CN115225733B - Identification resolution method and device based on direct routing and dynamic quantization resolution load - Google Patents

Abstract

The present invention proposes an identity resolution method and device based on direct routing and dynamic quantization resolution load, wherein the method includes: obtaining a data packet to be resolved and sending it to an LVS server; executing a destination network address translation protocol based on the LSV server, forwarding the data packet to be resolved to a real network node of the identity resolution service according to a polling algorithm; processing the data packet to be resolved by an identity resolution server corresponding to the real network node of the identity resolution service, and recording parameter indicators during the resolution process; and returning the processing result of the data packet to be resolved by the identity resolution server. The present invention realizes an identity resolution mechanism based on a direct routing model and a dynamic quantization resolution load algorithm.

Description

Identification resolution method and device based on direct routing and dynamic quantization resolution load

Technical Field

The invention belongs to the technical field of electronic information.

Background technique

The link load of industrial networks has characteristics that are completely different from those of traditional networks. When routing industrial equipment, problems such as the access of a large number of heterogeneous devices, limitations on latency-sensitive scenarios, and instantaneous high-concurrency impacts are faced. These problems require that the identity resolution network of the industrial Internet has extremely strong robustness and flexible load balancing strategies.

The identity resolution system is similar to the domain name resolution system of the traditional network. The industrial Internet also needs to convert the device's identity into the IP address of the server that stores the information, and then obtain information resources. The identity resolution system acts as the nerve hub of interconnection in the entire industrial Internet architecture. However, from the analysis of the OSI network seven-layer model, identity resolution works on the seventh layer application layer of the network. Therefore, the operation of the identity resolution protocol needs to be implemented on the basis of establishing a reliable connection link at the transport layer. Although the three-way handshake mechanism based on the TCP protocol can ensure the reliability of the transmission link, it causes huge loss to the system performance. In the face of massive concurrency scenarios of the industrial Internet, it is difficult to ensure complex TCP connection requirements while giving full play to the performance of the system. At the same time, the links of the industrial Internet are complex, the load fluctuation rate of a single link is large, and adjacent production networks may have time-sharing situations. This determines that if the link bandwidth of each production network is increased, it will cause a large waste of network idle time resources. Therefore, the load balancing problem in the industrial Internet scenario is very urgent.

At present, there are many technical solutions related to identity resolution. The industrial Internet identity is similar to the IP address in the Internet, which accurately locates the devices in the network. Identity resolution is the retrieval process of device resources. According to the different resolution architectures, the existing identity resolution solutions can be divided into ONS architecture-based and non-ONS architecture-based solutions. The identity resolution system based on the ONS architecture includes EPC and other solutions. The solutions based on the non-ONS architecture include Handle and the most widely used solution based on the distributed hash table (DHT). Each resolution node is networked in a point-to-point manner, and the resolution entries are mapped to different storage addresses according to the DHT. This peer-to-peer networking mode ensures the distributed architecture of the resolution nodes and prevents the impact of a single node on the global situation.

With the rapid development of identity resolution technology, many industry enterprises have realized product supply chain management and life cycle management by connecting to secondary nodes and using identity resolution technology. The number of identity registrations and resolutions has therefore reached a massive level, and it is urgent to make the identity resolution system effectively handle high-concurrency identity resolution service requests. For high-concurrency requests, load balancing is one of the main solutions.

In general solutions, requests are forwarded by prepending the server IP address, that is, a server dedicated to distributing requests is placed before the server cluster that actually provides the identity resolution service. This is the load balancing server. The load balancing server forwards the received requests in a prescribed manner according to the configured algorithm, such as the random forwarding algorithm, the fast polling algorithm, etc., to achieve the effect of balancing the identity resolution requests to the identity resolution server.

The traditional solution has the following problems:

1) Random storage of identifiers does not utilize aggregate queries

Whether it is an ONS-based identity resolution solution or a non-ONS identity resolution solution, neither considers clustering the identities for industry enterprises to facilitate quick queries within the industry.

2) Identification data is difficult to regulate

As the country pays more attention to data security and the advancement of relevant laws and regulations, the regulatory scheme for identity resolution data also needs to change.

3) Single point problem of load balancing server

This process seems to solve the problem of unbalanced server load, but it aggregates the load shared by multiple servers onto the load balancing server. The key to the problem is how the load balancing server can handle the concurrent requests that a single server could not handle before. It is not difficult to find that the crux of the problem lies in the delay of the identity resolution service request. The identity resolution service is located in the seventh application layer in the network model, which is the least efficient layer in the communication layer. The service of the application layer needs to call the transmission control layer for transmission, which is a cumbersome process.

4) Load balancing algorithm performance issues

In the industrial Internet scenario, the performance of different devices varies greatly. Ordinary load balancing algorithms based on minimum number of connections and polling make it difficult to fully utilize the identity resolution servers.

Summary of the invention

The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

To this end, the first purpose of the present invention is to propose an identity resolution method based on direct routing and dynamic quantization resolution load, which is used to solve the load balancing problem in the industrial Internet scenario.

The second objective of the present invention is to provide an identification resolution device based on direct routing and dynamic quantization resolution load.

A third object of the present invention is to provide a computer device.

A fourth object of the present invention is to provide a computer-readable storage medium.

To achieve the above-mentioned purpose, the first aspect of the embodiment of the present invention proposes an identity resolution method based on direct routing and dynamic quantization of resolution load, including: obtaining a data packet to be resolved and sending it to an LVS server; executing a destination network address translation protocol based on the LSV server, and forwarding the data packet to be resolved to the real network node of the identity resolution service according to a polling algorithm; processing the data packet to be resolved through the identity resolution server corresponding to the real network node of the identity resolution service, and recording parameter indicators during the resolution process; and returning the processing result of the data packet to be resolved through the identity resolution server.

The identification resolution method based on direct routing and dynamic quantitative resolution load proposed in the embodiment of the present invention is based on the evolution of the industrial Internet identification resolution system, and designs a hierarchical identification resolution architecture based on the Chord routing protocol, which reserves a global domain for national regulatory identification resolution data, and can achieve data security and controllability. In addition, a dynamic quantitative resolution load balancing model based on the direct routing model is designed according to the characteristics of the industrial Internet, and finally realizes an identification resolution network architecture that carries the massive concurrency of the industrial Internet.

In addition, the identity resolution method based on direct routing and dynamic quantization resolution load according to the above embodiment of the present invention may also have the following additional technical features:

Furthermore, in one embodiment of the present invention, when the number of identification resolutions reaches a set threshold, the polling algorithm is switched to a dynamic quantization resolution load algorithm, wherein the dynamic quantization resolution load algorithm includes:

Construct a server evaluation model with weighted quantitative and qualitative indicators:

S＝S _quantitative +S _qualitative ,

Among them, S is the quantitative index of the server, S _quantitative is the weighted result of the quantitative index, and S _qualitative is the weighted result of the qualitative index;

According to the parameter index in the parsing process, the quantitative index S of each server is obtained, the weight value of each server is initialized, and the polling algorithm of the LSV server is switched to a dynamic quantitative parsing load algorithm.

Furthermore, in one embodiment of the present invention, it also includes:

Periodically obtain the quantitative index S of each server and establish a mathematical model to quantify the server load value;

According to the real-time changes in load values, the weight value of each server is calculated to evaluate the real-time processing capability of the server, and the request volume is allocated according to the real-time processing capability to balance the load among the servers in each cycle.

Furthermore, in one embodiment of the present invention, it also includes:

A global supervision layer is established above the actual network nodes of the identity resolution service to trace and monitor the security of the identity resolution data.

Further, in one embodiment of the present invention, after the LVS server receives the data packet to be parsed, the method further includes:

Check whether the target port number of the data packet is the port number of the specified identity resolution process according to the setting. If so, forward the data packet to be resolved to the real network node of the identity resolution service according to the polling algorithm; otherwise, handle it according to the set exception handling plan.

To achieve the above-mentioned purpose, the second aspect of the embodiment of the present invention proposes an identity resolution device based on direct routing and dynamic quantization of resolution load, characterized in that it includes the following modules: an acquisition module, used to obtain the data packet to be resolved and send it to the LVS server; a transmission module, used to execute the destination network address translation protocol based on the LSV server, and forward the data packet to be resolved to the real network node of the identity resolution service according to the polling algorithm; a resolution module, which processes the data packet to be resolved through the identity resolution server corresponding to the real network node of the identity resolution service, and records the parameter indicators in the resolution process; a return module, which returns the processing result of the data packet to be resolved through the identity resolution server.

Furthermore, in one embodiment of the present invention, a dynamic update module is also included, which is used to:

Construct a server evaluation model with weighted quantitative and qualitative indicators:

S＝S _quantitative +S _qualitative ,

Among them, S is the quantitative index of the server, S _quantitative is the weighted result of the quantitative index, and S _qualitative is the weighted result of the qualitative index;

According to the parameter index in the parsing process, the quantitative index S of each server is obtained, the weight value of each server is initialized, and the polling algorithm of the LSV server is switched to a dynamic quantitative parsing load algorithm.

Furthermore, in one embodiment of the present invention, the dynamic update module is also used to:

Periodically obtain the quantitative index S of each server and establish a mathematical model to quantify the server load value;

According to the real-time changes in load values, the weight value of each server is calculated to evaluate the real-time processing capability of the server, and the request volume is allocated according to the real-time processing capability to balance the load among the servers in each cycle.

To achieve the above-mentioned purpose, the third aspect of the present invention proposes a computer device, characterized in that it includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the identification resolution method based on direct routing and dynamic quantization resolution load as described above is implemented.

To achieve the above-mentioned purpose, the fourth aspect of the present invention proposes a computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, the identification resolution method based on direct routing and dynamic quantization resolution load as described above is implemented.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and/or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

FIG1 is a schematic flow chart of an identity resolution method based on direct routing and dynamic quantization resolution load provided by an embodiment of the present invention.

FIG. 2 is a schematic diagram of a flow chart of an identity resolution device based on direct routing and dynamic quantization resolution load provided by an embodiment of the present invention.

FIG. 3 is a diagram of an identity resolution network architecture provided by an embodiment of the present invention.

FIG4 is a schematic diagram of an identity resolution network for implementing load balancing based on a DR model provided in an embodiment of the present invention.

FIG5 is a flowchart of identity resolution based on a direct routing model and a dynamic quantization resolution load algorithm provided by an embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

The following describes an identity resolution method and device based on direct routing and dynamic quantization resolution load according to an embodiment of the present invention with reference to the accompanying drawings.

FIG1 is a schematic flow chart of an identity resolution method based on direct routing and dynamic quantization resolution load provided by an embodiment of the present invention.

As shown in FIG1 , the identity resolution method based on direct routing and dynamic quantization resolution load includes the following steps:

S1: Get the data packet to be parsed and send it to the LVS server;

S2: Execute the destination network address translation protocol based on the LSV server, and forward the data packet to be resolved to the real network node of the identity resolution service according to the polling algorithm;

S3: The identity resolution server corresponding to the real network node of the identity resolution service processes the data packet to be resolved and records the parameter indicators in the resolution process;

S4: The processing result of the data packet to be resolved is returned to the identity resolution server.

Based on the evolution of the industrial Internet identity resolution system, this paper designs a hierarchical identity resolution architecture based on the Chord routing protocol, reserves a global domain for national regulatory identity resolution data, and can achieve data security and controllability. In addition, a dynamic quantitative resolution load balancing model based on a direct routing model is designed based on the characteristics of the industrial Internet, and finally realizes an identity resolution network architecture that carries massive concurrency of the industrial Internet.

Furthermore, in one embodiment of the present invention, it also includes:

A global supervision layer is established above the actual network nodes of the identity resolution service to trace and monitor the security of the identity resolution data.

The industrial Internet identity resolution architecture implemented by the present invention is shown in Figure 1. The architecture is based on the point-to-point protocol for local domain networking, and the IP of each node in the network and the port that opens the identity resolution service are hashed to obtain the node number. According to the Chord routing protocol, a DHT network is established, and each industry is networked according to this logic. A node is selected as a boundary node, and this boundary node is connected to the global domain network and connected to a node in the global domain. At the same time, the global domain uses logs to record when the resolution occurs, in order to meet the needs of national data supervision. The specific identity resolution process is shown in Figure 3.

Further, in one embodiment of the present invention, after the LVS server receives the data packet to be parsed, the method further includes:

Check whether the target port number of the data packet is the port number of the specified identity resolution process according to the setting. If so, forward the data packet to be resolved to the real network node of the identity resolution service according to the polling algorithm; otherwise, handle it according to the set exception handling plan.

LVS can achieve communication performance close to that of network cable transmission. Users do not need to perform a three-way handshake connection to connect to the LVS server. After the LVS server receives the data packet sent by the client, it can check whether the target port number of the data packet is the port number of the specified identity resolution process according to the settings. If it is, it runs the built-in load balancing algorithm and forwards it to the cluster that actually handles the identity resolution service according to the rules; otherwise, it directly forwards the data packet. The communication between the client and LVS is always controlled at the kernel level, and no data processing is performed. The real identity resolution service is processed by a server in the identity resolution cluster connected to LVS, including the TCP three-way handshake data packet. This is essentially different from Nginx's reverse proxy-based load balancing. The Nginx server needs to establish a TCP connection with the client. The official maximum concurrency is at the level of 50,000 QPS, but the maximum load of LVS can be infinitely expanded with the server performance. Therefore, this article can cluster the nodes in the identity resolution network as follows, and then use the load balancing server to load. The network structure diagram is shown in Figure 4. All user requests are directly sent to the public IP address of the load balancing server, and then forwarded by the load balancing server to the identity resolution server for processing.

The client request is loaded to a server in the identity resolution server cluster, but there is a problem. Because the destination IP address of the identity resolution request data packet sent by the client is the VIP of the load balancing server, rather than the RIP of the server to which it is loaded, the host will not process data packets whose destination address is not its own. This requires the NAT protocol to perform network address translation. In most cases, the IP addresses used are private network addresses that can only be used in a local area network. These IP addresses are not real IP addresses. Only the IP addresses on the router are real public network addresses. These private addresses are not visible on the Internet. There are two addresses on the router, one is a public network address and the other is a private address. The router connects to the operator ISP through the public network address, and finally connects to the host on the Internet to be accessed. This also shows that to access a host on the Internet, a public network address is required.

LVS is used to forward the load to the identity resolution server. After the identity resolution server processes the identity resolution request, it can return on its own. For the client, the load balancing server is transparent. The client does not know that the IP address it requested actually has nothing except the forwarding function. The real service is implemented in the cluster behind it. At the same time, this model does not need to use the NAT protocol to convert IP addresses, which greatly reduces the pressure on the load balancing server.

Furthermore, in one embodiment of the present invention, when the number of identification resolutions reaches a set threshold, the polling algorithm is switched to a dynamic quantization resolution load algorithm, wherein the dynamic quantization resolution load algorithm includes:

Construct a server evaluation model with weighted quantitative and qualitative indicators:

S＝S _quantitative +S _qualitative ,

Among them, S is the quantitative index of the server, S _quantitative is the weighted result of the quantitative index, and S _qualitative is the weighted result of the qualitative index;

According to the parameter index in the parsing process, the quantitative index S of each server is obtained, the weight value of each server is initialized, and the polling algorithm of the LSV server is switched to a dynamic quantitative parsing load algorithm.

Furthermore, in one embodiment of the present invention, it also includes:

Periodically obtain the quantitative index S of each server and establish a mathematical model to quantify the server load value;

According to the real-time changes in load values, the weight value of each server is calculated to evaluate the real-time processing capability of the server, and the request volume is allocated according to the real-time processing capability to balance the load among the servers in each cycle.

According to SPEC, HPCC and other methods, quantitative evaluations are given for processor performance, server system performance, and high-performance computer performance. These evaluation schemes are mainly aimed at the basic performance parameters of the server or the test of a certain performance alone, but these indicators are far from the evaluation indicators of the performance of the industrial Internet identity resolution server. Therefore, the present invention first sets up a model for the performance evaluation of the identity resolution server. First, some test parameters that reflect the various resources and performance of the server are needed to judge the pros and cons of the server performance. Referring to the server performance evaluation model, two types of parameters are mainly introduced: quantitative indicators and qualitative indicators. Among them, the quantitative indicators include: the concurrency of identity resolution requests, the resolution request response bandwidth, the identity resolution delay and delay jitter, the packet loss rate, the effective utilization rate of the identity resolution server CPU and the average waiting time of the server's I/O read and write operations; the qualitative indicators include: reliability, scalability, and availability. For the first type of indicators, because the accurate value can be obtained when the service is running, it is only necessary to determine the weighting coefficient for linear weighting based on the requirements of different industries for identity resolution services. For the second type of indicators, fuzzy processing can be performed, and the server evaluation model of weighted quantitative indicators and qualitative indicators can be derived based on this:

S＝S _quantitative +S _qualitative , (Formula 1)

Among them, there are positive and negative indicators in quantitative indicators. For example, delay and packet loss rate are negative indicators. Therefore, when processing, negative indicators can be processed as q _i = 1/q _i , and positive indicators can be directly multiplied by the weight coefficient to obtain a further expression:

S _quantitative = _∑wiqi (i in _qps , bandwidth, delay, vibration, loss, usage, wait), (Formula 2)

For qualitative indicators, by designing an evaluation set V = (v ₁ ,v ₂ ,v ₃ ) for each item, when the corresponding indicator of the server is v ₁ , its membership degree is fuzzified as:

At the same time, according to the relative comparison method, the weight coefficients of the three indicators are set as follows:

w ₂ = [0.5, 0.3, 0.2], (Formula 4)

At the same time, assuming that the objective evaluation result of the identity resolution server is P, the quantitative result of the qualitative indicator is:

S _{qualitatively} = w ₂ RP. (Formula 5)

The DQRB (Dynamic Quantitative Resolution Balance) algorithm implemented based on the above dynamic quantitative indicators quantifies the performance of the server based on the key performance indicators of the server, rather than subjectively setting a single weight value; the load is quantified based on the pressure caused by the request on each performance indicator of the server, and the number of connections is no longer used as the standard for load measurement. The specific process of the DQRB algorithm is as follows:

1) First, it is necessary to calculate each parameter according to the quantitative formula (Formula 2). Therefore, it is necessary to first use a polling algorithm to receive and respond to the identity resolution request. After running the calculation standard cycle, the quantitative index S of each server is obtained.

2) Based on the collected indicator S, initialize the weight value of each server and switch the LVS load balancing to the DQRB algorithm.

3) Periodically obtain the load parameters of each server and establish a reasonable mathematical model to quantify the load value of each server; according to the real-time changes of the quantified load, calculate the weight value of each server to evaluate the real-time processing capability of the server, allocate the request volume according to the real-time processing capability, and balance the load between the servers in each cycle.

The operation process of the load balancing algorithm of the entire identity resolution system is shown in Figure 5.

The embodiment of the present invention proposes an identity resolution method based on direct routing and dynamic quantization of resolution load. On the one hand, the identity resolution network architecture is realized by means of the Chord routing protocol. Since the identity resolution architecture currently used does not consider the clustering of identities for industry enterprises and the supervision scheme of identity resolution data, the present invention designs a layered scheme of local domains and global domains, which is convenient for inquiries within the industry and can meet the needs of national network security departments for the supervision of identity resolution data; on the other hand, an identity resolution node load balancing scheme based on Linux virtual server technology is designed. The identity resolution service is located at the seventh application layer in the network model, which is the least efficient in the communication layer. On the first layer, the services at the application layer need to call the transmission control layer for transmission, but the LVS technology can achieve load balancing at the network layer and enhance the overall concurrency capability of the network; on the third aspect, a model for the performance evaluation of the identity resolution server is first designed, which mainly introduces two types of parameters: quantitative indicators and qualitative indicators; for the first type of indicators, because the accurate value can be obtained through the service runtime, it is only necessary to determine the weighting coefficient for linear weighting according to the requirements of the identity resolution service in different industries, while for the second type of indicators, fuzzy processing can be used, based on which a server evaluation model of weighted quantitative indicators and qualitative indicators can be derived; on the fourth aspect, the DQRB (Dynamic Quantitative Resolution Balance) algorithm implemented according to the dynamic quantitative indicators quantifies the performance of the server according to the key performance indicators of the server, rather than subjectively setting a single weight value, and quantifies the load by the pressure caused by the request on the various performance indicators of the server, and no longer uses the number of connections as the standard for load measurement, and finally realizes the identity resolution mechanism based on the direct routing model and the dynamic quantitative resolution load algorithm.

In order to implement the above embodiment, the present invention also proposes an identity resolution device based on direct routing and dynamic quantization resolution load.

FIG2 is a schematic diagram of the structure of an identity resolution device based on direct routing and dynamic quantization resolution load provided by an embodiment of the present invention.

As shown in Figure 2, the identity resolution device based on direct routing and dynamic quantification resolution load includes: an acquisition module 10, a transmission module 20, a resolution module 30, and a return module 40, wherein the acquisition module is used to obtain the data packet to be resolved and send it to the LVS server; the transmission module is used to execute the destination network address translation protocol based on the LSV server, and forward the data packet to be resolved to the real network node of the identity resolution service according to the polling algorithm; the resolution module processes the data packet to be resolved through the identity resolution server corresponding to the real network node of the identity resolution service, and records the parameter indicators in the resolution process; the return module returns the processing result of the data packet to be resolved through the identity resolution server.

Furthermore, in one embodiment of the present invention, a dynamic update module is also included, which is used to:

Construct a server evaluation model with weighted quantitative and qualitative indicators:

S＝S _quantitative +S _qualitative ,

Among them, S is the quantitative index of the server, S _quantitative is the weighted result of the quantitative index, and S _qualitative is the weighted result of the qualitative index;

According to the parameter index in the parsing process, the quantitative index S of each server is obtained, the weight value of each server is initialized, and the polling algorithm of the LSV server is switched to a dynamic quantitative parsing load algorithm.

Furthermore, in one embodiment of the present invention, the dynamic update module is also used to:

Periodically obtain the quantitative index S of each server and establish a mathematical model to quantify the server load value;

According to the real-time changes in load values, the weight value of each server is calculated to evaluate the real-time processing capability of the server, and the request volume is allocated according to the real-time processing capability to balance the load among the servers in each cycle.

To achieve the above-mentioned purpose, the third aspect of the present invention proposes a computer device, characterized in that it includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the identification resolution method based on direct routing and dynamic quantization resolution load as described above is implemented.

To achieve the above-mentioned purpose, the fourth aspect of the present invention proposes a computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, the identification resolution method based on direct routing and dynamic quantization resolution load as described above is implemented.

In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims (6)

1. A method for identity resolution based on direct routing and dynamic quantization of resolution load, characterized by comprising the following steps: Get the data packet to be parsed and send it to the LVS server; Based on the LVS server executing the destination network address translation protocol, forwarding the data packet to be resolved to the real network node of the identity resolution service according to the polling algorithm; Processing the data packet to be resolved by the identity resolution server corresponding to the real network node of the identity resolution service, and recording parameter indicators in the resolution process; Return the processing result of the data packet to be resolved through the identity resolution server; When the number of identification resolution times reaches a set threshold, the polling algorithm is switched to a dynamic quantization resolution load algorithm, wherein the dynamic quantization resolution load algorithm includes: Construct a server evaluation model with weighted quantitative and qualitative indicators: S＝S _quantitative +S _qualitative , Among them, S is the quantitative index of the server, S _quantitative is the weighted result of the quantitative index, and S _qualitative is the weighted result of the qualitative index; among them, the negative index is processed as q _i =1/q _i , and the positive index is directly multiplied by the weight coefficient, and the further expression can be obtained: S _quantitative = _Σwiqi (i in _qps , bandwidth, delay, vibration, loss, usage, wait), For qualitative indicators, by designing an evaluation set V = (v ₁ ,v ₂ ,v ₃ ) for each item, when the corresponding indicator of the server is v ₁ , its membership degree is fuzzified as: At the same time, according to the relative comparison method, the weight coefficients of the three indicators are set as follows: w ₂ = [0.5, 0.3, 0.2], At the same time, assuming that the objective evaluation result of the identity resolution server is P, the quantitative result of the qualitative indicator is: S _qualitative = w ₂ RP; According to the parameter index in the analysis process, the quantitative index S of each server is obtained, the weight value of each server is initialized, and the polling algorithm of the LVS server is switched to a dynamic quantitative analysis load algorithm; Periodically obtain the quantitative index S of each server and establish a mathematical model to quantify the server load value; According to the real-time changes of the load value, the weight value of each server is calculated to evaluate the real-time processing capability of the server, the request amount is allocated according to the real-time processing capability, and the load between the servers is balanced in each cycle. 2. The method according to claim 1, further comprising: A global supervision layer is established above the real network nodes of the identity resolution service to trace and monitor the security of the identity resolution data. 3. The method according to claim 1, characterized in that after the LVS server receives the data packet to be parsed, it also includes: Check whether the target port number of the data packet is the port number of the specified identity resolution process according to the setting. If so, forward the data packet to be resolved to the real network node of the identity resolution service according to the polling algorithm; otherwise, handle it according to the set exception handling plan. 4. An identity resolution device based on direct routing and dynamic quantization resolution load, characterized by comprising the following modules: The acquisition module is used to obtain the data packet to be parsed and send it to the LVS server; A transmission module, configured to execute a destination network address translation protocol based on the LVS server and forward the data packet to be resolved to a real network node of the identity resolution service according to a polling algorithm; A parsing module, which processes the data packet to be parsed through the identity resolution server corresponding to the real network node of the identity resolution service, and records parameter indicators during the parsing process; A return module returns the processing result of the data packet to be parsed through the identity resolution server; Also included is a dynamic update module for: Construct a server evaluation model with weighted quantitative and qualitative indicators: S＝S _quantitative +S _qualitative , Among them, S is the quantitative index of the server, S _quantitative is the weighted result of the quantitative index, and S _qualitative is the weighted result of the qualitative index; The quantitative index S of each server is obtained according to the parameter index in the analysis process, the weight value of each server is initialized, and the polling algorithm of the LVS server is switched to a dynamic quantitative analysis load algorithm; wherein, the negative index is processed as q _i =1/q _i , and the positive index is directly multiplied by the weight coefficient, and a further expression can be obtained: S _quantitative = _Σwiqi (i in _qps , bandwidth, delay, vibration, loss, usage, wait), For qualitative indicators, by designing an evaluation set V = (v ₁ ,v ₂ ,v ₃ ) for each item, when the corresponding indicator of the server is v ₁ , its membership degree is fuzzified as: At the same time, according to the relative comparison method, the weight coefficients of the three indicators are set as follows: w ₂ = [0.5, 0.3, 0.2], At the same time, assuming that the objective evaluation result of the identity resolution server is P, the quantitative result of the qualitative indicator is: S _qualitative = w ₂ RP; According to the parameter index in the analysis process, the quantitative index S of each server is obtained, the weight value of each server is initialized, and the polling algorithm of the LVS server is switched to a dynamic quantitative analysis load algorithm; Periodically obtain the quantitative index S of each server and establish a mathematical model to quantify the server load value; According to the real-time changes of the load value, the weight value of each server is calculated to evaluate the real-time processing capability of the server, and the request amount is allocated according to the real-time processing capability, so as to balance the load among the servers in each cycle. 5. A computer device, characterized in that it comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method according to any one of claims 1 to 3 is implemented. 6. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.