive a response message sent by the coordinating node and including the target content added with the second identifier. In the embodiment of the present invention, the edge network node receives the response message sent by the coordinating node and includes the target content added with the second identifier, that is, the target content in the response message is the content that needs to be cached by the edge network node.

S702, calculating a cache probability. When the edge network node receives the response message containing the target content that needs to be cached, the cache probability of the edge network node is calculated. Specifically, the step of calculating the cache probability may include: calculating the cache probability by using a second preset formula, and the second preset formula may be expressed as:

In the formula, p _p (r) represents the cache probability of edge network node r, R represents the number of intermediate nodes between the requesting node and the coordinating node, and D _rS represents the distance between the edge network node r and the coordinating node.

S703, based on the caching probability, cache the target content, and send the target content to the requesting node. In this embodiment of the present invention, after calculating the cache probability, the calculated cache probability is input into a preset cache module to obtain the output of whether to cache the target content, further cache the target content, and send the target content to the requesting node. In the embodiment of the present invention, the preset caching module determines whether to cache the target content. Reference may be made to the module in the open source simulation tool NDNSim. The module can implement: input the caching probability and output the result of whether to cache the content.

In the embodiment of the present invention, the target content is cached based on the calculated caching probability, instead of caching the target content in all edge network nodes on the transmission path after receiving the response message, thereby reducing the redundancy of content caching , while reducing the occupation of the cache space, thereby saving network resources, further enabling the user to quickly obtain the content when the user requests the content, and improving the cache hit rate.

The following describes an edge computing-based coordinated cache provided by an embodiment of the present invention with reference to specific application examples.

It is assumed that a neighborhood in the edge network contains multiple edge network nodes, the multiple edge network nodes are controlled by a coordination node, the coordination node communicates with the cloud network node, and the cloud network node communicates with the source server. The specific information The transfer process is shown in Figure 8. The user request arrives at the edge network node. If the target content is cached in the edge network node, the target content is sent to the requesting node. If the target content is not cached in each edge network node, the user request passes through the edge network node and arrives at the coordinator. node. The coordinating node receives the user request and queries whether the tag corresponding to the target content is recorded in the ECT. If the tag corresponding to the target content is recorded in the ECT, the user request is forwarded to the cache with the target content by using the routing information recorded in the ECT. The edge network node, so that the edge network node that caches the target content returns the target content to the requesting node. If there is no mark corresponding to the target content recorded in the ECT, the user request is forwarded to the cloud network node closest to the coordinating node. The cloud network node receives the user request, calculates the HASH value corresponding to the identification information of the target content, and searches the DHT for the cloud network node corresponding to the HASH value. If the target content is cached in the cloud network node, the response message for the user request is sent to the coordination node, the coordination node determines whether the target content needs to be cached at the edge network node, and sends the response message to the edge network node; If the target content is not cached in the cloud network node, the user request is forwarded to the origin server. The origin server receives the user request, and returns a response message to the cloud network node according to the user request, so that the cloud network node caches the target content. The cloud network node receives the response message, calculates the HASH value of the target content in the response message, and based on the HASH value, caches the target content to the cloud network node corresponding to the HASH value, and forwards the response message to the coordinating node . The coordinating node receives the response message, calculates the cache utility value of the target content, and based on the cache utility value, adds an identifier to the target content to indicate whether the target content needs to be cached in the edge network node, and then adds the identifier to the response. The message is sent to the edge network node. The edge network node receives the response message, calculates the cache probability, caches the target content based on the cache probability, and sends the target content to the requesting node.

In the embodiment of the present invention, since the cloud network node can cache the target content to the cloud network node corresponding to the HASH value based on the obtained HASH value of the target content, the target content is cached corresponding to one cloud network node, and does not need to be cached in another cloud network node. The coordinating node can control the edge network nodes to cache the target content that needs to be cached, instead of caching all the target content, and cache the target content based on the calculated caching probability, instead of caching all the target content. After receiving the response message, the target content is cached in all edge network nodes on the transmission path, reducing the redundancy of content caching and reducing the occupation of cache space, thereby saving network resources, and further enabling users to request content when Get the content quickly and improve the cache hit rate.

Exemplarily, in the embodiment of the present invention, different caching methods are used to cache the target content, and the open-source simulation tool NDNSim is used for simulation experiments, wherein the open-source operating system Ubuntu uses 14.04, the standard library function libboost uses 1.55, and the simulator ndnSIM uses 1.0. ndnSIM1.0 is an NS-3 module that implements NDN (Named Data Networking). The experiment constructs a network topology that conforms to intra-domain and inter-domain characteristics. The number of nodes is 50, including 32 edge routers, 17 cloud routers and 1 origin server. Each router contains a certain amount of cache content and runs the corresponding cache. method. The user request distribution obeys the Zipf Zipf distribution, and the request arrival conforms to the Poisson process. The specific parameter settings are shown in Table 1 below.

Table 1 Parameter setting table

parameter settings Defaults scope Zipf 1 0.7～1.3 number of contents 10000 / cache block size/chunk 50 10～90 Request rate/(request s^-1) 100 /

In the example of the present invention, three performance evaluation indicators are considered: SLRR (server load reduction ratio, server load reduction ratio): the ratio of the number of Interest packets that do not reach the server in the network to the total number of Interest packets, and its expression is:

In the formula, L_sum represents the number of Interest packets that did not reach the server, and R_sum represents the total number of Interest packets. The server service load reduction rate reflects the reduction rate of the server-side load pressure caused by the cache responding to user requests. The higher the indicator is, the smaller the server-side load pressure is, and the more obvious the caching effect is.

AHRR (average hop reduction ratio, routing hop reduction ratio): The ratio of the distance saved by the interest packet hitting the node to the routing hops required by the user to obtain content on the source server.

In the formula, hop_c represents the number of hops passed by the Interest packet after using a certain caching method, and hop_nc represents the number of hops passed by the Interest packet when the cache is not used.

CR (cache redundancy, cache redundancy) is expressed as:

In the formula, Content_counts represents the number of all contents in the cache, and type_counts represents the number of all types of content generated. This indicator reflects the redundancy of the content in the cache. The lower the value, the lower the redundancy and the richer the cache types. The minimum value of 1 means that the cached content is not repeated.

In the embodiment of the present invention, the user's content request obeys the Zipf(N, θ) distribution, where N is the total content in the network, θ is the inclination factor, and the smaller the θ value, the more obvious the content concentration. In the example of the present invention, different caching methods are used to cache the target content, and the obtained simulation graphs are shown in Fig. 9a to Fig. 9f respectively. Among them, the NEC (No Edge Caching, no edge caching is set) caching method is as follows: content is only cached in cloud network nodes, and content is not cached in edge network nodes. The ECC (Edge Cloud Cacing, edge cloud caching) caching method is the caching method adopted in the embodiment of the present invention. The AEC (All Edge Caching, All Edge Caching) caching method is as follows: content is only cached in edge network nodes, and content is not cached in cloud network nodes. As the inclination factor θ increases, the concentration of content requests increases, and the popularity of the content becomes higher.

9a is a simulation diagram of the relationship between the server load reduction rate and the popularity factor change provided by the embodiment of the present invention. The figure shows that three different caching methods make the server load reduction rate and the popularity factor change different. The increase in popularity has made the cached content of each node converge, and the performance gap of the three methods has been reduced. ECC has established caches on both the edge and the cloud to increase the number of cached content to satisfy most content requests. In other words, the caching method of the embodiment of the present invention makes the server load reduction rate relatively high. Fig. 9b is a simulation diagram of the relationship between the reduction rate of routing hops and the change of popularity factor provided by the embodiment of the present invention. The figure shows that three different caching methods make the relationship between the reduction rate of routing hops and the change of popularity factor different. When the requests are concentrated, the edge cache can satisfy most of the content requests, while the NEC method does not perform edge caching. Compared with the other two methods, the location of the cache is farther from the requester, and the reduction rate of routing hops is less improved. Comparatively speaking, the caching method of the embodiment of the present invention enables a relatively high reduction rate of routing hops. FIG. 9c is a simulation diagram of the relationship between cache redundancy and popularity factor variation provided by an embodiment of the present invention. The figure shows that three different caching methods make the relationship between cache redundancy and popularity factor different. HASH cache, each kind of content is only cached by one node. Therefore, for the NEC method, the redundancy of the cache is always kept at 1, that is, there is no redundancy, while the AEC method uses cache everywhere, so compared with the other two methods. In comparison, the cache method of the embodiment of the present invention makes the cache redundancy relatively small. FIG. 9d is a simulation diagram of a server load reduction rate provided by an embodiment of the present invention. The diagram shows that three different caching methods make the server load reduction rate different. When the cache capacity is small, the node cache space is limited, and the cache content cannot be fully satisfied. In the AEC method, the edge nodes will perform a large number of replacements along the way, and the server load reduction rate is lower than that of NEC and ECC. As the cache space increases, the content that can be cached by the node increases. In comparison, the server load rate under the three methods has increased, but the gap between NEC and AEC has gradually decreased, and ECC has always maintained good performance. , the server load reduction rate is relatively high. FIG. 9e is a simulation diagram of the reduction rate of routing hops provided by an embodiment of the present invention. The figure shows that three different caching methods make the reduction rate of routing hops different. Due to the increase of node cache capacity, the number of contents that can be cached by the network increases. The average hop count reduction rate of the three methods has increased. In comparison, the caching method of the embodiment of the present invention makes the route hop count reduction rate relatively high. FIG. 9f is a simulation diagram of the change of cache redundancy with the increase of cache capacity according to an embodiment of the present invention. The figure shows that three different caching methods make the change of cache redundancy with the increase of cache capacity different, and the comparison can be seen. , the cache method of the embodiment of the present invention makes the change of the cache redundancy smaller with the increase of the cache capacity.

The above simulation results show that among the three caching methods, NEC has a great advantage in reducing cache redundancy, while other cache performance indicators are not as good as AEC and ECC, and AEC is not very bad in redundancy performance. , and proper content redundancy can bring better cache performance improvement. The ECC of the embodiment of the present invention can improve the server load reduction rate and reduce the cache redundancy at the same time.

Corresponding to the aforementioned first edge computing-based coordinated caching method, in another aspect of the implementation of the present invention, a first edge computing-based coordinated caching device is also provided. 10 is a schematic structural diagram of a first edge computing-based coordination cache device provided by an embodiment of the present invention, which is applied to the coordination layer, and the device includes:

The first receiving module 101 is configured to receive a first user request, where the first user request is a user request forwarded by an edge network node, and the first user request is used to acquire target content. The query module 102 is used for querying whether a mark corresponding to the target content is recorded in the edge content placement information table ECT; if a mark corresponding to the target content is recorded in the ECT, the routing information recorded in the ECT is used to request the first user Forward to the edge network node that caches the target content; if the mark corresponding to the target content is not recorded in the ECT, forward the first user request to the cloud network node closest to the coordinating node, so that the cloud network node processes the first user ask. In the embodiment of the present invention, the coordinating node queries whether a mark corresponding to the target content is recorded in the ECT for the received first user request, and if so, forwards the first user request to the edge network node that caches the target content. To prevent the requesting node from not requesting the target content in the edge network nodes on its request path, and further requesting the target content from the cloud network node, the requesting node can obtain the requested target content more quickly.

It should be noted that the device in this embodiment of the present invention is a device corresponding to the first edge computing-based coordinated caching method shown in FIG. 2 , and all implementations of the first edge computing-based coordinated caching method shown in FIG. 2 All examples are applicable to the device, and all can achieve the same or similar beneficial effects.

Corresponding to the aforementioned second edge computing-based coordinated caching method, in another aspect of the implementation of the present invention, a second edge computing-based coordinated caching device is also provided. 11 is a schematic structural diagram of a second edge computing-based coordination cache device provided by an embodiment of the present invention, which is applied to the coordination layer, and the device includes:

The second receiving module 111 is configured to receive a response message forwarded by the cloud network node, where the response message includes target content. The first calculation module 112 is configured to calculate the cache utility value of the target content by using the first preset formula. The first judging module 113 is configured to judge whether the cache utility value of the target content is less than a preset threshold. The first sending module 114 is configured to add a first identifier to the target content if the cache utility value is less than the preset threshold; the first identifier is used to indicate that the target content does not need to be cached in the edge network node; target content, and send a response message containing the target content to the edge network node. The second sending module 115 is configured to add a second identifier to the target content if the cache utility value is greater than or equal to the preset threshold; the second identifier is used to indicate that the target content needs to be cached in the edge network node; for adding the second identifier After the target content, the response message containing the target content is sent to the edge network node.

It should be noted that the apparatus in this embodiment of the present invention is an apparatus corresponding to the second edge computing-based coordinated caching method shown in FIG. 3 , and all implementations of the second edge computing-based coordinated caching method shown in FIG. 3 All examples are applicable to the device, and all can achieve the same or similar beneficial effects.

Corresponding to the aforementioned third edge computing-based coordinated caching method, in another aspect of the implementation of the present invention, a third edge computing-based coordinated caching device is also provided. 12 is a schematic structural diagram of a third edge computing-based coordination cache device provided by an embodiment of the present invention, which is applied to the cloud network layer, and the device includes:

The third receiving module 121 is configured to receive a second user request, where the second user request is a user request forwarded by the coordinating node, the second user request is used to obtain target content, and the second user request carries identification information of the target content. The second calculation module 122 is configured to calculate the hash HASH value corresponding to the identification information of the target content, and search for the cloud network node corresponding to the HASH value in the DHT. The second judging module 123 is configured to judge whether the target content has been cached in the cloud network node when the cloud network node corresponding to the HASH value is found. The third sending module 124 is configured to send a response message for the second user request to the coordinating node if the target content is cached in the cloud network node. The fourth sending module 125 is configured to forward the second user request to the origin server if the target content is not cached in the cloud network node, so that the origin server returns a response message according to the second user request.

It should be noted that the device in this embodiment of the present invention is a device corresponding to the third edge computing-based coordinated caching method shown in FIG. 4 , and all implementations of the third edge computing-based coordinated caching method shown in FIG. 4 All examples are applicable to the device, and all can achieve the same or similar beneficial effects.

Corresponding to the foregoing fourth edge computing-based coordinated caching method, in another aspect of the implementation of the present invention, a fourth edge computing-based coordinated caching device is also provided. 13 is a schematic structural diagram of a fourth edge computing-based coordination cache device provided by an embodiment of the present invention, which is applied to the cloud network layer, and the device includes:

The fourth receiving module 131 is configured to receive a response message. The third calculation module 132 is configured to calculate the HASH value of the target content in the response message. The first cache module 133 is configured to cache the target content to the cloud network node corresponding to the HASH value based on the obtained HASH value of the target content, and forward the response message to the coordination node.

It should be noted that the apparatus in this embodiment of the present invention is an apparatus corresponding to the fourth edge computing-based coordinated caching method shown in FIG. 5 , and all implementations of the fourth edge computing-based coordinated caching method shown in FIG. 5 All examples are applicable to the device, and all can achieve the same or similar beneficial effects.

Corresponding to the aforementioned fifth edge computing-based coordinated caching method, in another aspect of the implementation of the present invention, a fifth edge computing-based coordinated caching device is also provided. 14 is a schematic structural diagram of a fifth edge computing-based coordination cache device provided by an embodiment of the present invention, which is applied to the edge network layer, and the device includes:

The fifth receiving module 141 is configured to obtain a third user request, where the third user request is a user request sent by the requesting node, and the third user request is used to obtain target content. The third judgment module 142 is configured to judge whether the target content is cached in each edge network node. The fifth sending module 143 is configured to send the target content to the requesting node if the target content is cached in the edge network node. The sixth sending module 144 is configured to forward the third user request to the coordinating node if the target content is not cached in the edge network nodes, so that the coordinating node queries the edge content placement information table ECT whether there is a record corresponding to the target content. logo.

It should be noted that the device in this embodiment of the present invention is a device corresponding to the fifth edge computing-based coordinated caching method shown in FIG. 6 , and all implementations of the fifth edge computing-based coordinated caching method shown in FIG. 6 All examples are applicable to the device, and all can achieve the same or similar beneficial effects.

Corresponding to the foregoing sixth edge computing-based coordinated caching method, in another aspect of the implementation of the present invention, a sixth edge computing-based coordinated caching device is also provided. FIG. 15 is a schematic structural diagram of a sixth edge computing-based coordination cache device provided by an embodiment of the present invention, which is applied to the edge network layer, and the device includes:

The sixth receiving module 151 is configured to receive a response message sent by the coordinating node and including the target content added with the second identifier. The fourth calculation module 152 is used to calculate the cache probability. The second caching module 153 is configured to cache the target content based on the caching probability, and send the target content to the requesting node.

It should be noted that the apparatus in this embodiment of the present invention is an apparatus corresponding to the sixth edge computing-based coordinated caching method shown in FIG. 7 , and all implementations of the sixth edge computing-based coordinated caching method shown in FIG. 7 All examples are applicable to the device, and all can achieve the same or similar beneficial effects.

Corresponding to the aforementioned coordinated caching method based on edge computing, in another aspect of the implementation of the present invention, a coordinated caching system based on edge computing is also provided. The system includes: a source server, a cloud network node, a coordination node and an edge network node, the source server communicates directly with the cloud network node, and the coordination node communicates with the cloud network node and the edge network node respectively;

The edge network node is used for: receiving the third user request sent by the requesting node for obtaining the target content; judging whether the target content is cached in each edge network node; if the target content is cached in the edge network node, sending the target content to requesting node; if the target content is not cached in the edge network nodes, the third user request is forwarded to the coordinating node, so that the coordinating node inquires whether the edge content placement information table ECT records an identifier corresponding to the target content; or, receiving Coordinating the response message sent by the node including the target content after adding the second identifier; calculating the cache probability; based on the cache probability, cache the target content, and send the target content to the requesting node;

The coordinating node is used to: receive the first user request for acquiring the target content forwarded by the edge network node; query whether the edge content placement information table ECT records a mark corresponding to the target content; if there is a mark corresponding to the target content recorded in the ECT mark, then use the routing information recorded in the ECT to forward the first user request to the edge network node that caches the target content; if the mark corresponding to the target content is not recorded in the ECT, then forward the first user request to the distance coordination the cloud network node closest to the node, so that the cloud network node can process the first user request; or, receive a response message containing the target content forwarded by the cloud network node; calculate the cache utility value of the target content through the first preset formula; If the cache utility value is less than the preset threshold, add a first identifier to the target content; the first identifier is used to indicate that the target content does not need to be cached in the edge network node; for the target content after adding the first identifier, the target content will be included The response message of the content is sent to the edge network node; if the cache utility value is greater than or equal to the preset threshold, a second identifier is added to the target content; the second identifier is used to indicate that the target content needs to be cached in the edge network node; for adding For the target content after the second identification, a response message containing the target content is sent to the edge network node;

The cloud network node is used to: receive the second user request carrying the identification information of the target content forwarded by the coordination node, calculate the HASH value corresponding to the identification information of the target content, and search the distributed hash table DHT for the corresponding HASH value When the cloud network node corresponding to the HASH value is found, it is determined whether the target content has been cached in the cloud network node; if the target content is cached in the cloud network node, it will respond to the second user request The message is sent to the coordination node; if the target content is not cached in the cloud network node, the second user request is forwarded to the origin server, so that the origin server returns a response message according to the second user request; or, receives the response message ; Calculate the HASH value of the target content in the response message; Based on the obtained HASH value of the target content, cache the target content to the cloud network node corresponding to the HASH value, and forward the response message to the coordination node;

The origin server is used to: return a response message to the user request forwarded by the cloud network node.

Exemplarily, in this embodiment of the present invention, the cloud network node may be a cloud router, the coordination node may be a server, and the edge network node may be an edge router. Set with edge network nodes, which is not limited in this embodiment of the present invention.

In the coordinated caching system according to the embodiment of the present invention, since the cloud network node can cache the target content to the cloud network node corresponding to the HASH value based on the obtained HASH value of the target content, the target content corresponds to a cloud network node for caching, and does not It needs to be cached repeatedly in other cloud network nodes, and the coordinating node can control the edge network nodes to cache the target content that needs to be cached, instead of caching all the target content, and cache the target content based on the calculated caching probability. , instead of caching the target content in all edge network nodes on the transmission path after receiving the response message, reducing the redundancy of content caching and reducing the occupancy of cache space, thereby saving network resources and further enabling users to When requesting content, quickly obtain the content to improve the cache hit rate.

An embodiment of the present invention further provides an electronic device, as shown in FIG. 16 , including a processor 161 , a communication interface 162 , a memory 163 and a communication bus 164 , wherein the processor 161 , the communication interface 162 , and the memory 163 pass through the communication bus 164 To complete the mutual communication, the memory 163 is used to store computer programs; the processor 161 is used to execute the program stored in the memory 163 to realize a method provided by the embodiment of the present invention and applied to the coordination nodes in the information center network. A coordinated caching method based on edge computing.

In the electronic device provided by the embodiment of the present invention, since the coordinating node can control the edge network node to cache the target content that needs to be cached, instead of caching all the target content, the embodiment of the present invention can reduce the content cache The degree of redundancy, while reducing the occupancy of cache space, thus saving network resources.

An embodiment of the present invention further provides another electronic device, as shown in FIG. 17 , including a processor 171, a communication interface 172, a memory 173, and a communication bus 174, wherein the processor 171, the communication interface 172, and the memory 183 pass through the communication bus 174 completes mutual communication, the memory 173 is used to store computer programs; the processor 171 is used to implement the cloud network node applied in the information center network provided by the embodiment of the present invention when executing the program stored in the memory 173 A coordinated caching method based on edge computing.

Another electronic device provided by the embodiment of the present invention can cache the target content to a cloud network node corresponding to the HASH value based on the obtained HASH value of the target content, so the target content is cached corresponding to a cloud network node, and does not It is necessary to repeat the caching in other cloud network nodes to reduce the redundancy of content caching, reduce space occupation, and save network resources.

An embodiment of the present invention further provides another electronic device, as shown in FIG. 18 , including a processor 181, a communication interface 182, a memory 183, and a communication bus 184, wherein the processor 181, the communication interface 182, and the memory 183 pass through the communication bus 184 completes mutual communication, the memory 183 is used to store the computer program; the processor 181 is used to implement the edge network node applied in the information center network provided by the embodiment of the present invention when executing the program stored in the memory 183 A coordinated caching method based on edge computing.

Another electronic device provided by the embodiment of the present invention caches the target content based on the calculated caching probability, instead of caching the target content in all edge network nodes on the transmission path after receiving the response message , reducing the redundancy of the content cache and reducing the occupancy of the cache space, thereby saving network resources, further enabling the user to obtain the content quickly when the content is requested, and improving the cache hit rate.

The communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI for short) bus or an Extended Industry Standard Architecture (EISA for short) bus or the like. The communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the above electronic device and other devices. The memory may include random access memory (Random Access Memory, RAM for short), and may also include non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor. The above-mentioned processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, referred to as CPU), a network processor (Network Processor, referred to as NP), etc.; may also be a digital signal processor (Digital Signal Processing, referred to as DSP), Application Specific Integrated Circuit (ASIC for short), Field-Programmable Gate Array (FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

An embodiment of the present invention also provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, when the computer-readable storage medium runs on a computer, the computer is made to execute any one of the foregoing embodiments and apply to an information center An edge computing-based coordination caching method for coordinating nodes in the network, cloud network nodes, or origin servers to obtain the same technical effect.

It should be noted that the device, system, electronic device, and storage medium in the embodiments of the present invention are the device, electronic device, and storage medium that apply the above-mentioned edge computing-based coordination caching method, respectively, and the above-mentioned edge computing-based coordination All embodiments of the caching method are applicable to the apparatus, system, electronic device and storage medium, and can achieve the same or similar beneficial effects.

It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any relationship between these entities or operations. any such actual relationship or sequence exists. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

Each embodiment in this specification is described in a related manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. Especially, for the apparatus/system electronic device/storage medium embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the partial description of the method embodiment.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims (9)

1. a kind of coordination caching method based on edge calculations, which is characterized in that applied to the coordination section in information centre's network Point, information centre's network include: source server, cloud network layer, cooperation layer and edge network layer, the cloud network Layer includes several cloud network nodes, and the cooperation layer includes several coordinator nodes, and the edge network layer includes several edges Network node, the cooperation layer are communicated to connect with the cloud network layer and the edge network layer respectively, which comprises

The first user request is received, first user request is user's request of edge network node forwarding, described the One user requests for obtaining object content；

Label corresponding with the object content whether has been recorded in inquiry edge content placement information table ECT；

If record has label corresponding with the object content in the ECT, believed using the routing recorded in the ECT First user request is forwarded to the edge network node for being cached with the object content by breath；

If not recorded label corresponding with the object content in the ECT, first user request is forwarded to The cloud network node nearest apart from the coordinator node, so that the first user described in the cloud network node processing requests；

The response message forwarded by the cloud network node is received, the response message includes the object content；

By the first preset formula, the caching value of utility of the object content is calculated；

If the caching value of utility is less than preset threshold, first identifier is added in the object content；First mark Knowledge does not need to cache in edge network node for indicating the object content；

For the object content after addition first identifier, the response message comprising the object content is sent to the edge network Node；

If the caching value of utility is greater than or equal to preset threshold, second identifier is added in the object content；It is described Second identifier caches in edge network node for indicating the object content needs；

For the object content after addition second identifier, the response message comprising the object content is sent to the edge network Node.

2. the method according to claim 1, wherein first preset formula indicates are as follows:

In formula, CU (i) indicates that the caching value of utility of object content i, P (i) indicate that the popularity of object content i, Size (i) indicate The memory space that object content i is occupied, CR (i) indicate the network profit ratio of object content i caching；Wherein, the object content Popularity for indicating the requested frequent degree of the object content；The network profit ratio of the object content caching is used for Indicate the percentage of time that the object content is saved in edge network node caching.

3. a kind of coordination caching method based on edge calculations, which is characterized in that applied to the cloud net in information centre's network Network node, information centre's network include: source server, cloud network layer, cooperation layer and edge network layer, the cloud Network layer packet contains several cloud network nodes, and the cooperation layer includes several coordinator nodes, and the edge network layer includes several Edge network node, the cooperation layer are communicated to connect with the cloud network layer and the edge network layer respectively, the method Include:

Second user request is received, the second user request is user's request of coordinator node forwarding, and described second uses Family request carries the identification information of the object content for obtaining object content, and in second user request；

Calculate Hash HASH value corresponding to the identification information of the object content, and in distributed hashtable DHT search with The corresponding cloud network node of the HASH value；

When finding the corresponding cloud network node of the HASH value, judge whether to have cached in the cloud network node described Object content；

It, will be for the response report of second user request if being cached with the object content in the cloud network node Text is sent to the coordinator node；

If the uncached object content in the cloud network node, second user request is forwarded to source service Device, so that the source server requests returning response message according to the second user；

Receive the response message；

Calculate the HASH value of the object content in the response message；

Based on the HASH value of the obtained object content, the object content is cached to cloud net corresponding with the HASH value Network node, and the response message is forwarded to coordinator node.

4. a kind of coordination caching method based on edge calculations, which is characterized in that applied to the edge net in information centre's network Network node, information centre's network include: source server, cloud network layer, cooperation layer and edge network layer, the cloud Network layer packet contains several cloud network nodes, and the cooperation layer includes several coordinator nodes, and the edge network layer includes several Edge network node, the cooperation layer are communicated to connect with the cloud network layer and the edge network layer respectively, the method Include:

Third user request is received, the third user request is user's request that requesting node is sent, and the third user asks It asks for obtaining object content；

Judge the object content whether is cached in each edge network node；

If being cached with the object content in the edge network node, the object content is sent to the request and is saved Point；

If the uncached object content in the edge network node, third user request is forwarded to described Coordinator node, so as to whether recorded and the object content in coordinator node inquiry edge content placement information table ECT Corresponding mark；

Receive the response message comprising the object content after addition second identifier that the coordinator node is sent；

Calculate caching probability；

Based on the caching probability, the object content is cached, and the object content is sent to the request and is saved Point.

5. according to the method described in claim 4, it is characterized in that, the step of calculating caches probability, comprising:

Caching probability is calculated by the second preset formula, second preset formula indicates are as follows:

In formula, p_p(r) indicate that the caching probability of edge network node r, R indicate requesting node to the middle node between coordinator node The quantity of point, D_rSIndicate edge network node r at a distance from coordinator node.

6. a kind of coordination buffer storage based on edge calculations, which is characterized in that applied to the cooperation layer in information centre's network, Information centre's network includes: source server, cloud network layer, cooperation layer and edge network layer, the cloud network layer packet Containing several cloud network nodes, the cooperation layer includes several coordinator nodes, and the edge network layer includes several edge networks Node, the cooperation layer are communicated to connect with the cloud network layer and the edge network layer respectively, and described device includes:

First receiving module, for receiving the first user request, the first user request is that the edge network node forwards User's request, first user request is for obtaining object content；

Enquiry module, for inquiring in edge content placement information table ECT whether recorded mark corresponding with the object content Note；If record has label corresponding with the object content in the ECT, using the routing iinformation recorded in the ECT, First user request is forwarded to the edge network node for being cached with the object content；If do not recorded in the ECT There is label corresponding with the object content, then first user request is forwarded to the cloud nearest apart from the coordinator node Network node is held, so that the first user described in the cloud network node processing requests；

Second receiving module, for receiving the response message forwarded by the cloud network node, the response message includes institute State object content；

First computing module, for calculating the caching value of utility of the object content by the first preset formula；

First judgment module, for judging whether the caching value of utility of the object content is less than preset threshold；

First sending module, if adding the in the object content for the caching value of utility less than preset threshold One mark；The first identifier does not need to cache in edge network node for indicating the object content；For addition the Object content after one mark, is sent to the edge network node for the response message comprising the object content；

Second sending module, if being greater than or equal to preset threshold for the caching value of utility, in the object content Add second identifier；The second identifier caches in edge network node for indicating the object content needs；For adding Response message comprising the object content is sent to the edge network node by the object content after adding second identifier.

7. a kind of coordination buffer storage based on edge calculations, which is characterized in that applied to the cloud net in information centre's network Network layers, information centre's network include: source server, cloud network layer, cooperation layer and edge network layer, the cloud net Network layers include several cloud network nodes, and the cooperation layer includes several coordinator nodes, and the edge network layer includes several sides Edge network node, the cooperation layer are communicated to connect with the cloud network layer and the edge network layer respectively, described device packet It includes:

Third receiving module, for receiving second user request, the second user request is the use of coordinator node forwarding Family request, the second user request carry in the target for obtaining object content, and in second user request The identification information of appearance；

Second computing module, for calculating Hash HASH value corresponding to the identification information of the object content, and in distribution Cloud network node corresponding with the HASH value is searched in Hash table DHT；

Second judgment module, for judging the cloud network section when finding the corresponding cloud network node of the HASH value Whether the object content has been cached in point；

Third sending module, if for being cached with the object content in the cloud network node, it will be for described the The response message of two users request is sent to the coordinator node；

4th sending module, if used for the uncached object content in the cloud network node by described second Family request is forwarded to source server, so that the source server requests returning response message according to the second user；

4th receiving module, for receiving the response message；

Third computing module, for calculating the HASH value of the object content in the response message；

First cache module, for the HASH value based on the obtained object content, by the object content cache to this The corresponding cloud network node of HASH value, and the response message is forwarded to coordinator node.

8. a kind of coordination buffer storage based on edge calculations, which is characterized in that applied to the edge net in information centre's network Network layers, information centre's network include: source server, cloud network layer, cooperation layer and edge network layer, the cloud net Network layers include several cloud network nodes, and the cooperation layer includes several coordinator nodes, and the edge network layer includes several sides Edge network node, the cooperation layer are communicated to connect with the cloud network layer and the edge network layer respectively, described device packet It includes:

5th receiving module, for obtaining third user request, the third user request is that the user that requesting node is sent asks It asks, the third user request is for obtaining object content；

Third judgment module, for judging whether be cached with the object content in each edge network node；

5th sending module, if for being cached with the object content in the edge network node, it will be in the target Appearance is sent to the requesting node；

6th sending module, if for the uncached object content in the edge network node, by the third Whether user's request is forwarded to the coordinator node, so as to remember in coordinator node inquiry edge content placement information table ECT Record has mark corresponding with the object content；

6th receiving module, for receiving that the coordinator node sends comprising the object content after addition second identifier Response message；

4th computing module, for calculating caching probability；

Second cache module caches the object content, and for being based on the caching probability by the object content It is sent to the requesting node.

9. a kind of coordination caching system based on edge calculations characterized by comprising source server, cloud network node, association Point of adjustment and edge network node, the source server and the cloud network node direct communication, the coordinator node point It is not communicated to connect with the cloud network node and the edge network node；

The edge network node is used for: receiving the third user request for obtaining object content that requesting node is sent；Sentence Break and whether is cached with the object content in each edge network node；If being cached with the target in the edge network node The object content is then sent to the requesting node by content；If the uncached mesh in the edge network node Content is marked, then third user request is forwarded to the coordinator node, so that coordinator node inquiry edge content is put It sets and whether has recorded mark corresponding with the object content in information table ECT；

Alternatively, receiving the response message comprising the object content after addition second identifier that the coordinator node is sent；Meter Calculate caching probability；Based on the caching probability, the object content is cached, and the object content is sent to described Requesting node；

The coordinator node is used for: the first user for obtaining object content for receiving the edge network node forwarding asks It asks；Label corresponding with the object content whether has been recorded in inquiry edge content placement information table ECT；If the ECT Middle record has label corresponding with the object content, then using the routing iinformation recorded in the ECT, by first user Request is forwarded to the edge network node for being cached with the object content；If do not recorded in the ECT in the target Hold corresponding label, then first user request is forwarded to the cloud network node nearest apart from the coordinator node, with Request the first user described in the cloud network node processing；

Alternatively, receiving the response message comprising the object content forwarded by the cloud network node；It is default by first Formula calculates the caching value of utility of the object content；If the caching value of utility is less than preset threshold, in the target First identifier is added in content；The first identifier does not need to delay in edge network node for indicating the object content It deposits；For the object content after addition first identifier, the response message comprising the object content is sent to the edge network Node；If the caching value of utility is greater than or equal to preset threshold, second identifier is added in the object content；It is described Second identifier caches in edge network node for indicating the object content needs；For the target after addition second identifier Response message comprising the object content is sent to the edge network node by content；

The cloud network node is used for: receiving the second of the identification information for carrying object content of the coordinator node forwarding User's request, calculates Hash HASH value corresponding to the identification information of the object content, and look into distributed hashtable DHT Look for cloud network node corresponding with the HASH value；When finding the corresponding cloud network node of the HASH value, judgement The object content whether has been cached in the cloud network node；If be cached in the target in the cloud network node Hold, then will be sent to the coordinator node for the response message of second user request；If in the cloud network node In the uncached object content, then by the second user request be forwarded to source server so that the source server according to The second user requests returning response message；

Alternatively, receiving the response message；Calculate the HASH value of the object content in the response message；Described in obtaining The HASH value of object content caches the object content to cloud network node corresponding with the HASH value, and by the sound Message is answered to be forwarded to coordinator node；

The source server is used for: being requested for the user that the cloud network node is forwarded, returning response message.