WO2020211365A1

WO2020211365A1 - Flow load balancing method and apparatus, computer device and storage medium

Info

Publication number: WO2020211365A1
Application number: PCT/CN2019/118386
Authority: WO
Inventors: 宋杰
Original assignee: 平安科技（深圳）有限公司
Priority date: 2019-04-19
Filing date: 2019-11-14
Publication date: 2020-10-22
Also published as: CN110011928A; CN110011928B

Abstract

Provided by the present application are a flow load balancing method and apparatus based on data processing, a computer device and a storage medium. The method comprises: obtaining a network request initiated by a user through a target system, and sending a call instruction; calling a preset strategy set according to the call instruction, the strategy set comprising a plurality of load strategies; matching a plurality of load strategies in the strategy set with the network request one by one according to a preset matching rule to find a target load strategy corresponding to the network request; obtaining a target service address corresponding to the network request from the target load strategy; and forwarding the network request to a server corresponding to the target service address. The method enables different network requests to be forwarded to appropriate servers for processing, improves the stability of the system, and realizes direct and immediately effective on-line modification of the load strategy as needed.

Description

Flow balance load method, device, computer equipment and storage medium

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on April 19, 2019 with the application number 2019103190689 and the application titled "Flow Balanced Load Method, Device, Computer Equipment and Storage Medium", the entire content of which is by reference Incorporated in this application.

Technical field

This application relates to the technical field of data processing, and in particular to a method, device, computer equipment, and storage medium for traffic balancing load.

Background technique

With the development of business, the usage of various business systems by users has gradually increased. When network requests continue to increase and the load on the system increases, the response delay will also increase, and even data congestion and traffic will occur. Problems such as imbalance, so system stability becomes more important.

The traditional method uses load balancing equipment to load traffic to multiple servers to achieve simple system high availability, but the stability of the system is not high. Once a failure occurs, it will affect all users. In addition, the existing load In the method of balancing the load by strategy, when the load strategy needs to be modified, the process is more complicated and the real-time performance is also poor.

technical problem

The main purpose of this application is to provide a method, device, computer equipment, and storage medium for traffic balancing load, aiming to solve the technical problem of poor system stability when load balancing traffic.

Technical solutions

Based on the above-mentioned purpose of the invention, this application proposes a load balancing method for traffic, including:

Get the network request initiated by the user through the target system;

Generating a call instruction according to the network request;

Calling a preset strategy set according to the calling instruction, the strategy set including a plurality of load strategies;

Matching a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

Obtaining the target service address corresponding to the network request from the target load strategy;

Forward the network request to the server corresponding to the target service address.

This application also provides a flow balancing load device, including:

The acquisition request unit is used to acquire the network request initiated by the user through the target system;

Generating and calling unit for generating a calling instruction according to the network request;

The invoking strategy unit is configured to invoke a preset strategy set according to the invoking instruction, and the strategy set includes multiple load strategies;

A matching strategy unit, configured to match a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

An obtaining address unit, configured to obtain a target service address corresponding to the network request from the target load strategy;

The forwarding request unit is configured to forward the network request to the server corresponding to the target service address.

The present application also provides a computer device, including a memory and a processor, the memory stores computer-readable instructions, and the processor implements the steps of the foregoing method when the computer-readable instructions are executed.

The present application also provides a computer-readable storage medium on which computer-readable instructions are stored, and when the computer-readable instructions are executed by a processor, the steps of the foregoing method are implemented.

Beneficial effect

The beneficial effects of this application are: forwarding the network request to the load component and forwarding it to the corresponding server through the load strategy, so that different network requests can be forwarded to a suitable server for processing, improving the stability of the system, and when the load needs to be modified When making a strategy, directly modify the strategy online and take effect in real time.

Description of the drawings

FIG. 1 is a schematic diagram of steps of a method for balancing load in an embodiment of this application;

2 is a schematic block diagram of the structure of a traffic balancing load device in an embodiment of the application;

FIG. 3 is a schematic block diagram of the structure of a computer device according to an embodiment of the application.

The realization, functional characteristics, and advantages of the purpose of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

The best mode of the invention

It should be understood that the specific embodiments described here are only used to explain the application, and are not used to limit the application.

Referring to Figure 1, the method for traffic balancing load in this embodiment includes:

Step S1: Obtain the network request initiated by the user through the target system;

Step S2: Generate a call instruction according to the network request;

Step S3: Invoking a preset strategy set according to the invoking instruction, the strategy set including multiple load strategies;

Step S4: Match a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

Step S5: Obtain the target service address corresponding to the network request from the target load strategy;

Step S6: Forward the network request to the server corresponding to the target service address.

In this embodiment, the above method is implemented in a network environment. In order to prevent users from interfering with each other when the network fails, and the system work will not be interrupted, a set (such as allocating servers, deploying applications) is set up in different areas. New network environment, and ensure that the new network environment and the infrastructure used in the old network environment (such as host, network, storage, etc.) are physically separated. For example, the new computer room in Shenzhen is physically separated from the old computer room in Shanghai, but the two share the same system. When obstacles occur, the other can be used, and the two sets can be used at the same time to further share the load pressure.

As mentioned in step S1 and step S2 above, it is known that an enterprise may have multiple business systems, and each business system has a different function, and the request of each business system must be allocated to a suitable server, so that the system Load balancing. The above-mentioned target system is any business system in this embodiment. When a user initiates a network request through the target system, and the load system receives the network request, it forwards the network request to the load component in the load system. It is a program used to temporarily store network requests and forward them to a suitable server. When the network request is forwarded to the load component, the load component starts to work. At this time, the load component triggers the generation of a call instruction according to the network request, and the call instruction is used to call a preset strategy set.

As described in the above steps S3 and S4, the load component invokes the strategy set according to the call instruction, such as obtained from the cache or obtained from a preset database. The above strategy set contains multiple load strategies, and these load strategies are used to determine whether the network request Which server to forward to for processing, because different business systems have different load strategies corresponding to the configuration, it is necessary to find the target load strategy corresponding to the network request. In this embodiment, the load strategy and network request are preset by The matching rules are matched to find the target load strategy corresponding to the network request in the strategy set. The above matching rule may be that the two are considered to match when the network request meets the preset condition in the load strategy. Among them, the load strategy includes policy name, keyword key, request attribute type (any one of request header hd, parameter pm, source org), preset condition cn, routing array, strategy type, etc. The above routing arrays are multiple The collection of service addresses, that is, the routing array contains multiple service addresses, and each service address corresponds to a group of processing servers. These load strategies can be configured in advance through the developer using the command line, or through the user using the online editing page to configure and configure Then store it in the database. When the load strategy needs to be modified, get the load strategy from the database and modify it, and then save the modified load strategy in the database.

As described in the above steps S5 and S6, after finding the corresponding target load strategy, the routing array in the target load strategy can be obtained. Since there are multiple service addresses in the routing array, it is necessary to select the target corresponding to the network request Service address. It is known that the network request includes request line, request header, and request body. The request header includes multiple attributes in the format of "attribute name: attribute value". These attributes include the customer information of the business system that initiated the request (ID) or corresponding client, these can respectively correspond to different target service addresses. For different business systems, the method of selecting the target service address is different. For example, it is selected by preset rules or preset algorithms. When the target service address is found, the network request can be forwarded to the server corresponding to the target service address. The server processes the aforementioned network requests. Through this method, the network requests of different business systems are allocated to suitable servers for processing, and the balance is reasonable. On the one hand, the system can run more quickly and stably, and on the other hand, the load strategy can be modified in real time, which is simple and convenient for immediate use.

In one embodiment, in order to ensure the high availability of load components, multiple load components are deployed in a cluster manner, that is, a load cluster composed of multiple load components is set up. After the network request is obtained, the service discovery program can be used randomly It is forwarded to any load component in the load cluster. The aforementioned service discovery program is a preset program for randomly selecting load components from the load cluster. After the service discovery program selects one of the load components in the load cluster, it forwards the above network request to the load component for further distribution and forwarding.

In another example, it can also be distributed according to the current flow (workload) of the load components. First, obtain the current flow of each load component and compare it to find the load component with the smallest flow, and then forward the network request to this Load components.

In an embodiment, the above step S3 includes:

Step S31: Load the policy set in the preset database into the local memory according to the call instruction to call the policy set.

In this embodiment, in order to improve the operating speed of the system, the policy set can be cached from the database to redis (Redis is an open source Key-Value storage system written in ANSIC language). It is known that the above policy set is stored in a preset database When the load component is started, the above strategy set is loaded into the local memory according to the calling instruction, so that it can be called during subsequent work, which improves work efficiency.

In one embodiment, the strategy set includes a private load strategy corresponding to a designated business system and a public load strategy corresponding to all business systems. The above step S4 includes:

Step S41: Match the private load policy with the network request;

Step S42: If the target load strategy cannot be matched, the public load strategy is matched with the network request to find the target load strategy.

In this embodiment, the above-mentioned load strategy includes two types of strategies, namely, a private load strategy and a public load strategy. The private load strategy corresponds to a designated business system, that is, it is configured for a certain business system, such as a verification code system. Only the load strategy corresponding to the verification code system can be used; the public load strategy corresponds to most business systems. When matching, first match the private load strategy. If the target load strategy is found in the private load strategy, it indicates that the business system has set a targeted load strategy, then the load strategy can be used first, that is, the load strategy will The network request is forwarded to the corresponding server. If the private load policy cannot be matched, the public policy is matched with the network request to find the target load policy, and then the network request is forwarded to the corresponding server through the load policy.

In one embodiment, the load strategy includes the strategy name, keywords, request attributes, preset conditions, routing array, and strategy type. The above step S4 includes:

Step S43: Obtain the attribute value corresponding to the load strategy from the network request according to the request attribute of each load strategy;

Step S44: sequentially determine whether the attribute value meets the preset condition of the corresponding load strategy, until the target load strategy whose attribute value meets the preset condition is found.

In this embodiment, the above-mentioned load policy includes policy name, keyword key, request attribute type (any of request header hd, parameter pm, source org), preset condition (cn condition), routing array, policy type (whether Global st(0\1), global is public, non-global is private), etc. The request attributes and preset conditions of each load strategy configuration may be different. When configuring, start with the request header hd and parameters of the request attributes Choose any configuration from pm and source org, and then obtain the attribute value of the requested attribute corresponding to the network request according to the request attribute, determine whether the attribute value meets the preset conditions of the load strategy, and judge each load strategy in turn according to the above method. Until the target load strategy that meets the preset conditions is found.

For example, the request attribute of the first load strategy is configured as the parameter pm. When matching, the parameter pm value (that is, the attribute value) of the network request is obtained, and the parameter pm value is calculated according to the preset calculation in the preset conditions To check whether the parameter pm value meets the preset condition. For example, if the parameter pm value is equal to 0 for 1000, the network request satisfies the preset condition. At this time, the first load strategy defaults to Target load strategy; if the value of the parameter pm is not equal to 0 for 1000, that is, the preset condition is not met, the second load strategy in the strategy set needs to be judged again, and the request attribute of the second load strategy is configured as request Head hd, then get the hd value (attribute value) of the request header of the network request. The preset condition of the second load strategy is that if the hd value is equal to 0 after the remainder of 10, the network request meets the preset condition. The second load strategy is the target load strategy by default, otherwise continue to judge the third load strategy until the target load strategy is found.

For example, a load strategy can be configured as follows: the name p1, the keyword ip, the request attribute is the parameter pm, the preset condition is that the value of the parameter is equal to 0 for 1000, and the routing array includes two service addresses, 100.78.98.111 and 100.78.98.111 respectively. 100.78.98.11, the policy type is a private load policy, which is expressed as follows: {"name":"p1","key":"ip","type":"pm","cn":"$var%1000= ＝0","tar":"100.78.98.111|||100.78.98.112||||4344|||433","st":"1"}

In an embodiment, the above step S5 includes:

Step S51: Obtain the mobile phone number of the user and the system ID of the target system;

Step S52: calculating the mobile phone number and the system ID through a hash function to obtain a first value;

Step S53: Obtain the number of all service addresses in the routing array;

Step S54: Calculate the first value and the number through a preset formula to obtain the target service address.

In this embodiment, it is known that the target load strategy also includes a routing array. The process of selecting the target service address from the routing array can be as follows: first obtain the user’s mobile phone number and the system ID of the target system, since the network request includes For user information and target system information, the corresponding system ID can be obtained according to the network request, and the corresponding user can be found according to the network request to obtain the mobile phone number. Then, the mobile phone number and the system ID are calculated through a hash function to obtain the first value. The above hash function is a hash function, also known as a "hash" function. In this embodiment, the expression of the hash function is n=hash(phone+id), the mobile phone number and the system ID are spliced through the hash function operation to obtain the first value n between 0 and 31, and then the route of the target load strategy is obtained The number of all service addresses in the array.

In the above step S54, the preset calculation formula may be: add=g[n%l], where add is the target service address, n is the first value, and l is the number of service addresses in the routing array, g is the routing array, and g[n%l] is the n%l service address in the routing array. Specifically, first obtain the remainder m of the first value divided by the number, that is, m=n% l, and g[m] is the m-th service address in the routing array. For example, when m is found to be 5, the above-mentioned target service address is the fifth service address in the routing array.

In an embodiment, the above step S5 includes:

Step S50: Select the target service address from the routing array of the target load strategy through a random algorithm.

Since the routing array contains multiple service addresses, each service address corresponds to a group of processing servers. When the network request is to be forwarded to the server, the service address of the server needs to be selected from multiple service addresses. In this embodiment, The selection is made by a random algorithm, and one of multiple service addresses is randomly selected. The above random algorithm can be any of the known numerical probability algorithm, Monte Carlo algorithm, Sherwood algorithm, and Sherwood algorithm. The calculation process is In the prior art, which will not be repeated here, a service address in the routing array is calculated through a random algorithm, and the service address is the aforementioned target service address.

Referring to Figure 2, the flow balancing load device in this embodiment includes:

The obtaining request unit 100 is configured to obtain a network request initiated by a user through a target system;

The generating and calling unit 200 is configured to generate a calling instruction according to the network request;

The invoking strategy unit 300 is configured to invoke a preset strategy set according to the invoking instruction, and the strategy set includes multiple load strategies;

The matching strategy unit 400 is configured to match a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

An address obtaining unit 500, configured to obtain a target service address corresponding to the network request from the target load policy;

The forwarding request unit 600 is configured to forward the network request to the server corresponding to the target service address.

In this embodiment, the above-mentioned device is applied in a network environment. In order to prevent interference between users when the network fails, and the system work will not be interrupted, a set of devices (such as allocating servers, deploying applications) is built in different areas. New network environment, and ensure that the new network environment and the infrastructure used in the old network environment (such as host, network, storage, etc.) are physically separated. For example, the new computer room in Shenzhen is physically separated from the old computer room in Shanghai, but the two share the same system. When obstacles occur, the other can be used, and the two sets can be used at the same time to further share the load pressure.

As described in the above-mentioned acquisition request unit 100 and the generation and invocation unit 200, it is known that an enterprise may have multiple business systems, and each business system has a different function, and the request for processing each business system must be allocated to a suitable The server makes the system load balanced. The above-mentioned target system is any business system in this embodiment. When a user initiates a network request through the target system, after the acquisition request unit 100 in the load system obtains the network request, it forwards the network request to the load system. Load component, the above load component is a program for temporarily storing network requests, and distributing and forwarding network requests to a suitable server. When the network request is forwarded to the load component, the load component starts to work. At this time, the load component triggers the generation of a call instruction according to the network request, and the call instruction is used to call a preset strategy set.

As described in the invoking strategy unit 300 and the matching strategy unit 400, the invoking strategy unit 300 of the load component invokes the strategy set according to the invocation instruction, such as obtained from a cache or a preset database, the strategy set includes multiple load strategies, These load strategies are used to determine which server to forward the network request distribution to for processing. Since different business systems have different load strategies corresponding to the configuration, it is necessary to find the target load strategy corresponding to the network request. In this embodiment, match The policy unit 400 matches the load policy with the network request according to a preset matching rule, thereby finding a target load policy corresponding to the network request in the policy set. The matching rule may satisfy the preset condition in the load policy for the network request, namely It can be considered that the two match. Among them, the load strategy includes policy name, keyword key, request attribute type (any one of request header hd, parameter pm, source org), preset condition cn, routing array, strategy type, etc. The above routing arrays are multiple The collection of service addresses, that is, the routing array contains multiple service addresses, and each service address corresponds to a group of processing servers. These load strategies can be configured in advance through the developer using the command line, or through the user using the online editing page to configure and configure Then store it in the database, and when the load strategy needs to be modified, get the load strategy from the database and modify it, and then save the modified load strategy in the database.

As described in the address unit 500 and the forwarding request unit 600 described above, after finding the corresponding target load strategy, the routing array in the target load strategy can be obtained. Since there are multiple service addresses in the routing array, it is necessary to select a network The target service address corresponding to the request. It is known that the network request includes the request line, the request header, and the request body. The request header includes multiple attributes in the format of "attribute name: attribute value". These attributes include the business that initiated the request The customer information (ID) of the system or the corresponding client, which can respectively correspond to different target service addresses. For different business systems, the method of selecting the target service address is different. For example, the selection is made by preset rules or preset algorithms. When the target service address is found, the forwarding request unit 600 can forward the network request to the corresponding target service address. The server, which processes the above network request. By distributing the network requests of different business systems to suitable servers for processing, balanced and reasonable, on the one hand, the system can run faster and more stable, on the other hand, the load strategy can be modified in real time, which is simple and convenient for immediate use.

In one embodiment, in order to ensure the high availability of load components, multiple load components are deployed in a cluster manner, that is, a load cluster composed of multiple load components is set up. After the network request is obtained, the service discovery program can be used randomly It is forwarded to any load component in the load cluster. The above service discovery program is a preset program for randomly selecting load components from the load cluster. After the service discovery program selects one of the load components in the load cluster, it forwards the above network request to the load component for further distribution and forwarding.

In an embodiment, the above-mentioned calling strategy unit 300 includes:

The loading strategy subunit is configured to load the strategy set in the preset database to the local memory according to the calling instruction to invoke the strategy set.

In this embodiment, in order to improve the operating speed of the system, the policy set can be cached from the database to redis (Redis is an open source Key-Value storage system written in ANSIC language). It is known that the above policy set is stored in a preset database When the load component is started, the above strategy set is loaded into the local memory according to the calling instruction, so that it can be called during subsequent work, and work efficiency is improved.

In one embodiment, the strategy set includes a private load strategy corresponding to a designated business system and a public load strategy corresponding to all business systems. The above-mentioned matching strategy unit 400 includes:

The first matching subunit is used to match the private load policy with the network request;

The second matching subunit is configured to match the public load strategy with the network request when the target load strategy cannot be matched to find the target load strategy.

In one embodiment, the load strategy includes a strategy name, keywords, request attributes, preset conditions, routing arrays, and strategy types. The above matching strategy unit 400 includes:

The acquiring attribute subunit is configured to acquire the attribute value corresponding to the load strategy from the network request according to the request attribute of each load strategy;

The judgment condition subunit is used to sequentially judge whether the attribute value meets the corresponding preset condition of the load strategy, until the target load strategy whose attribute value meets the preset condition is found.

In an embodiment, obtaining the address unit 500 above includes:

The number obtaining subunit is used to obtain the mobile phone number of the user and the system ID of the target system;

An arithmetic value subunit for calculating the mobile phone number and system ID through a hash function to obtain the first value;

The quantity obtaining subunit is used to obtain the quantity of all service addresses in the routing array;

The address calculation subunit is configured to calculate the first value and the number through a preset calculation formula to obtain the target service address.

In the above calculation address subunit, the preset calculation formula may be: add=g[n%l], where add is the target service address, n is the first value, and l is the value of the service address in the routing array Number, g is the routing array, g[n%l] is the n%l service address in the routing array. Specifically, first obtain the remainder m of the first value divided by the number, that is, m= n%l, and g[m] is the m-th service address in the routing array. For example, when m is found to be 5, the above-mentioned target service address is the fifth service address in the routing array.

In an embodiment, obtaining the address unit 500 above includes:

The address selection subunit is used to select the target service address from the routing array of the target load strategy through a random algorithm.

3, an embodiment of the present application also provides a computer device. The computer device may be a server, and its internal structure may be as shown in FIG. 3. The computer equipment includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the computer designed processor is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer readable instructions, and a database. The memory provides an environment for the operation of the operating system and computer readable instructions in the non-volatile storage medium. The database of the computer equipment is used to store all the data required for the flow load process. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer-readable instruction is executed by the processor to realize a method of traffic balancing load.

The above-mentioned processor executes the steps of the above-mentioned traffic balancing load method: acquiring a network request initiated by a user through a target system; generating a calling instruction according to the network request; calling a preset policy set according to the calling instruction, and the policy set includes a plurality of Load strategy; a plurality of load strategies in the strategy group are matched with the network request according to a preset matching rule to find the target load strategy corresponding to the network request; and the target load strategy is obtained from the target load strategy. The target service address corresponding to the network request; forwarding the network request to the server corresponding to the target service address.

In the above computer device, the step of invoking a preset strategy set according to the invoking instruction includes: loading the strategy set in the preset database to a local memory according to the invoking instruction to invoke the strategy set.

In one embodiment, the above-mentioned strategy set includes a private load strategy corresponding to a designated business system and a public load strategy corresponding to all business systems, and the plurality of load strategies in the strategy set are matched one by one with the network request according to a preset The step of matching rules to find the target load strategy corresponding to the network request includes: matching the private load strategy with the network request; if the target load strategy cannot be matched, then the public load The strategy is matched with the network request to find the target load strategy.

In one embodiment, the aforementioned load strategy includes request attributes and preset conditions, and the multiple load strategies in the strategy set are matched one by one with the network request according to a preset matching rule to find the corresponding network request The step of the target load strategy includes: obtaining the attribute value corresponding to the load strategy from the network request according to the request attribute of each load strategy; sequentially determining whether the attribute value meets the corresponding load strategy The preset condition is reached until the target load strategy whose attribute value meets the preset condition is found.

In an embodiment, the above-mentioned target load strategy further includes a routing array, the routing array includes a plurality of service addresses, and the step of obtaining a target service address corresponding to the network request from the target load strategy , Including: obtaining the mobile phone number of the user and the system ID of the target system; calculating the mobile phone number and the system ID through a hash function to obtain the first value; obtaining the number of all service addresses in the routing array; The first value and the number are calculated by a preset formula to obtain the target service address.

In one embodiment, the step of calculating the first value and the number to obtain the target service address through a preset formula includes: calculating the target service address using the following preset formula: add=g[n %L], where add is the target service address, n is the first value, l is the number of service addresses in the routing array, g is the routing array, and g[n%l] is the value The n%lth service address in the routing array.

In an embodiment, the above step of obtaining the target service address corresponding to the network request from the target load strategy includes: selecting the target service address from the routing array of the target load strategy through a random algorithm .

Those skilled in the art can understand that the structure shown in FIG. 3 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied.

An embodiment of the present application further provides a computer-readable storage medium. The computer-readable storage medium is, for example, a non-volatile computer-readable storage medium or a volatile computer-readable storage medium on which Stored with computer-readable instructions, when the computer-readable instructions are executed by the processor, a method for traffic balancing load is realized, specifically: obtaining a network request initiated by a user through the target system; generating a call instruction according to the network request; The instruction calls a preset strategy set, the strategy set includes a plurality of load strategies; the plurality of load strategies in the strategy set are matched with the network request according to the preset matching rule to find the corresponding network request Target load strategy; obtain the target service address corresponding to the network request from the target load strategy; forward the network request to the server corresponding to the target service address.

In the computer-readable storage medium, the step of invoking a preset strategy set according to the invoking instruction includes: loading the strategy set in a preset database to a local memory according to the invoking instruction to invoke the strategy set .

The above are only the preferred embodiments of this application, and do not limit the scope of this application. Any equivalent structure or equivalent process transformation made using the content of this application description and drawings, or directly or indirectly applied to other related The technical field is equally included in the scope of patent protection of this application.

Claims

A load balancing method for traffic, characterized in that it comprises:

Get the network request initiated by the user through the target system;

Generating a call instruction according to the network request;

Calling a preset strategy set according to the calling instruction, the strategy set including a plurality of load strategies;

Matching a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

Obtaining the target service address corresponding to the network request from the target load strategy;

Forward the network request to the server corresponding to the target service address.
The traffic balancing load method according to claim 1, wherein the step of invoking a preset policy set according to the invoking instruction comprises:

The strategy set in the preset database is loaded into the local memory according to the calling instruction to invoke the strategy set.
The traffic balancing load method according to claim 1, wherein the strategy set includes a private load strategy corresponding to a designated business system and a public load strategy corresponding to all business systems, and the strategy is grouped into multiple load strategies The step of matching the network request according to the preset matching rule to find the target load strategy corresponding to the network request includes:

Matching the private load policy with the network request;

If the target load strategy cannot be matched, the public load strategy is matched with the network request to find the target load strategy.
The method for traffic balancing load according to claim 1, wherein the load strategy includes request attributes and preset conditions, and the plurality of load strategies in the strategy group are matched with the network request according to the preset conditions. The step of matching the rules to find the target load strategy corresponding to the network request includes:

Obtaining the attribute value corresponding to the load strategy from the network request according to the request attribute of each load strategy;

It is determined in turn whether the attribute value meets the corresponding preset condition of the load strategy, until the target load strategy with the attribute value meeting the preset condition is found.
The method of traffic balancing load according to claim 1, wherein the target load strategy further includes a routing array, the routing array includes a plurality of service addresses, and the data obtained from the target load strategy and The step of the network requesting the corresponding target service address includes:

Acquiring the mobile phone number of the user and the system ID of the target system;

Calculating the mobile phone number and the system ID through a hash function to obtain the first value;

Obtaining the number of all service addresses in the routing array;

The target service address is obtained by calculating the first value and the number through a preset calculation formula.
The method of traffic balancing load according to claim 5, wherein the step of calculating the first value and the number to obtain the target service address through a preset calculation formula comprises:

The target service address is calculated using the following preset formula:

add=g[n%l]

Where add is the target service address, n is the first value, l is the number of service addresses in the routing array, g is the routing array, and g[n%l] is the number of The n%l service address.
The traffic balancing load method according to claim 1, wherein the step of obtaining a target service address corresponding to the network request from the target load strategy comprises:

The target service address is selected from the routing array of the target load strategy through a random algorithm.
A flow balancing load device, characterized in that it comprises:

Obtaining request unit for obtaining network requests initiated by users through the target system

Generating and calling unit for generating a calling instruction according to the network request;

The invoking strategy unit is configured to invoke a preset strategy set according to the invoking instruction, and the strategy set includes multiple load strategies;

A matching strategy unit, configured to match a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

An obtaining address unit, configured to obtain a target service address corresponding to the network request from the target load strategy;

The forwarding request unit is configured to forward the network request to the server corresponding to the target service address.
The traffic balancing load device according to claim 8, wherein the invoking policy unit comprises:

The loading strategy subunit is configured to load the strategy set in the preset database to the local memory according to the calling instruction to invoke the strategy set.
The traffic balancing load device according to claim 8, wherein the matching strategy unit comprises:

The first matching subunit is used to match the private load policy with the network request;

The second matching subunit is configured to match the public load strategy with the network request when the target load strategy cannot be matched to find the target load strategy.
The traffic balancing load device according to claim 8, wherein the matching strategy unit:

The acquiring attribute subunit is configured to acquire the attribute value corresponding to the load strategy from the network request according to the request attribute of each load strategy;

The judgment condition subunit is used to sequentially judge whether the attribute value meets the corresponding preset condition of the load strategy, until the target load strategy whose attribute value meets the preset condition is found.
The traffic balancing load device according to claim 8, wherein the obtaining address unit comprises:

The number obtaining subunit is used to obtain the mobile phone number of the user and the system ID of the target system;

An arithmetic value subunit for calculating the mobile phone number and system ID through a hash function to obtain the first value;

The quantity obtaining subunit is used to obtain the quantity of all service addresses in the routing array;

The address calculation subunit is configured to calculate the first value and the number through a preset calculation formula to obtain the target service address.
A computer device includes a memory and a processor, the memory stores computer-readable instructions, and is characterized in that when the processor executes the computer-readable instructions, a traffic balancing load method is implemented, and the traffic balancing load method includes:

Get the network request initiated by the user through the target system;

Generating a call instruction according to the network request;

Calling a preset strategy set according to the calling instruction, the strategy set including a plurality of load strategies;

Matching a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

Obtaining the target service address corresponding to the network request from the target load strategy;

Forward the network request to the server corresponding to the target service address.
The computer device according to claim 13, wherein the step of invoking a preset strategy set according to the calling instruction comprises:

The strategy set in the preset database is loaded into the local memory according to the calling instruction to invoke the strategy set.
The computer device according to claim 13, wherein the strategy set includes a private load strategy corresponding to a specified business system and a public load strategy corresponding to all business systems, and the strategy is set to a plurality of load strategies one by one. The step of matching the network request according to a preset matching rule to find the target load strategy corresponding to the network request includes:

Matching the private load policy with the network request;

If the target load strategy cannot be matched, the public load strategy is matched with the network request to find the target load strategy.
The computer device according to claim 13, wherein the load strategy includes request attributes and preset conditions, and the grouping of multiple load strategies one by one in the strategy is performed according to a preset matching rule with the network request The step of matching to find the target load strategy corresponding to the network request includes:

Obtaining the attribute value corresponding to the load strategy from the network request according to the request attribute of each load strategy;

It is determined in turn whether the attribute value meets the corresponding preset condition of the load strategy, until the target load strategy with the attribute value meeting the preset condition is found.
A computer-readable storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions are executed by a processor to implement a flow balancing load method, and the flow balancing load method includes:

Get the network request initiated by the user through the target system;

Generating a call instruction according to the network request;

Calling a preset strategy set according to the calling instruction, the strategy set including a plurality of load strategies;

Matching a plurality of load strategies in the strategy set one by one with the network request according to a preset matching rule to find a target load strategy corresponding to the network request;

Obtaining the target service address corresponding to the network request from the target load strategy;

Forward the network request to the server corresponding to the target service address.
18. The computer-readable storage medium of claim 17, wherein the step of invoking a preset strategy set according to the invoking instruction comprises:

The strategy set in the preset database is loaded into the local memory according to the calling instruction to invoke the strategy set.
The computer-readable storage medium according to claim 17, wherein the strategy set includes a private load strategy corresponding to a specified business system and a public load strategy corresponding to all business systems, and the strategy is centralized to multiple loads The step of matching the strategy one by one with the network request according to a preset matching rule to find the target load strategy corresponding to the network request includes:

Matching the private load policy with the network request;

If the target load strategy cannot be matched, the public load strategy is matched with the network request to find the target load strategy.
The computer-readable storage medium according to claim 17, wherein the load strategy includes request attributes and preset conditions, and the plurality of load strategies are combined with the network request according to a preset condition. The step of matching the matching rules to find the target load strategy corresponding to the network request includes:

Obtaining the attribute value corresponding to the load strategy from the network request according to the request attribute of each load strategy;

It is determined in turn whether the attribute value meets the corresponding preset condition of the load strategy, until the target load strategy with the attribute value meeting the preset condition is found.