An intelligent load balancing method and system for gateway based on task data preloading

Abstract

The present invention discloses a method and system for intelligent load balancing of gateway based on task data preloading, which relates to the field of communication technologies. The method includes obtaining gateway node information, obtaining gateway node load information according to the gateway node information, obtaining a gateway node load index according to the gateway node load information and gateway node parameter information, and obtaining a node task preloading coefficient according to the task data preloading node information and the gateway node load index. The present invention accurately evaluates the load status of gateway nodes through the gateway node load index, obtains a node data transmission index according to the task data information and gateway node information, obtains task data preloading node information according to the node data transmission index, improves the task data preloading efficiency, evaluates the task data preloading status through the node task preloading coefficient, realizes the load balancing of gateway nodes, improves the system performance, and enhances the stability and adaptability of the system.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and particularly to an intelligent gateway load balancing method and system based on task data preloading. Background Art

[0002] With the development of the Internet, the increase in business volume, the separation of front and back ends, the increase in the number of developers, the consistency of development, testing, and production environments, as well as more refined operations such as CDN, traffic control, and permission control, and the current prevalence of the microservices architecture, a large number of gateway designs have emerged. Under the current prevalence of cloud computing technologies, the deployment of system service clusters has become a norm. When a service has multiple node deployments, issues such as request forwarding and distribution need to be considered, and requests should be distributed as evenly as possible to each node to balance the request load of each service. Currently, there are already many devices for load balancing, such as LVS, Nginx, etc., which support request forwarding strategies such as IP_HASH, round-robin, and weight. The purpose of these strategies is to distribute requests as evenly as possible to each node and ensure that the load of each node remains within a reasonable range as much as possible.

[0003] Currently, for gateway node load balancing, there are still problems such as being unable to accurately evaluate the load status of gateway nodes based on gateway node information, unable to screen gateway nodes according to task data information to obtain task data preloading node information, unable to accurately analyze task data preloading nodes, and unable to accurately evaluate the load status of task data preloading nodes. Summary of the Invention

[0004] To solve the above technical problems, an intelligent gateway load balancing method and system based on task data preloading are provided. The technical solution of the present invention solves the problems in the above background art, namely, being unable to accurately evaluate the load status of gateway nodes based on gateway node information, unable to screen gateway nodes according to task data information to obtain task data preloading node information, unable to accurately analyze task data preloading nodes, and unable to accurately evaluate the load status of task data preloading nodes.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] An intelligent gateway load balancing method based on task data preloading, comprising:

[0007] Obtaining gateway node information, where the gateway node information includes gateway node topology structure information and gateway node parameter information;

[0008] Obtaining gateway node load information according to the gateway node information, where the gateway node load information includes gateway node connection information and gateway node data transmission volume information;

[0009] Obtain the gateway node load index according to the gateway node load information and the gateway node parameter information;

[0010] Obtain the task data information, where the task data information includes task data type information and task data volume information;

[0011] Obtain the task data preloading node information according to the task data information;

[0012] Obtain the node task preloading coefficient according to the task data preloading node information and the gateway node load index;

[0013] Obtain the node task preloading coefficient threshold based on the gateway load balancing requirement;

[0014] Judge whether the task data preloading node meets the requirement according to the node task preloading coefficient and the node task preloading coefficient threshold. If the node task preloading coefficient is lower than the node task preloading coefficient threshold, the task data preloading node is available, and preload the task data according to the task data preloading node information;

[0015] If the node task preloading coefficient is higher than the node task preloading coefficient threshold, the task data preloading node is unavailable, and adjust the task data preloading node according to the node task preloading coefficient;

[0016] Preferably, the obtaining the task data preloading node information according to the task data information specifically includes:

[0017] Obtain the gateway node data transmission information and the gateway parallel connection information according to the gateway node load information;

[0018] Obtain the gateway node load index according to the gateway node data transmission information, the gateway parallel connection information and the gateway node parameter information;

[0019] Obtain the node data transmission index according to the task data information and the gateway node information;

[0020] Obtain the task data preloading node information according to the node data transmission index, where the task data preloading node is the gateway node corresponding to the maximum value of the node data transmission index;

[0021] Among them, the calculation formula of the gateway node load index is:

[0022]

[0023] In the formula, E is the gateway node load index, m is the total number of gateway node data, σ is the parallel connection load coefficient, b j is the size of the jth data, δ jis the gateway load coefficient of the j-th data, where b0 is the standard value of data transmission of the gateway node, ε is the threshold value of data size affected by gateway node transmission, α is the influence coefficient of gateway node bandwidth data, A is the gateway node bandwidth, β is the influence coefficient of gateway node memory data, B is the gateway node memory, ω s is the s-th performance index of the gateway node, and h is the total number of gateway node performance indexes;

[0024] The calculation formula of the node data transmission index is:

[0025]

[0026] In the formula, W(g) is the node data transmission index of the g-th gateway node, D is the task data size, V g is the transmission rate of the task data preloading node, T g is the standard queuing delay of the data of the g-th gateway node, E is the gateway node load index, P g is the packet loss rate of the g-th gateway node.

[0027] Preferably, obtaining the node task preloading coefficient according to the task data preloading node information and the gateway node load index specifically includes:

[0028] Obtaining the node data transmission index of the task data preloading node according to the task data preloading node information;

[0029] Obtaining the node task preloading coefficient according to the node data transmission index and the gateway node load index;

[0030] The calculation formula of the node task preloading coefficient is:

[0031]

[0032] In the formula, k is the node task preloading coefficient, and δ is the gateway load coefficient of the task data.

[0033] Preferably, adjusting the task data preloading node according to the node task preloading coefficient specifically includes:

[0034] Dividing the task data according to the task data information to obtain task data shard information, where the task data shard information includes task data shard size information and task data shard type information;

[0035] Obtaining the task data shard characteristic coefficient according to the task data shard information, where the task data shard characteristic coefficient is used to represent the importance of the task data shard;

[0036] Performing gateway node planning on the task data shards according to the task data shard characteristic coefficient to obtain shard preloading node information;

[0037] Obtain the task data shard quality index according to the task data shard information, the task data shard characteristic coefficient, and the shard preloading node information;

[0038] Adjust the task data shards according to the task data shard quality index until the task data shard quality index reaches the maximum value.

[0039] Preferably, the obtaining of the task data shard quality index according to the task data shard information, the task data shard characteristic coefficient, and the shard preloading node information specifically includes:

[0040] Obtain the task data shard characteristic coefficient according to the task data shard information;

[0041] Obtain the task data shard quality index according to the task data shard information, the task data shard characteristic coefficient, and the shard preloading node information;

[0042] Among them, the calculation formula for the task data shard characteristic coefficient is:

[0043]

[0044] In the formula, μ(x) is the task data shard characteristic coefficient of the x-th task data shard, dx is the size of the x-th task data shard, is the access demand coefficient of the x-th task data shard, where if the type of the x-th task data shard is random access, then If the type of the x-th task data shard is sequential access, then D is the task data size;

[0045] The calculation formula for the task data shard quality index is:

[0046]

[0047] In the formula, Q is the task data shard quality index, Vx is the preloading gateway node transmission speed of the x-th task data shard, P x is the packet loss rate of the preloading gateway node of the x-th task data shard, tx is the task data recombination time of the x-th task data shard, and n is the total number of task data shards.

[0048] Preferably, the splitting of the task data according to the task data information to obtain the task data shard information specifically includes:

[0049] Split the task data according to the task data information and the node task preloading coefficient to obtain the task data shard information, and the task data shard information includes task data main shard information and task data secondary shard information;

[0050] Among them, the shard preloading node of the task data main shard is the task data preloading node, and the node task preloading coefficient of the task data main shard is lower than the node task preloading coefficient threshold;

[0051] According to the task data main shard information, obtain the task data main shard characteristic coefficient and the node data transmission index of the task data main shard;

[0052] According to the task data main shard characteristic coefficient and the node data transmission index, determine whether the task data main shard meets the data shard requirements;

[0053] If Then the task data main shard meets the standard. According to the task data shard characteristic coefficient, perform gateway node planning on the task data secondary shard to obtain shard preloading node information;

[0054] If Then the task data main shard does not meet the standard. According to the task data information, split the task data to obtain task data shard information;

[0055] Among them, μ(0) is the task data shard characteristic coefficient of the task data main shard, W(0) is the node data transmission index of the task data main shard, and θ is the task data main shard quality coefficient.

[0056] Furthermore, a gateway load intelligent balancing system based on task data preloading is proposed to implement the above-mentioned balancing method, including:

[0057] The main control module is used to determine whether the task data preloading node meets the requirements according to the node task preloading coefficient and the node task preloading coefficient threshold, determine whether the task data main shard meets the data shard requirements according to the task data main shard characteristic coefficient and the node data transmission index, obtain the task data preloading node information according to the task data information, split the task data according to the task data information to obtain task data shard information, and perform gateway node planning on the task data shard according to the task data shard characteristic coefficient to obtain shard preloading node information;

[0058] The information acquisition module is used to acquire gateway node information, gateway node topology structure information, gateway node parameter information, and task data information. The task data information includes task data type information and task data volume information. According to the gateway node information, obtain the gateway node load information, gateway node connection information, and gateway node data transmission volume information, and transmit them to the evaluation module;

[0059] An evaluation module, which is used to obtain the data transmission information and parallel connection information of the gateway node according to the gateway node load information, obtain the gateway node load index according to the gateway node data transmission information, gateway parallel connection information and gateway node parameter information, obtain the node task preloading coefficient according to the node data transmission index and the gateway node load index, obtain the node data transmission index according to the task data information and the gateway node information, obtain the task data sharding characteristic coefficient according to the task data sharding information, and obtain the task data sharding quality index according to the task data sharding information, task data sharding characteristic coefficient and sharding preloading node information;

[0060] A display module, which interacts with the main control module and is used to display the gateway node load index, node data transmission index, node task preloading coefficient and task data preloading node information.

[0061] Optionally, the main control module specifically includes:

[0062] A control unit, which is used to obtain the task data preloading node information according to the task data information, split the task data according to the task data information to obtain the task data sharding information, and perform gateway node planning on the task data sharding according to the task data sharding characteristic coefficient to obtain the sharding preloading node information;

[0063] An information receiving unit, which interacts with the information acquisition module and the evaluation module and is used to receive data and transmit it to the judgment unit;

[0064] A judgment unit, which is used to judge whether the task data preloading node meets the requirements according to the node task preloading coefficient and the node task preloading coefficient threshold, and judge whether the task data main shard meets the data sharding requirements according to the task data main shard characteristic coefficient and the node data transmission index.

[0065] Optionally, the information acquisition module specifically includes:

[0066] A first acquisition unit, which is used to acquire the gateway node information, gateway node topology structure information, gateway node parameter information, and task data information, and the task data information includes task data type information and task data volume information;

[0067] A second acquisition unit, which is used to acquire the gateway node load information, gateway node connection information and gateway node data transmission volume information according to the gateway node information and transmit them to the evaluation module.

[0068] Optionally, the evaluation module specifically includes:

[0069] A node evaluation unit, which is used to obtain the data transmission information and parallel connection information of the gateway node according to the gateway node load information, obtain the gateway node load index according to the gateway node data transmission information, gateway parallel connection information and gateway node parameter information, and obtain the node task preloading coefficient according to the node data transmission index and the gateway node load index;

[0070] A task data evaluation unit, which is used to obtain the node data transmission index according to the task data information and the gateway node information, obtain the task data sharding feature coefficient according to the task data sharding information, and obtain the task data sharding quality index according to the task data sharding information, the task data sharding feature coefficient and the sharding preloading node information.

[0071] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0072] The present invention provides an intelligent gateway load balancing method and system based on task data preloading. By accurately evaluating the load status of the gateway node through the gateway node load index, obtaining the node data transmission index according to the task data information and the gateway node information, and obtaining the task data preloading node information according to the node data transmission index, the task data preloading efficiency is improved. By using the node task preloading coefficient to evaluate the task data preloading status, the load balancing of the gateway node is achieved, the system performance is improved, and the stability and adaptability of the system are enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a flowchart of an intelligent gateway load balancing method based on task data preloading proposed by the present invention;

[0074] Figure 2 It is a flowchart for obtaining the task data preloading node information in the present invention;

[0075] Figure 3 It is a flowchart for obtaining the task data sharding quality index in the present invention;

[0076] Figure 4 It is a block diagram of the structure of an intelligent gateway load balancing system based on task data preloading proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0077] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations.

[0078] Referring to Figure 1 - Figure 3 As shown, an intelligent gateway load balancing method based on task data preloading in an embodiment of the present invention includes:

[0079] Obtain gateway node information, where the gateway node information includes gateway node topology structure information and gateway node parameter information;

[0080] According to the gateway node information, obtain gateway node load information, where the gateway node load information includes gateway node connection information and gateway node data transmission volume information;

[0081] According to the gateway node load information and gateway node parameter information, obtain the gateway node load index;

[0082] Obtain task data information, where the task data information includes task data type information and task data volume information;

[0083] According to the task data information, obtain task data preloading node information;

[0084] Specifically, according to the task data information, obtaining task data preloading node information specifically includes:

[0085] According to the gateway node load information, obtain gateway node data transmission information and gateway parallel connection information;

[0086] According to the gateway node data transmission information, gateway parallel connection information and gateway node parameter information, obtain the gateway node load index;

[0087] According to the task data information and gateway node information, obtain the node data transmission index;

[0088] According to the node data transmission index, obtain task data preloading node information, where the task data preloading node is the gateway node corresponding to the maximum value of the node data transmission index;

[0089] Among them, the calculation formula for the gateway node load index is:

[0090]

[0091] In the formula, E is the gateway node load index, m is the total number of gateway node data, σ is the parallel connection load coefficient, b j is the size of the jth data, δ j is the gateway load coefficient of the jth data, where b0 is the standard value of the gateway node transmission data, ε is the threshold value of the data size affected by the gateway node transmission, α is the coefficient affecting the gateway node bandwidth data, A is the gateway node bandwidth, β is the coefficient affecting the gateway node memory data, B is the gateway node memory, ω s is the sth performance index of the gateway node, and h is the total number of gateway node performance indexes;

[0092] The calculation formula for the node data transmission index is:

[0093]

[0094] Wherein, W(g) is the node data transmission index of the g-th gateway node, D is the task data size, V g is the task data preloading node transmission rate, T g is the data queuing standard delay of the g-th gateway node, E is the gateway node load index, P g is the packet loss rate of the g-th gateway node.

[0095] In this solution, the gateway node load index is obtained through the gateway node data transmission information, gateway parallel connection information, and gateway node parameter information. The gateway node load status is accurately evaluated through the gateway node load index. The node data transmission index is obtained through the task data information and gateway node information. According to the node data transmission index, the task data preloading node information is obtained, improving the task data preloading efficiency;

[0096] It can be understood that the preloading efficiency of different gateway nodes for task data is also very different, and the most suitable gateway node is selected through the node data transmission index.

[0097] According to the task data preloading node information and gateway node load index, the node task preloading coefficient is obtained;

[0098] Specifically, according to the task data preloading node information and gateway node load index, the node task preloading coefficient is obtained, which specifically includes:

[0099] According to the task data preloading node information, the node data transmission index of the task data preloading node is obtained;

[0100] According to the node data transmission index and gateway node load index, the node task preloading coefficient is obtained;

[0101] The calculation formula of the node task preloading coefficient is:

[0102]

[0103] Wherein, k is the node task preloading coefficient, and δ is the gateway load coefficient of the task data.

[0104] In this solution, the node task preloading coefficient is obtained through the node data transmission index and gateway node load index. Through the node task preloading coefficient, the task data preloading status is evaluated, achieving load balancing of the gateway node, improving system performance, and enhancing the stability and adaptability of the system;

[0105] It is understandable that the task data preloading node may cause unbalanced load on the gateway node during task data preloading. Evaluating the task data preloading node through the node task preloading coefficient ensures the stability and reliability of the gateway node load.

[0106] Based on the gateway load balancing requirement, obtain the threshold of the node task preloading coefficient;

[0107] According to the node task preloading coefficient and the threshold of the node task preloading coefficient, determine whether the task data preloading node meets the requirement. If the node task preloading coefficient is lower than the threshold of the node task preloading coefficient, the task data preloading node is available, and preload the task data according to the task data preloading node information;

[0108] If the node task preloading coefficient is higher than the threshold of the node task preloading coefficient, the task data preloading node is unavailable, and adjust the task data preloading node according to the node task preloading coefficient.

[0109] Specifically, adjusting the task data preloading node according to the node task preloading coefficient specifically includes:

[0110] According to the task data information, split the task data to obtain task data shard information, where the task data shard information includes task data shard size information and task data shard type information;

[0111] According to the task data shard information, obtain the task data shard characteristic coefficient, where the task data shard characteristic coefficient is used to represent the importance degree of the task data shard;

[0112] According to the task data shard characteristic coefficient, perform gateway node planning on the task data shard to obtain shard preloading node information;

[0113] According to the task data shard information, the task data shard characteristic coefficient, and the shard preloading node information, obtain the task data shard quality index;

[0114] According to the task data shard quality index, adjust the task data shard until the task data shard quality index reaches the maximum value.

[0115] Specifically, obtaining the task data shard quality index according to the task data shard information, the task data shard characteristic coefficient, and the shard preloading node information specifically includes:

[0116] According to the task data shard information, obtain the task data shard characteristic coefficient;

[0117] According to the task data shard information, the task data shard characteristic coefficient, and the shard preloading node information, obtain the task data shard quality index;

[0118] Among them, the calculation formula for the task data shard feature coefficient is as follows:

[0119]

[0120] In the formula, μ(x) is the task data shard feature coefficient of the x-th task data shard, dx is the size of the x-th task data shard, is the access requirement coefficient of the x-th task data shard. If the type of the x-th task data shard is random access, then If the type of the x-th task data shard is sequential access, then D is the task data size;

[0121] The calculation formula for the task data shard quality index is as follows:

[0122]

[0123] In the formula, Q is the task data shard quality index, Vx is the transmission speed of the preloading gateway node of the x-th task data shard, P x is the packet loss rate of the preloading gateway node of the x-th task data shard, tx is the task data recombination time of the x-th task data shard, and n is the total number of task data shards.

[0124] Furthermore, according to the task data information, the task data is segmented to obtain the task data shard information, which specifically includes:

[0125] The task data is segmented according to the task data information and the node task preloading coefficient to obtain the task data shard information, and the task data shard information includes the task data main shard information and the task data secondary shard information;

[0126] Among them, the shard preloading node of the task data main shard is the task data preloading node and the node task preloading coefficient of the task data main shard is lower than the node task preloading coefficient threshold;

[0127] According to the task data main shard information, the task data main shard feature coefficient and the node data transmission index of the task data main shard are obtained;

[0128] According to the task data main shard feature coefficient and the node data transmission index, it is judged whether the task data main shard meets the data shard requirements;

[0129] If then the task data main shard meets the standard, and according to the task data shard feature coefficient, the gateway node planning is carried out for the task data secondary shard to obtain the shard preloading node information;

[0130] If Then the main shard of the task data does not meet the standard. The task data is segmented according to the task data information to obtain the task data shard information;

[0131] Among them, μ(0) is the task data shard feature coefficient of the main shard of the task data, W(0) is the node data transmission index of the main shard of the task data, and θ is the task data main shard quality coefficient.

[0132] In this solution, through the task data information, the task data is segmented to obtain the task data shard information. According to the task data shard information, the task data shard feature coefficient is obtained. According to the task data shard feature coefficient, the gateway node planning for the task data shards is carried out to obtain the preloaded node information for the shards. According to the task data shard information, the task data shard feature coefficient, and the preloaded node information for the shards, the task data shard quality index is obtained. According to the task data shard quality index, the task data shards are adjusted until the task data shard quality index reaches the maximum value;

[0133] It can be understood that when the task data preloaded node does not meet the actual requirements, by segmenting the task data into the main shard of the task data and the secondary shard of the task data, the main shard of the task data is still preloaded through the original task data preloaded node, and the gateway node planning for the secondary shard of the task data is carried out. While saving resources, the task data preloading efficiency is improved. Through the task data main shard feature coefficient and the node data transmission index, the main shard of the task data is evaluated to avoid the situation where the main shard of the task data accounts for too small a proportion in the task data, resulting in too low data shard quality.

[0134] Refer to Figure 4 As shown in, further, in combination with the above-mentioned intelligent gateway load balancing method based on task data preloading, an intelligent gateway load balancing system based on task data preloading is proposed, including:

[0135] The main control module is used to judge whether the task data preloaded node meets the requirements according to the node task preloading coefficient and the node task preloading coefficient threshold, judge whether the main shard of the task data meets the data shard requirements according to the task data main shard feature coefficient and the node data transmission index, obtain the task data preloaded node information according to the task data information, segment the task data according to the task data information to obtain the task data shard information, and carry out gateway node planning for the task data shards according to the task data shard feature coefficient to obtain the preloaded node information for the shards;

[0136] An information acquisition module, which is used to acquire gateway node information, gateway node topology information, gateway node parameter information, and task data information. The task data information includes task data type information and task data volume information. According to the gateway node information, gateway node load information, gateway node connection information, and gateway node data transmission volume information are acquired and transmitted to the evaluation module;

[0137] An evaluation module, which is used to acquire gateway node data transmission information and gateway parallel connection information according to the gateway node load information. According to the gateway node data transmission information, gateway parallel connection information, and gateway node parameter information, the gateway node load index is acquired. According to the node data transmission index and the gateway node load index, the node task preloading coefficient is acquired. According to the task data information and the gateway node information, the node data transmission index is acquired. According to the task data sharding information, the task data sharding feature coefficient is acquired. According to the task data sharding information, the task data sharding feature coefficient, and the sharding preloading node information, the task data sharding quality index is acquired;

[0138] A display module, which interacts with the main control module and is used to display the gateway node load index, the node data transmission index, the node task preloading coefficient, and the task data preloading node information.

[0139] The main control module specifically includes:

[0140] A control unit, which is used to acquire task data preloading node information according to the task data information, split the task data according to the task data information to acquire task data sharding information, and perform gateway node planning on the task data shards according to the task data sharding feature coefficient to acquire sharding preloading node information;

[0141] An information receiving unit, which interacts with the information acquisition module and the evaluation module and is used to receive data and transmit it to the judgment unit;

[0142] A judgment unit, which is used to judge whether the task data preloading node meets the requirements according to the node task preloading coefficient and the node task preloading coefficient threshold, and judge whether the main task data shard meets the data sharding requirements according to the main task data shard feature coefficient and the node data transmission index.

[0143] The information acquisition module specifically includes:

[0144] A first acquisition unit, which is used to acquire gateway node information, gateway node topology information, gateway node parameter information, and task data information. The task data information includes task data type information and task data volume information;

[0145] A second acquisition unit, which is configured to acquire gateway node load information, gateway node connection information, and gateway node data transmission volume information according to gateway node information and transmit same to an evaluation module.

[0146] The evaluation module specifically includes:

[0147] A node evaluation unit, which is configured to acquire gateway node data transmission information and gateway parallel connection information according to gateway node load information, acquire a gateway node load index according to the gateway node data transmission information, the gateway parallel connection information, and gateway node parameter information, and acquire a node task preloading coefficient according to the node data transmission index and the gateway node load index;

[0148] A task data evaluation unit, which is configured to acquire a node data transmission index according to task data information and gateway node information, acquire a task data fragmentation characteristic coefficient according to task data fragmentation information, and acquire a task data fragmentation quality index according to the task data fragmentation information, the task data fragmentation characteristic coefficient, and shard preloading node information.

[0149] In summary, the advantages of the present invention are as follows: A gateway node load index is acquired through gateway node data transmission information, gateway parallel connection information, and gateway node parameter information, the gateway node load condition is accurately evaluated through the gateway node load index, a node data transmission index is acquired according to task data information and gateway node information, task data preloading node information is acquired according to the node data transmission index, the task data preloading efficiency is improved, a node task preloading coefficient is acquired through the task data preloading node information and the gateway node load index, and the task data preloading condition is evaluated through the node task preloading coefficient, achieving load balancing of the gateway node, improving the system performance, and enhancing the stability and adaptability of the system.

[0150] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and what is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and all these changes and improvements fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A gateway load intelligent balancing method based on task data preloading, characterized in that: include: Acquire gateway node information, wherein the gateway node information includes gateway node topology information and gateway node parameter information; According to the gateway node information, the gateway node load information is acquired, wherein the gateway node load information includes the gateway node connection information and the gateway node data transmission volume information; Obtaining a gateway node load index according to the gateway node load information and the gateway node parameter information; Acquire task data information, wherein the task data information includes task data type information and task data amount information; According to the task data information, obtain the task data preload node information; Obtain the node task preload coefficient based on the task data preload node information and gateway node load index; Based on the gateway load balancing requirements, obtain the node task preload coefficient threshold; According to the node task preloading coefficient and the node task preloading coefficient threshold, determine whether the task data preloading node meets the requirements. If the node task preloading coefficient is lower than the node task preloading coefficient threshold, the task data preloading node is available, and the task data is preloaded according to the task data preloading node information. If the node task preload coefficient is higher than the node task preload coefficient threshold, the task data preload node is unavailable, and the task data preload node is adjusted according to the node task preload coefficient.

2. According to claim 1, a gateway load intelligent balancing method based on task data preloading is characterized in that: The step of obtaining task data preload node information according to the task data information specifically includes: According to the gateway node load information, the gateway node data transmission information and the gateway parallel connection information are obtained; Obtaining a gateway node load index according to gateway node data transmission information, gateway parallel connection information, and gateway node parameter information; According to the task data information and gateway node information, obtain the node data transmission index; According to the node data transmission index, the task data preloading node information is obtained, and the task data preloading node is the gateway node corresponding to the maximum value of the node data transmission index; The calculation formula of the gateway node load index is: Where E is the gateway node load index, m is the total number of gateway node data, σ is the parallel connection load factor, and b j is the jth data size, δ j is the gateway load coefficient of the jth data, where b0 is the standard value of the gateway node transmission data, ε is the threshold of the size of the data affected by the gateway node transmission, α is the gateway node bandwidth data impact coefficient, A is the gateway node bandwidth, β is the gateway node memory data impact coefficient, B is the gateway node memory, ω s is the sth performance index of the gateway node, and h is the total number of performance indexes of the gateway nodes; The calculation formula of node data transmission index is: Where W(g) is the node data transmission index of the g-th gateway node, D is the task data size, and V g Task data preload node transmission rate, T g is the data queuing standard delay of the g-th gateway node, E is the gateway node load index, P g is the packet loss rate of the g-th gateway node.

3. According to claim 2, a gateway load intelligent balancing method based on task data preloading is characterized in that: The step of obtaining the node task preloading coefficient according to the task data preloading node information and the gateway node load index specifically includes: According to the task data preloading node information, obtain the node data transmission index of the task data preloading node; Obtain the node task preload coefficient according to the node data transmission index and the gateway node load index; The calculation formula of node task preload coefficient is: Where k is the node task preload coefficient, and δ is the gateway load coefficient of task data.

4. According to claim 1, a gateway load intelligent balancing method based on task data preloading is characterized in that: The step of adjusting the task data preloading node according to the node task preloading coefficient specifically includes: According to the task data information, the task data is segmented to obtain task data segmentation information, wherein the task data segmentation information includes task data segmentation size information and task data segmentation type information; According to the task data slicing information, a task data slicing characteristic coefficient is obtained, where the task data slicing characteristic coefficient is used to indicate the importance of the task data slicing; According to the task data shard characteristic coefficient, the gateway node is planned for the task data shard, and the shard preload node information is obtained; Obtain the task data shard quality index according to the task data shard information, the task data shard characteristic coefficient and the shard preload node information; According to the task data shard quality index, the task data shard is adjusted until the task data shard quality index reaches a maximum value.

5. According to claim 4, a gateway load intelligent balancing method based on task data preloading is characterized in that: The step of obtaining the task data slice quality index according to the task data slice information, the task data slice characteristic coefficient and the slice preload node information specifically includes: According to the task data sharding information, obtain the task data sharding characteristic coefficient; Obtain the task data shard quality index according to the task data shard information, the task data shard characteristic coefficient and the shard preload node information; Among them, the calculation formula of the task data sharding characteristic coefficient is: Where μ(x) is the task data partition characteristic coefficient of the xth task data partition, d x is the size of the x-th task data shard, is the access demand coefficient of the x-th task data shard, where if the x-th task data shard type is random access, then If the data shard type of the xth task is sequential access, then D is the task data size; The calculation formula of the task data shard quality index is: Where Q is the task data shard quality index, V x is the preload gateway node transmission speed of the xth task data shard, P x is the packet loss rate of the preloaded gateway node of the xth task data shard, t x is the task data reorganization time of the xth task data shard, and n is the total number of task data shards.

6. A gateway load intelligent balancing method based on task data preloading according to claim 4, characterized in that: The step of segmenting the task data according to the task data information to obtain the task data segmentation information specifically includes: According to the task data information and the node task preload coefficient, the task data is divided to obtain task data shard information, wherein the task data shard information includes task data primary shard information and task data secondary shard information; The shard preload node of the task data primary shard is the task data preload node and the node task preload coefficient of the task data primary shard is lower than the node task preload coefficient threshold; According to the task data primary shard information, obtain the task data primary shard characteristic coefficient and the node data transmission index of the task data primary shard; According to the characteristic coefficient of the task data main shard and the node data transmission index, determine whether the task data main shard meets the data sharding requirements; like If the primary shard of the task data meets the standard, the gateway node planning is performed on the secondary shard of the task data according to the characteristic coefficient of the task data shard, and the shard preload node information is obtained; like If the main shard of the task data does not meet the standard, the task data is segmented according to the task data information to obtain the task data shard information; Among them, μ(0) is the task data shard characteristic coefficient of the task data main shard, W(0) is the node data transmission index of the task data main shard, and θ is the task data main shard quality coefficient.

7. A gateway load intelligent balancing system based on task data preloading, used to implement the balancing method according to any one of claims 1 to 6, characterized in that: include: A main control module, wherein the main control module is used to determine whether the task data preloading node meets the requirements according to the node task preloading coefficient and the node task preloading coefficient threshold, determine whether the task data main shard meets the data sharding requirements according to the task data main sharding characteristic coefficient and the node data transmission index, obtain the task data preloading node information according to the task data information, divide the task data according to the task data information, obtain the task data sharding information, perform gateway node planning for the task data shard according to the task data sharding characteristic coefficient, and obtain the shard preloading node information; An information acquisition module, the information acquisition module is used to acquire gateway node information, gateway node topology information, gateway node parameter information, and task data information, the task data information includes task data type information and task data volume information, and based on the gateway node information, acquire gateway node load information, gateway node connection information, and gateway node data transmission volume information, and transmit the information to the evaluation module; An evaluation module, wherein the evaluation module is used to obtain gateway node data transmission information and gateway parallel connection information according to gateway node load information, obtain a gateway node load index according to the gateway node data transmission information, the gateway parallel connection information and the gateway node parameter information, obtain a node task preload coefficient according to the node data transmission index and the gateway node load index, obtain a node data transmission index according to the task data information and the gateway node information, obtain a task data segmentation characteristic coefficient according to the task data segmentation information, and obtain a task data segmentation quality index according to the task data segmentation information, the task data segmentation characteristic coefficient and the segmentation preload node information; A display module interacts with the main control module and is used to display the gateway node load index, the node data transmission index, the node task preload coefficient and the task data preload node information.

8. The gateway load intelligent balancing system based on task data preloading according to claim 7 is characterized in that: The main control module specifically includes: A control unit, the control unit is used to obtain task data preload node information according to task data information, segment the task data according to the task data information, obtain task data slicing information, perform gateway node planning on the task data slicing according to the task data slicing characteristic coefficient, and obtain slicing preload node information; An information receiving unit, which interacts with the information acquisition module and the evaluation module to receive data and transmit it to the judgment unit; A judgment unit is used to judge whether the task data preloading node meets the requirements based on the node task preloading coefficient and the node task preloading coefficient threshold, and to judge whether the task data main shard meets the data sharding requirements based on the task data main shard characteristic coefficient and the node data transmission index.

9. The gateway load intelligent balancing system based on task data preloading according to claim 7, characterized in that: The information acquisition module specifically includes: A first acquisition unit, the first acquisition unit is used to acquire gateway node information, gateway node topology information, gateway node parameter information, and task data information, wherein the task data information includes task data type information and task data volume information; The second acquisition unit is used to acquire gateway node load information, gateway node connection information and gateway node data transmission volume information according to the gateway node information, and transmit the information to the evaluation module.

10. The gateway load intelligent balancing system based on task data preloading according to claim 7, characterized in that: The evaluation module specifically includes: A node evaluation unit, the node evaluation unit is used to obtain gateway node data transmission information and gateway parallel connection information according to gateway node load information, obtain a gateway node load index according to the gateway node data transmission information, the gateway parallel connection information and the gateway node parameter information, and obtain a node task preload coefficient according to the node data transmission index and the gateway node load index; A task data evaluation unit, wherein the task data evaluation unit is used to obtain a node data transmission index based on task data information and gateway node information, obtain a task data slice characteristic coefficient based on task data slice information, and obtain a task data slice quality index based on task data slice information, task data slice characteristic coefficient and slice preload node information.

Images