CN114065095A

CN114065095A - Network node acceleration method and device, computer readable medium and electronic equipment

Info

Publication number: CN114065095A
Application number: CN202111446711.8A
Authority: CN
Inventors: 陈凡尔; 黄哲武; 冯紫隽; 卞正皑
Original assignee: China Telecom Corp Ltd
Current assignee: China Telecom Corp Ltd
Priority date: 2021-11-30
Filing date: 2021-11-30
Publication date: 2022-02-18

Abstract

The application relates to the field of network communication, and discloses a network node acceleration method, a network node acceleration device, a computer readable medium and electronic equipment. The method comprises the following steps: acquiring access data to a website; acquiring the application types of the detection data and the appointed website of the depth data packet; training according to the detection data of the depth data packet and the application type to obtain an application type identification model; calculating to obtain various weights based on various weight calculation methods by taking the website as a network node; determining the important value of the network node according to the multiple weights; constructing characteristic data of the network nodes; training based on the feature data and the important values to obtain a network node important value identification model; acquiring an important value of a target network node according to the network node important value identification model; determining the application type of the target network node according to the application type recognition model and the target depth data packet detection data; and performing acceleration operation according to the application type and the importance value. The method improves the accuracy and flexibility of website acceleration.

Description

Network node acceleration method and device, computer readable medium and electronic equipment

Technical Field

The present application relates to the field of network communication technologies, and in particular, to a network node acceleration method, an apparatus, a computer-readable medium, and an electronic device.

Background

At present, when network acceleration is carried out, acceleration can only be carried out on a preset fixed website, and due to the fact that the number of websites on the internet is large, a large amount of labor cost is needed for selecting the website needing acceleration, and once the website is selected, the website can not be changed, so that the flexibility is poor; meanwhile, the accelerated website is still in the experience of people seriously when being selected, and the behavior of a user for accessing the website is changed all the time, so the website acceleration accuracy is poor.

Disclosure of Invention

In the field of network communication technologies, to solve the foregoing technical problems, an object of the present application is to provide a network node acceleration method, apparatus, computer readable medium, and electronic device.

According to an aspect of an embodiment of the present application, there is provided a network node acceleration method, including:

simulating to access a website of a website and a multi-level external link website linked with the website through a network request to obtain access data corresponding to each website, wherein the access data comprises access performance data;

acquiring deep data packet detection data of a user and an application type of a specified website, wherein the deep data packet detection data records behavior data of the user accessing the specified website;

training according to the application types of the deep data packet detection data and the specified website to obtain an application type identification model;

taking the website as a network node, and calculating to obtain various weights respectively corresponding to the network nodes based on various weight calculation methods according to the access data and the depth data packet detection data corresponding to the network nodes;

aiming at each network node, determining the important value of the network node according to various weights corresponding to the network node;

constructing feature data of the network node according to the access performance data and the depth data packet detection data of the network node;

training based on the feature data and the important value of each network node to obtain a network node important value identification model;

when target depth data packet detection data is acquired, feature data of the target network node is constructed according to the target depth data packet detection data, the feature data is input into the network node important value identification model, and the important value of the target network node is acquired, wherein the target depth data packet detection data records behavior data of a user accessing the target network node;

determining the application type of the target network node according to the application type identification model and the target depth data packet detection data;

and carrying out network node acceleration operation according to the application type of the target network node and the important value of the target network node.

According to an aspect of an embodiment of the present application, there is provided a network node acceleration apparatus, including:

the simulation access unit is used for simulating and accessing the website of the website and the multi-level external link website linked by the website through a network request to obtain access data corresponding to each website, wherein the access data comprises access performance data;

the device comprises an acquisition unit, a storage unit and a processing unit, wherein the acquisition unit is used for acquiring the application type of a depth data packet detection data and an appointed website of a user, and the depth data packet detection data records the behavior data of the user accessing the appointed website;

the first training unit is used for training according to the deep data packet detection data and the application type of the specified website to obtain an application type recognition model;

the computing unit is used for computing various weights respectively corresponding to the network nodes based on various weight computing methods according to the access data and the depth data packet detection data corresponding to the network nodes by taking the website as the network node;

the important value determining unit is used for determining the important value of each network node according to the multiple weights corresponding to the network node;

a constructing unit, configured to construct feature data of the network node according to the access performance data and the deep packet probe data of the network node;

the second training unit is used for training based on the feature data and the important values of all the network nodes to obtain a network node important value recognition model;

the system comprises a constructing and inputting unit, a judging unit and a judging unit, wherein the constructing and inputting unit is used for constructing characteristic data of a target network node according to target depth data packet detection data when the target depth data packet detection data is obtained, inputting the characteristic data to a network node important value identification model and obtaining an important value of the target network node, and the target depth data packet detection data records behavior data of a user accessing the target network node;

the application type determining unit is used for determining the application type of the target network node according to the application type identification model and the target deep data packet detection data;

and the accelerating unit is used for carrying out network node accelerating operation according to the application type of the target network node and the important value of the target network node.

According to an aspect of embodiments of the present application, there is provided a computer-readable medium, on which a computer program is stored, which, when executed by a processor, implements the method as described in the above embodiments.

According to an aspect of an embodiment of the present application, there is provided an electronic device including:

one or more processors;

storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method as described in the embodiments above.

The technical scheme provided by the embodiment of the application can have the following beneficial effects:

the network node acceleration method provided by the application comprises the following steps: simulating to access a website of a website and a multi-level external link website linked with the website through a network request to obtain access data corresponding to each website, wherein the access data comprises access performance data; acquiring deep data packet detection data of a user and an application type of a specified website, wherein the deep data packet detection data records behavior data of the user accessing the specified website; training according to the application types of the deep data packet detection data and the specified website to obtain an application type identification model; taking the website as a network node, and calculating to obtain various weights respectively corresponding to the network nodes based on various weight calculation methods according to the access data and the depth data packet detection data corresponding to the network nodes; aiming at each network node, determining the important value of the network node according to various weights corresponding to the network node; constructing feature data of the network node according to the access performance data and the depth data packet detection data of the network node; training based on the feature data and the important value of each network node to obtain a network node important value identification model; when target depth data packet detection data is acquired, feature data of the target network node is constructed according to the target depth data packet detection data, the feature data is input into the network node important value identification model, and the important value of the target network node is acquired, wherein the target depth data packet detection data records behavior data of a user accessing the target network node; determining the application type of the target network node according to the application type identification model and the target depth data packet detection data; and carrying out network node acceleration operation according to the application type of the target network node and the important value of the target network node.

In the method, a website and a multi-level external link of the website are crawled in a request simulating manner to obtain corresponding access data, meanwhile, deep data packet detection data and the application type of the website when a user accesses the website are obtained, and then deep data packet detection data and an application type identification model are respectively established according to the obtained data; when a user accesses some websites, the corresponding application type identification model is obtained, then, the important value and the application type of the website accessed by the user and the external link thereof can be determined according to the established deep data packet detection data and the application type identification model, and the network node acceleration operation is carried out based on the important value and the application type. Therefore, the network nodes needing to be accelerated and the strategy of accelerating the network nodes can be dynamically determined in real time according to the access behaviors of the user, and the accuracy and the flexibility of website acceleration are improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.

Fig. 1 is a system architecture diagram illustrating a network node acceleration method in accordance with an exemplary embodiment;

FIG. 2 is a flow diagram illustrating a network node acceleration method in accordance with an exemplary embodiment;

FIG. 3A is a diagram illustrating data obtained by calling an interface to access a page in accordance with an illustrative embodiment;

fig. 3B is a schematic diagram illustrating data obtained after cleaning the data acquired by calling the interface according to an exemplary embodiment;

FIG. 4 is a schematic diagram illustrating simplified processing of link relationships of network nodes in accordance with an illustrative embodiment;

FIG. 5 is a schematic diagram illustrating a calculation process for calculating weights based on a weight calculation method, according to an exemplary embodiment;

FIG. 6 is a diagram illustrating the partitioning of application types for different web sites in accordance with an illustrative embodiment;

FIG. 7 is a weighted network node association graph shown in accordance with an example embodiment;

FIG. 8 is an overall flow diagram illustrating a network node acceleration method in accordance with an exemplary embodiment;

FIG. 9 is a schematic diagram illustrating technical gist in steps of an embodiment of the present application according to an exemplary embodiment;

FIG. 10 is a comparative page schematic diagram illustrating the effects of network node acceleration in accordance with an exemplary embodiment;

FIG. 11 is a block diagram illustrating a network node acceleration apparatus in accordance with an exemplary embodiment;

FIG. 12 illustrates a schematic structural diagram of a computer system suitable for use in implementing the electronic device of an embodiment of the present application.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

Furthermore, the drawings are merely schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities.

In the related art, only the acceleration can be performed according to the manually specified website, but the website access behavior of the user is constantly changing, so that the final website may have a small user access amount, and thus acceleration resources are wasted.

In addition, some schemes are based on tests of four-layer TCP protocol, seven-layer HTTP(s) protocol and websocket protocol to detect network performance, and meanwhile, protocol feature analysis and mathematical statistics such as TCP session time, message analysis and other schemes are developed to identify and accelerate network application. However, since data on the network is encrypted, and meanwhile, the network applications are various and the protocol types are complex, the acceleration efficiency is low, and it is difficult to accurately determine the network applications that need to be accelerated.

Therefore, the application firstly provides a network node acceleration method, which can overcome the defects, efficiently and accurately determine the network node needing acceleration, accelerate, and dynamically adjust the acceleration strategy according to the network access behavior of the user, thereby more accurately allocating the acceleration resource.

The implementation terminal of the present application may be any device having computing, processing, and communication functions, and the device may be connected to an external device for receiving or sending data, and specifically may be a portable mobile device, such as a smart phone, a tablet computer, a notebook computer, a pda (personal Digital assistant), or the like, or may be a fixed device, such as a computer device, a field terminal, a desktop computer, a server, a workstation, or the like, or may be a set of multiple devices, such as a physical infrastructure of cloud computing or a server cluster.

Optionally, the implementation terminal of the present application may be a server or a physical infrastructure of cloud computing.

Fig. 1 is a system architecture diagram illustrating a network node acceleration method according to an example embodiment. As shown in fig. 1, the system architecture 100 includes a terminal device 101, a server 102, and a network 103. The terminal apparatus 101 and the server 102, the server 102 and the network 103, and the terminal apparatus 101 and the network 103 are connected by communication links. The terminal device 101 is an implementation terminal in this embodiment, and when the network node acceleration method provided by the present application is applied to the system architecture shown in fig. 1, a specific process may be as follows: firstly, the terminal device 101 accesses various websites in the network 103 continuously by calling an interface for simulating the access of the websites to obtain access data of the websites and multilevel external links of the websites; then, the terminal device 101 acquires, from the server 102, deep data packet probe data and an application type of the specified website, where the deep data packet probe data is behavior data of a user in the network 103 when accessing the specified website, and is collected by the server 102; then, the terminal device 101 performs model training based on the obtained depth data packet detection data and the application type of the specified website to obtain an application type recognition model; then, on one hand, the terminal device 101 calculates the important value of the network node according to the access data and the depth data packet detection data of the network node, and on the other hand, constructs the feature data of the network node according to the access performance data and the depth data packet detection data, so as to obtain a network node important value recognition model based on the feature data and the important value training; next, the terminal device 101 obtains target depth data packet detection data to be monitored from the server 102 again, and determines the application type and the important value of the target website accessed by the user according to the target depth data packet detection data, so as to determine whether the target website and the external link need to be accelerated, an acceleration strategy, and the like according to the application type and the important value; finally, the terminal apparatus 101 sends instructions to some controllers in the network 103, thereby accelerating the determined website.

In one embodiment of the present application, the terminal device 101 is accessed in a multi-threaded manner.

It should be noted that fig. 1 is only one embodiment of the present application, and although in the embodiment of fig. 1, the implementation terminal is a common terminal device such as a desktop computer, in other embodiments of the present application, the implementation terminal may be various types of devices, such as a server; although in the embodiment of the present application, the deep packet probe data is obtained by the implementing terminal from the server, in other embodiments of the present application, the deep packet probe data may also be obtained by the implementing terminal probing the network. The embodiments of the present application are not limited in this respect, and the scope of protection of the present application should not be limited thereby.

Fig. 2 is a flow chart illustrating a network node acceleration method according to an example embodiment. The network node acceleration method provided by the embodiment is executed by a server, and as shown in fig. 2, the method may include the following steps:

step 210, simulating to access the website of the website and the multi-level external link websites linked by the website through a network request, and obtaining access data corresponding to each website.

The access data includes access performance data.

Specifically, a tool such as Selenium can be used, based on technologies such as simulated login, webpage driving and browser agent, the website and the second-level and third-level website links thereof can be traversed and cyclically crawled in a web crawler mode, and in the crawling process, for the website needing to input the verification code, the verification code link can be automatically skipped; for a website needing to input an account password for logging in, the account password can be automatically input. The second-level website external link of the website is an external link directly linked with the website, and the third-level website external link is an external link directly linked with the second-level website external link.

The access performance data is loaded into the web page, and specifically may include a response state, a response time, and a content length, where the response state is a state code obtained by accessing the web address, the response time is a time for obtaining the state code, and the content length is a file size of the accessed page. Since the website is not only crawled in the step, but also the website external link of the website is crawled, the access data can include the link relation data among the websites besides the access performance data, and the link relation data can include the website, the website name, the specific description information and the like. After the access data is obtained through the emulation access, the access data can be packaged into an output data interface for system call.

FIG. 3A is a diagram illustrating data obtained by calling an interface to access a page in accordance with an illustrative embodiment. As shown in fig. 3A, when the page of each website is accessed through the call interface, information such as response state, response time, and length related to the accessed page may be obtained, and the information is related to the access performance of the page.

In an embodiment of the application, the simulating, by the network request, access to a website of a website and a website of a multi-level external link linked to the website to obtain access data corresponding to each website includes:

acquiring interface data corresponding to each website by simulating the website for accessing the website and the multistage external link websites linked by the website;

and carrying out data cleaning operation on the interface data to obtain access data corresponding to each website.

Specifically, the interface data may be parameter data returned by the interface in JSON format, and the parameter data may be subjected to data cleansing processing in DATAFRAME format. Performing data cleansing operations may also include regular processing based on regular expressions, and the like.

Fig. 3B is a schematic diagram illustrating data obtained after cleaning the data acquired by calling the interface according to an exemplary embodiment. Referring to fig. 3B, the finally obtained data may include fields of an INDEX (INDEX), a NAME (NAME), an ADDRESS (ADDRESS), a status code (ResponseCode), and a before-node performance map (before-acceleration-times graph), where the horizontal axis (xAxis) is the number of accesses and the vertical axis (data) is the response time obtained in each access.

In an embodiment of the present application, the simulating, by a network request, access to a website of a website and a website of a multi-level external link linked to the website includes:

the network access performance of websites far away from continents can be detected in real time in batches by simulating the website of the accessed website and the website of the multilevel external link linked by the website through the network request in a multithreading mode.

In the embodiment of the application, the website and the external links thereof are accessed in batches in real time in a multithreading mode, so that the access performance of the webpage can be monitored more efficiently.

In one embodiment of the present application, the method further comprises: and skipping the simulated access to the website to be monitored according to the monitoring abnormality when the simulated access to the website to be monitored is requested through the network.

For example, when a page is accessed, if a response state is not obtained within a predetermined time period or a complete page is not obtained within a predetermined time period or a response state is obtained 404, the simulated access to the page is skipped. The predetermined time period may be set empirically, for example, may be set to 5 seconds.

In the embodiment of the application, the access to the current website is skipped when the abnormality is monitored, so that the situation of blockage during batch access of websites can be avoided, and the performance monitoring of the websites can be smoothly carried out.

In an embodiment of the present application, the skipping the simulated access to the website to be monitored according to the monitoring of the abnormality when the simulated access to the website to be monitored is requested through the network includes:

when the simulation access to the website to be monitored is requested through the network, the simulation access is performed on the website to be monitored again according to the monitoring abnormality;

and if the abnormality is monitored again, skipping the simulated access to the website to be monitored.

In the embodiment of the application, when a website is visited for the first time, if the abnormality is monitored, the simulation access is performed again instead of directly skipping, the simulation access to the website to be monitored is skipped only when the abnormality is found again, and when the abnormality is not monitored during the re-access, corresponding access data can be obtained at the moment. Therefore, the embodiment of the application can avoid missing access to the normal website, can obtain the access data of the website more accurately and thoroughly, and improves the comprehensiveness of website monitoring.

Step 220, obtaining the depth data packet detection data of the user and the application type of the appointed website.

The detection data of the depth data packet records the behavior data of the user accessing the specified website.

The deep Packet inspection data is DPI (deep Packet inspection) data, which is data obtained by the DPI device through performing inspection analysis on traffic and Packet content at key points of the network, and includes IP quintuple data, and may also include other data related to services accessed by a user. The five-element group data of the IP comprises a source IP address, a source port, a destination IP address, a destination port and a transport layer protocol, wherein the destination IP address is the address of a specified website accessed by a user. Therefore, the DPI data can be used for analyzing the internet surfing behavior of the user.

The application type is the category of the website at the application level, and can be a specific website (such as love art, Tencent video and the like) and can also be the category of the website (such as a video website, a picture website, a communication tool, a game website and the like).

The application type of the designated website may be obtained in a preset manner, and specifically, related personnel may be arranged to access the designated website belonging to the designated application type, and then extract corresponding DPI data, so that the application type of each designated website is known.

And step 230, training according to the depth data packet detection data and the application type of the specified website to obtain an application type identification model.

Because the deep data packet detection data comprises data related to the service, the deep data packet detection data corresponds to the specified website, and the service is associated with the application type, the application type recognition model can be obtained through training of the deep data packet detection data and the application type of the specified website.

The application type recognition model can be realized by algorithms such as random forest and the like.

And 240, taking the website as a network node, and calculating various weights respectively corresponding to the network nodes based on various weight calculation methods according to the access data and the depth data packet detection data corresponding to the network nodes.

The weight calculation method may use, for example, a pagerank algorithm based on pagerank, an association analysis algorithm based on Apriori, or the like. A weight corresponding to each network node may be obtained based on a weight calculation method.

In one embodiment of the present application, the method further comprises: drawing a network node directed structure diagram according to the access data and the depth data packet detection data; and associating the important value to the network node directed structure chart to obtain a weighted network node association chart, and outputting the weighted network node association chart.

In the embodiment of the application, the association graph of the weighted network nodes is output, so that a user can more intuitively master the association relation of the network nodes and the importance condition of the network nodes, and decision basis information is provided for accelerating the network nodes.

Fig. 4 is a schematic diagram illustrating simplified processing of link relationships of network nodes according to an example embodiment. As shown in fig. 4, the access data includes link relationship data between websites, and the link relationship data between websites shows directionality between nodes, so that simplified processing is performed to obtain corresponding simplified data. The depth data packet detection data comprises IP quintuple data, and the addresses of websites visited by the user can be determined according to the IP quintuple data, so that the websites continuously visited by the user can be determined according to the depth data packet detection data.

For example, "A B" and "A C" respectively represent that the a node and the B node, and the a node and the C node, which are accessed by the user in sequence, establish a link relationship. "A B" may further indicate that the node B is an extranet link of the node a, so that the corresponding network node directed structure diagram may be drawn according to the depth data packet probe data.

By associating the important values of the network nodes to the directed structure diagram of the network nodes, the information which can be displayed by the established weighted network node association diagram is more comprehensive.

The weighted network node association graph comprises network nodes and edges for connecting the network nodes, and the edges can be associated with relationship weights. Specifically, the relationship weight may be calculated when calculating the weight based on a weight calculation method, and a relationship weight corresponding to a pair of network nodes may be calculated based on a weight calculation method. The total relationship weight corresponding to a pair of network nodes can be calculated according to various relationship weights corresponding to the pair of network nodes, and then the edges in the weighted network node association diagram are drawn according to the size of the total relationship weight. For example, when the total relationship weight of a pair of network nodes exceeds a predetermined threshold, the corresponding edge may be drawn as a solid line; when the overall relationship weight for a pair of network nodes does not exceed a predetermined threshold, the corresponding edge may be drawn as a dashed line. For another example, the color of the corresponding edge may be drawn according to the total relationship weight of the network node pair, and specifically, the color of the corresponding edge gradually deepens along with the total relationship weight.

Fig. 7 is a weighted network node association graph shown in accordance with an example embodiment. Referring to fig. 7, the network nodes and the multi-level external link nodes thereof are connected by edges, which are respectively represented by solid lines and dotted lines, and represent the difference of the relationship weights between the corresponding nodes.

When the weight is calculated based on the weight calculation method, the calculation may be performed directly from the access data, or may be performed by using a network node directed structure diagram drawn based on the access data.

Fig. 5 is a diagram illustrating a calculation process for calculating weights based on a weight calculation method according to an exemplary embodiment, the calculation process being based on pagerank. See fig. 5, which shows data for five links. The data of the first link includes [ 'a', 'B' ], which represents a pair of network nodes establishing an external link relationship, and is equivalent to an edge between the pair of network nodes, and the data can be obtained by accessing the data or by reading the edge in the directed structure diagram of the network nodes; the data of the second link includes 'a', 'B', etc. network nodes, which are a collection of network nodes obtained from edges. The data of the third link is a digital node relation which is correspondingly generated according to the network nodes and the edges obtained in the previous two links. Wherein, the nodes of 'A', 'B', 'C', 'D' E 'are represented by 0,1,2,3,4,5 respectively, so that the node symbols are mapped from alphabets to Arabic numerals, and then [0,1] represents the edge formed between the nodes of A' and 'B'. The fourth link is to show a node relationship matrix, the elements of which are the relationship weights between the corresponding two network nodes. The data of the fifth link is the weight of the network node finally obtained through 8 iterations.

And step 250, aiming at each network node, determining the important value of the network node according to the multiple weights corresponding to the network node.

In an embodiment of the present application, the determining, for each network node, an importance value of the network node according to a plurality of weights corresponding to the network node includes: and inputting various weights corresponding to the network nodes into a predetermined function aiming at each network node to obtain the important value of the network node.

The predetermined function is a function set for calculating the importance value, and may be expressed in various forms of formulas. Thus, the manner in which the importance values of the network nodes are calculated may be varied. The significance values of the network nodes can be calculated in the following manner: the importance value is calculated by calculating a weighted sum of the various weights, or by removing the maximum and minimum values of the various weights and then averaging the remaining weights.

In an embodiment of the present application, the determining, for each network node, an importance value of the network node according to a plurality of weights corresponding to the network node includes: aiming at each network node, determining a candidate important value of the network node according to various weights corresponding to the network node; and adjusting the candidate important values according to expert experience to obtain important values.

In the embodiment of the application, after the candidate important value is calculated, the expert adjusts the important value, so that the accuracy of the obtained important value is improved, and the performance of a network node important value identification model obtained through subsequent training is improved.

Step 260, constructing feature data of the network node according to the access performance data and the deep packet probe data of the network node.

The feature data includes feature values respectively corresponding to the plurality of features.

Specifically, a feature value corresponding to features such as response time and content length may be constructed from the access performance data, and a feature value corresponding to features such as network node access amount may be constructed from the deep packet probe data. Of course, features of other dimensions may also be provided as desired.

And 270, training to obtain a network node important value identification model based on the feature data and the important values of the network nodes.

The network node importance value identification model can be constructed based on various algorithms, such as a linear regression algorithm. The established network node important value identification model can obtain corresponding important values according to the input feature data.

Step 280, when target depth data packet detection data is acquired, feature data of the target network node is constructed according to the target depth data packet detection data, and the feature data is input to the network node important value identification model to obtain an important value of the target network node.

The target depth data packet detection data records behavior data of a user accessing a target network node.

The target depth data packet probe data is different from the depth data packet probe data acquired in step 220. The target depth packet probe data may be, for example, DPI data that has not yet been analyzed. It is uncertain what application type the target network node belongs to. The target network node may be one or more network nodes.

Step 290, determining the application type of the target network node according to the application type identification model and the target deep data packet probe data.

The application type recognition model is trained by using the detection data of the depth data packet, so that the application type recognition model can output the application type of the corresponding target network node for the detection data of the target depth data packet.

In an embodiment of the present application, the determining an application type to which the target network node belongs according to the application type recognition model and the target deep packet probe data includes:

and if the importance value of the target network node is greater than a preset importance threshold value, determining the application type of the target network node according to the application type recognition model and the target depth data packet detection data.

In the embodiment of the application, the application type of the target network node is determined only when the importance value of the target network node is greater than the predetermined importance threshold, so that the network node can be accelerated, and the expenditure of computing resources is reduced.

clustering the target depth data packet detection data to obtain detection data corresponding to each target network node;

and respectively inputting the detection data corresponding to each target network node into the application type identification model to obtain the application type of each target network node.

In the embodiment of the application, the detection data of different network nodes can be divided by clustering the detection data of the target depth data packet, so that the application type identification model can be accurately identified.

FIG. 6 is a diagram illustrating the partitioning of application types for different web sites, according to an example embodiment.

It can be seen that different web sites can be divided into multiple application types.

Step 2110, performing network node acceleration operation according to the application type of the target network node and the important value of the target network node.

After the application type of the target network node and the important value of the target network node are obtained, the network node can be accelerated according to various strategies, and not only can the target network node be accelerated, but also other nodes related to the target network node can be accelerated. Network node acceleration operations may be performed based on SDN (Software Defined Network) technology.

In an embodiment of the present application, the performing, according to the application type to which the target network node belongs and the importance value of the target network node, a network node acceleration operation includes:

accelerating the target network node if the importance value of the target network node is greater than a predetermined importance threshold and the application type to which the target network node belongs is a target application type.

For the network node with greater importance, the network node is a high-value node in the network, and the access performance of the network node is ensured by accelerating the network node.

In one embodiment of the present application, the method further comprises:

determining the application type of the external link node linked by the target network node according to the application type identification model; and/or

Determining the important value of the external link node linked by the target network node according to the network node important value identification model;

and accelerating the external link node according to the application type of the external link node and/or the important value of the external link node.

The external link node can be accelerated only according to the application type or the important value of the external link node, and the external link node can be accelerated simultaneously according to the application type and the important value of the external link node.

In an embodiment of the present application, the accelerating the external link node according to the application type to which the external link node belongs and/or the importance value of the external link node includes: determining a corresponding acceleration strategy according to the application type of the external link node and the important value of the external link node; and carrying out acceleration operation on the external link node according to the acceleration strategy.

Corresponding acceleration strategies can be formulated aiming at the external link nodes with different application types and different important values; in addition, the outer link nodes with higher importance values can be synchronously accelerated for the target network nodes with the importance values larger than the preset importance threshold or the application types of the target network nodes.

FIG. 8 is an overall flow diagram illustrating a network node acceleration method in accordance with an exemplary embodiment; fig. 9 is a schematic diagram illustrating technical points in steps of an embodiment of the present application according to an exemplary embodiment. The solution of the embodiment of the present application is further described below with reference to fig. 8 and 9.

Referring to fig. 8, first, in a network request phase, continuously and cyclically accessing a domain name of more than two levels from a top-level domain name, so as to obtain external link information and interface data, where the domain name is a network node; then, drawing a node directed graph (a network node directed structure graph) according to the external link information and the interface data, wherein the nodes in the node directed graph comprise a website home page and a multi-level external link linked with the website home page, and all the nodes are connected through directed edges; meanwhile, DPI data is used as a training set, an application type recognition model can be obtained through training based on the training set, node weight values can be calculated according to the training set and corresponding interface data, the node weight values are important values of the network nodes, and a network node important value recognition model can be obtained through training based on the node weight values; then, taking other DPI data as a test set, and carrying out application identification through an application type identification model based on quintuple, link direction, data flow and other data in the DPI data to identify applications such as Sina and Game; meanwhile, an association analysis algorithm based on Apriori excavates frequent items to perform association analysis or performs importance value identification on network nodes through a network node importance value identification model to identify key nodes, and performs classification clustering on the key nodes to determine the application types of the key nodes. Based on the application type, key nodes needing acceleration can be determined, and then the key nodes are added into an acceleration channel to carry out network acceleration operation. Network parameters may be continuously monitored before and after network acceleration.

Referring to FIG. 9, after the log-in is simulated by inputting the website address, the method includes the following steps:

the method comprises the following steps: network request

The technology on which the step is carried out comprises a request module, regular processing, automatic interface data acquisition, webpage driving, proxy setting, multi-thread processing and the like.

Step two: data analysis processing

The technical points on which this step is based are divided into three links:

1. recording interface parameters: specifically, information such as a response status code, response time, content length, and the like may be recorded.

2. Weight calculation and application identification: the technology adopted in the link comprises webpage sequencing, association analysis, linear regression, classification clustering and the like.

3. Visualization weighting node: the technology adopted in the link comprises the following steps: network node tree, undirected and directed graph, visualization techniques.

Step three: developing applications

The scheme of the steps is applied to an actual scene, and can be particularly applied to the aspects of network monitoring, node acceleration, user internet behavior analysis and the like.

In one embodiment of the present application, the method further comprises: and correspondingly outputting a comparison page of the acceleration effect of the network node aiming at the accelerated network node.

FIG. 10 is a comparative page schematic illustrating the effects of network node acceleration according to an example embodiment. Fig. 10 shows an acceleration effect list, which includes names of network nodes, web addresses, a pre-acceleration performance data presentation graph, a post-acceleration performance data presentation graph, and performance improvement data caused by acceleration.

In the embodiment of the application, the user can intuitively master the acceleration effect of the network node by outputting the corresponding comparison page of the acceleration effect of the network node aiming at the accelerated network node.

According to a third aspect of the present application, the present application further provides a network node acceleration apparatus, and the following are apparatus embodiments of the present application.

Fig. 11 is a block diagram illustrating a network node acceleration apparatus according to an example embodiment. As shown in fig. 11, the apparatus 1100 includes:

a simulation access unit 1110, configured to simulate, through a network request, access to a website of a website and a website of a multi-level external link linked to the website, and obtain access data corresponding to each website, where the access data includes access performance data;

an obtaining unit 1120, configured to obtain deep data packet detection data of a user and an application type of a specified website, where the deep data packet detection data records behavior data of the user accessing the specified website;

a first training unit 1130, configured to obtain an application type identification model according to the deep data packet probe data and the application type training of the specified website;

a calculating unit 1140, configured to calculate, by using the website as a network node, relationship weights between network node pairs with established external link relationships according to the access data and the depth data packet detection data corresponding to the network nodes, based on multiple relationship weight calculation methods, and obtain multiple relationship weights corresponding to the network node pairs;

an important value determining unit 1150, configured to determine, for each network node, an important value of the network node according to multiple relationship weights corresponding to the network node pair where the network node is located;

a construction unit 1160, configured to construct feature data of the network node according to the access performance data and the deep packet probe data of the network node;

a second training unit 1170, configured to train to obtain a network node importance value identification model based on the feature data and the importance value of each network node;

a constructing and inputting unit 1180, configured to, when target depth data packet detection data is acquired, construct feature data of the target network node according to the target depth data packet detection data, and input the feature data to the network node important value identification model to obtain an important value of the target network node, where the target depth data packet detection data records behavior data of a user accessing the target network node;

an application type determining unit 1190, configured to determine, according to the application type identification model and the target deep data packet probe data, an application type to which the target network node belongs;

the acceleration unit 11100 is configured to perform network node acceleration operation according to the application type of the target network node and the important value of the target network node.

According to another aspect of the present application, there is also provided an electronic device capable of implementing the above method.

As will be appreciated by one skilled in the art, aspects of the present application may be embodied as a system, method or program product. Accordingly, various aspects of the present application may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.

It should be noted that the computer system 1200 of the electronic device shown in fig. 12 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.

As shown in fig. 12, the computer system 1200 includes a Central Processing Unit (CPU)1201, which can perform various appropriate actions and processes, such as performing the methods described in the above embodiments, according to a program stored in a Read-Only Memory (ROM) 1202 or a program loaded from a storage section 1208 into a Random Access Memory (RAM) 1203. In the RAM 1203, various programs and data necessary for system operation are also stored. The CPU 1201, ROM 1202, and RAM 1203 are connected to each other by a bus 1204. An Input/Output (I/O) interface 1205 is also connected to bus 1204.

The following components are connected to the I/O interface 1205: an input section 1206 including a keyboard, a mouse, and the like; an output section 1207 including a Display device such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and a speaker; a storage section 1208 including a hard disk and the like; and a communication section 1209 including a Network interface card such as a LAN (Local Area Network) card, a modem, or the like. The communication section 1209 performs communication processing via a network such as the internet. A driver 1210 is also connected to the I/O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is mounted on the drive 1210 as necessary, so that a computer program read out therefrom is mounted into the storage section 1208 as necessary.

In particular, according to embodiments of the application, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 1209, and/or installed from the removable medium 1211. The computer program executes various functions defined in the system of the present application when executed by a Central Processing Unit (CPU) 1201.

It should be noted that the computer readable medium shown in the embodiments of the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a Read-Only Memory (ROM), an Erasable Programmable Read-Only Memory (EPROM), a flash Memory, an optical fiber, a portable Compact Disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. Each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present application may be implemented by software, or may be implemented by hardware, and the described units may also be disposed in a processor. Wherein the names of the elements do not in some way constitute a limitation on the elements themselves.

As an aspect, the present application also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by an electronic device, cause the electronic device to implement the method described in the above embodiments.

It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the application. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.

Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.

Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains.

It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

1. A method for network node acceleration, the method comprising:

2. The method of claim 1, wherein the performing network node acceleration operations according to the application type to which the target network node belongs and the importance value of the target network node comprises:

3. The method of claim 2, further comprising:

4. The method of claim 1, further comprising:

drawing a network node directed structure diagram according to the access data and the depth data packet detection data;

and associating the important value to the network node directed structure chart to obtain a weighted network node association chart, and outputting the weighted network node association chart.

5. The method of claim 1, wherein simulating access to the web site of the web site and the web sites of the multi-level external links linked by the web site via the network request comprises:

and simulating to access the website of the website and the multi-level external link website linked by the website in real time through the network request in batches in a multi-thread mode.

6. The method of claim 1, further comprising:

and skipping the simulated access to the website to be monitored according to the monitoring abnormality when the simulated access to the website to be monitored is requested through the network.

7. The method of claim 6, wherein skipping the simulated access to the website to be monitored based on monitoring the abnormality when the simulated access to the website to be monitored is requested through the network comprises:

8. A network node acceleration apparatus, the apparatus comprising:

9. A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.

10. An electronic device, comprising:

one or more processors;

storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to carry out the method of any one of claims 1 to 7.