Push service node load balancing management method based on big data analysis

By collecting and analyzing the performance data of the push service node, calculating the performance status coefficient and dynamically adjusting the load balancing strategy, the problem of difficulty in dynamically adjusting the weights of the push service node load balancing management in the existing technology is solved, and the resource utilization rate and user experience are improved.

CN118945170AActive Publication Date: 2024-11-12GUANGZHOU YANCHENG TECHNOLOGY CO LTD

Patent Information

Application Number
CN202411051482.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-12
Estimated Expiration
2044-08-01

AI Technical Summary

Technical Problem

The load balancing management of existing push service nodes is difficult to dynamically adjust weights, resulting in insufficient resource availability and affecting user experience and satisfaction.

Method used

By collecting and analyzing the performance data of each push service node, calculating the performance status coefficient, and generating a load balancing strategy based on the comparison between the coefficient and the preset threshold interval, monitoring performance changes in real time, and optimizing load balancing using weighted polling.

Benefits of technology

It realizes dynamic adjustment of weights according to the performance status changes in each work cycle, optimizes load balancing strategies, improves resource utilization, and improves user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118945170A_ABST
    Figure CN118945170A_ABST
Patent Text Reader

Abstract

The invention relates to the field of push services, and discloses a push service node load balancing management method based on big data analysis, which comprises the following steps: S1, collecting performance data of each push service node, including response time, error rate, connection number, request rate, response rate, data transmission speed, memory utilization rate and CPU utilization rate; s2, analyzing the collected performance data of each push service node, calculating to obtain a performance state coefficient of each push service node, and performing comparative analysis on the performance state coefficient and a preset performance state threshold interval; s3, generating a load balancing strategy according to a comparison analysis result in S2; and S4, monitoring the performance data of each push service node in real time, detecting the performance change of each push service node, and dynamically adjusting a load balancing strategy according to the change condition. According to the invention, analysis is carried out according to the performance state change of each recommended service node in the previous work period, and the performance state in the current work period is predicted, so that the weight of each recommended service node is dynamically changed, and the purpose of adjusting a load balancing strategy is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of push services, and in particular to a push service node load balancing management method based on big data analysis. Background Art

[0002] Push service refers to the technology of pushing data or information from the server to the client. Push service can be used in many scenarios, such as: 1. Real-time message push: push real-time messages to the client, such as chat software, social media; 2. Application update push: push application updates to the client, such as software updates, game updates; 3. Notification push: push notifications to the client, such as push notifications, warning notifications; 4. Data push: push data to the client, such as stock prices, weather information.

[0003] The above push service requires the use of multiple push service nodes. Whether the multiple push service nodes match the push tasks will lead to waste or shortage of resources. Therefore, big data is introduced. The use of big data analysis technology can help administrators better understand the performance of nodes, such as which node has the best performance and which node has the worst performance. However, in the process of implementing load balancing management, there are still some problems. For example, it is impossible to dynamically adjust the weight according to the performance changes of the push service nodes to achieve the purpose of dynamically adjusting the load balancing management strategy, resulting in insufficient resource availability and affecting user experience and satisfaction. Summary of the invention

[0004] The purpose of the present invention is to provide a push service node load balancing management method based on big data analysis to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A push service node load balancing management method based on big data analysis, comprising:

[0007] S1. Collect performance data of each push service node, including response time, error rate, number of connections, request rate, response rate, data transmission speed, memory usage and CPU usage;

[0008] S2. Analyze the collected performance data of each push service node, calculate the performance status coefficient of each push service node, and compare and analyze the performance status coefficient with a preset performance status threshold interval;

[0009] S3, generating a load balancing strategy according to the comparison and analysis results in S2;

[0010] S4. Monitor the performance data of each push service node in real time, detect the performance changes of each push service node, and dynamically adjust the load balancing strategy according to the changes.

[0011] As a further technical solution, the process of obtaining the performance status coefficient of each push service node is:

[0012] Get the response time A of each push service node j , error rate B j , request rate C j , response rate D j , data transmission speed E j 、Number of connections F j , memory usage G j and CPU usage H j ;

[0013] By formula:

[0014]

[0015] Calculate the performance status coefficient x of each push service node j ;

[0016] Among them, α1, α2, α3, α4, α5, α6, and α7 are preset weight coefficients, and K i (f j , g j ,h j ) is the number of remaining connections f j 、Remaining memory capacity g j 、Remaining CPU resources h j The judgment function of .

[0017] As a further technical solution, obtaining the K i (f j , g j ,h j ) process is:

[0018] Get the number of connections f required for the current push task ji 、Memory capacity g ji and CPU resources ji ;

[0019] According to the number of connections f ji 、Memory capacity g ji and CPU resources ji Establish:

[0020]

[0021] in, is the preset weight factor.

[0022] As a further technical solution, the performance state coefficient x j and the preset performance status threshold interval [x jth , x jch The process of comparative analysis is as follows:

[0023] If x j ∈[x jth , x jch ], it is judged that the performance status of the current push service node for the current push task is good;

[0024] like It is determined that the performance status of the current push service node for the current push task is poor.

[0025] As a further technical solution, the process of generating a load balancing strategy according to the comparison and analysis results in S2 is:

[0026] S31, analyzing and comparing the current push task with each push service node;

[0027] S32, remove push service nodes with poor performance;

[0028] S33, after the remaining push service nodes are arranged in descending order according to the performance status coefficient, weights X are set for the push service nodes from high to low. j .

[0029] As a further technical solution, the process of dynamically adjusting the load balancing strategy according to the changing situation is:

[0030] S41. Obtain a curve x showing the performance status coefficient of each push service node changing over time within a statistical period. j (t);

[0031] S42, based on big data, obtain the parameter curve x of the performance status coefficient of each push service node changing over time j0 (t);

[0032] S43. Acquisition The corresponding time points take two adjacent time points as a time period;

[0033] S44, according to x j (t) and x j0 (t) The deviation coefficient of each push service node is obtained by comparing the situation analysis;

[0034] S45. According to the comparison result between the deviation coefficient and the deviation threshold, determine whether to adjust the load balancing strategy of the current push service node.

[0035] As a further technical solution, the process of obtaining the deviation coefficient of each push service node is:

[0036] By formula:

[0037]

[0038] Calculate the deviation coefficient θ of each push service node j ;

[0039] Among them, β y is the preset proportional coefficient corresponding to y time periods, t i ,t i+1 are the starting time point and the ending time point of the time period respectively, and N is the total number of time periods.

[0040] As a further technical solution, the working process of S45 is as follows:

[0041] The deviation coefficient θ j With the deviation threshold θ j0 Compare;

[0042] If θ j ≥θ j0 , it is judged that the performance status of the recommended service node fluctuates greatly, and the load balancing strategy of the recommended service node needs to be adjusted, and the adjustment process is:

[0043] By formula:

[0044]

[0045] Calculate the weight of the recommended service node after adjustment

[0046] If θ j <θ j0 , it is determined that the performance status of the service node fluctuates slightly and no load balancing strategy adjustment is performed.

[0047] Beneficial effects of the present invention:

[0048] (1) The present invention analyzes the performance status changes of each recommended service node in the previous working cycle and predicts the performance status in the current working cycle to dynamically change the weight of each recommended service node and achieve the purpose of adjusting the load balancing strategy;

[0049] (2) The present invention compares the performance status coefficient with the preset performance status threshold interval. It is obvious that the push service node within the preset performance status threshold interval has a moderate match with the current push task, while the push service node that does not belong to the preset performance status threshold interval indicates that the requirements are not met or exceed the requirements too much, which will cause a waste of resources. Therefore, the push service nodes that do not belong to the preset performance status threshold interval are not recommended. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The present invention will be further described below in conjunction with the accompanying drawings.

[0051] Figure 1 It is a diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] See also Figure 1 As shown, the present invention is a push service node load balancing management method based on big data analysis, comprising:

[0054] S1. Collect performance data of each push service node, including response time, error rate, number of connections, request rate, response rate, data transmission speed, memory usage and CPU usage;

[0055] S2. Analyze the collected performance data of each push service node, calculate the performance status coefficient of each push service node, and compare and analyze the performance status coefficient with a preset performance status threshold interval;

[0056] S3, generating a load balancing strategy according to the comparison and analysis results in S2;

[0057] S4. Monitor the performance data of each push service node in real time, detect the performance changes of each push service node, and dynamically adjust the load balancing strategy according to the changes.

[0058] In this embodiment, the performance data of each push service node is first collected, and then the collected performance data of each push service node is analyzed to calculate the performance status coefficient of each push service node, and the performance status coefficient is compared and analyzed with the preset performance status threshold interval, and a load balancing strategy is generated according to the comparison and analysis results; then the performance data of each push service node is monitored in real time, the performance changes of each push service node are detected, and the load balancing strategy is dynamically adjusted according to the changes; through the above technical scheme, the performance status of each recommended service node can be monitored by weighted polling to implement different load balancing strategies according to the push task and achieve the optimal push method; it should be noted that the process of adjusting the load balancing strategy in the present invention is based on the analysis of the performance status changes of each recommended service node in the previous working cycle, and the performance status in the current working cycle is predicted to dynamically change the weight of each recommended service node to achieve the purpose of adjusting the load balancing strategy.

[0059] The process of obtaining the performance status coefficient of each push service node is as follows:

[0060] Get the response time A of each push service node j , error rate B j , request rate C j , response rate D j , data transmission speed E j 、Number of connections F j , memory usage G j and CPU usage H j ;

[0061] By formula:

[0062]

[0063] Calculate the performance status coefficient x of each push service node j ;

[0064] Among them, α1, α2, α3, α4, α5, α6, and α7 are preset weight coefficients, and K i (f j , g j ,h j ) is the number of remaining connections F j 、Remaining memory capacity g j 、Remaining CPU resources h j The judgment function.

[0065] In this embodiment, a method for obtaining the performance status coefficient of each push service node is provided. Specifically, the data of each parameter is first obtained through an existing detection tool, and then substituted into Calculate the performance status coefficient x of each push service node j ,pass It can be seen that when the response time A j , error rate B j 、Number of connections F j , memory usage G j and CPU usage H j The smaller the value, the better the performance of the current recommended service node. j , response rate D j , data transmission speed E j The larger the value is, the better the performance of the current recommended service node is. i (f j , g j ,h j ) evaluates each push task, comprehensively considers the requirements of each push task and the performance of each push service node, so as to obtain a more accurate performance status coefficient, thereby truly reflecting the actual performance of each recommended service node.

[0066] Get the K i (f j , g j ,h j ) process is:

[0067] Get the number of connections f required for the current push task ji 、Memory capacity g ji and CPU resources ji ;

[0068] According to the number of connections f ji 、Memory capacity g ji and CPU resources ji Establish:

[0069]

[0070] in, is the preset weight factor.

[0071] In this embodiment, a method for obtaining K is provided. i (f j , g j ,h j ) method, first obtain the number of connections F required for the current push task ji 、Memory capacity g ji and CPU resources ji ,Depend on Obviously, as long as f j ≤f ji or j ≤g jior j ≤h ji If any one of the above is true, it means that the current push service node does not meet the requirements of the current push task, so the value is 0, and the current push service node also cannot meet the performance requirements of the current push task. j >f ji And g j >g ji And h j >h ji When By taking the weighted sum of the ratios of the actual status of the three parameters and the actual status required by the current push task, the actual status of the current push service node that meets the current push task can be obtained, thereby obtaining a more accurate performance status coefficient x j , to improve the effective implementation of subsequent load balancing strategies.

[0072] The performance status coefficient x j and the preset performance status threshold interval [x jth , x jch The process of comparative analysis is as follows:

[0073] If x j ∈[x jth , x jch ], it is judged that the performance status of the current push service node for the current push task is good;

[0074] like It is determined that the performance status of the current push service node for the current push task is poor.

[0075] In this embodiment, the performance state coefficient x j and the preset performance status threshold interval [x jth , x jch ] for comparison, it is obvious that the push service nodes within the preset performance status threshold range have a moderate match with the current push task, while the push service nodes that do not belong to the preset performance status threshold range indicate that they do not meet the requirements or exceed the requirements too much, which will cause a waste of resources. Therefore, those that do not belong to the preset performance status threshold range are not recommended.

[0076] The process of generating a load balancing strategy based on the comparison and analysis results in S2 is as follows:

[0077] S31, analyzing and comparing the current push task with each push service node;

[0078] S32, remove push service nodes with poor performance;

[0079] S33, after the remaining push service nodes are arranged in descending order according to the performance status coefficient, weights X are set for the push service nodes from high to low. j .

[0080] The process of dynamically adjusting the load balancing strategy according to the changing situation is as follows:

[0081] S41. Obtain a curve x showing the performance status coefficient of each push service node changing over time within a statistical period. j (t);

[0082] S42, based on big data, obtain the parameter curve x of the performance status coefficient of each push service node changing over time j0 (t);

[0083] S43. Acquisition The corresponding time points take two adjacent time points as a time period;

[0084] S44, according to x j (t) and x j0 (t) The deviation coefficient of each push service node is obtained by comparing the situation analysis;

[0085] S45. According to the comparison result between the deviation coefficient and the deviation threshold, determine whether to adjust the load balancing strategy of the current push service node.

[0086] The process of obtaining the deviation coefficient of each push service node is:

[0087] By formula:

[0088]

[0089] Calculate the deviation coefficient θ of each push service node j ;

[0090] Among them, β y is the preset proportional coefficient corresponding to y time periods, t i ,t i+1 are the starting time point and the ending time point of the time period respectively, and N is the total number of time periods.

[0091] The working process of S45 is as follows:

[0092] The deviation coefficient θ j With the deviation threshold θ j0 Compare;

[0093] If θ j ≥θ j0 , it is judged that the performance status of the recommended service node fluctuates greatly, and the load balancing strategy of the recommended service node needs to be adjusted, and the adjustment process is:

[0094] By formula:

[0095]

[0096] Calculate the weight of the recommended service node after adjustment γ j is the conversion coefficient, which is selected and determined based on historical experimental data;

[0097] If θ j <θ j0 , it is determined that the performance status of the service node fluctuates slightly and no load balancing strategy adjustment is performed.

[0098] In this embodiment, the current push task is first analyzed and compared with each push service node, and then the push service nodes with poor performance status are eliminated; then the remaining push service nodes are arranged in descending order according to the performance status coefficient, and the weights X are set for the push service nodes from high to low. j , in order to realize the generation process of load balancing strategy, but in the actual operation process, the performance status coefficient of each push service node is in a dynamically changing state, so it is necessary to adjust the weight in combination with the dynamic change to achieve the purpose of dynamically adjusting the load balancing strategy;

[0099] Specifically, first obtain the curve x of the performance status coefficient of each push service node over time within a statistical period j (t); then based on the big data, obtain the parameter curve x of the performance status coefficient of each push service node changing with time j0 (t); then by obtaining The corresponding time points are divided into two adjacent time points as a time period to achieve the time period division of a statistical cycle; then the formula Calculate the deviation coefficient θ of each push service node j ; The deviation coefficient θ j With the deviation threshold θ j0 Compare, if θ j ≥θ j0 , it is judged that the performance status of the recommended service node fluctuates greatly, and the load balancing strategy of the recommended service node needs to be adjusted, and the adjustment process is: through the formula: Calculate the weight of the recommended service node after adjustment If θ j <θ j0 , it is determined that the performance status of the service node fluctuates slightly, and no load balancing strategy adjustment is performed;

[0100] Through the above technical solution, the deviation between the change curve of the performance status coefficient of the recommended service node in the previous statistical period and the reference change curve can be used to adjust the weight of the recommended service node in the next statistical period, thereby avoiding the recommended service node after the performance status coefficient changes still occupying a large weight, resulting in the inability to effectively implement the load balancing strategy, and ultimately affecting the overall push quality.

[0101] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0102] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A push service node load balancing management method based on big data analysis, characterized in that: include: S1. Collect performance data of each push service node, including response time, error rate, number of connections, request rate, response rate, data transmission speed, memory usage and CPU usage; S2. Analyze the collected performance data of each push service node, calculate the performance status coefficient of each push service node, and compare and analyze the performance status coefficient with a preset performance status threshold range; S3, generating a load balancing strategy according to the comparison and analysis results in S2; S4. Monitor the performance data of each push service node in real time, detect the performance changes of each push service node, and dynamically adjust the load balancing strategy according to the changes.

2. The push service node load balancing management method based on big data analysis according to claim 1 is characterized in that: The process of obtaining the performance status coefficient of each push service node is as follows: Get the response time A of each push service node j , error rate B j , request rate C j , response rate D j , data transmission speed E j 、Number of connections F j , memory usage G j and CPU usage H j ; By formula: Calculate the performance status coefficient x of each push service node j ; Among them, α1, α2, α3, α4, α5, α6, and α7 are preset weight coefficients, and K i (f j , g j ,h j ) is the number of remaining connections f j 、Remaining memory capacity g j , the judgment function of the remaining CPU resources hj.

3. The push service node load balancing management method based on big data analysis according to claim 2 is characterized in that: Get the K i (f j , g j ,h j ) process is: Get the number of connections f required for the current push task ji 、Memory capacity g ji and CPU resources ji ; According to the number of connections f ji 、Memory capacity g ji and CPU resources ji Establish: in, is the preset weight factor.

4. The push service node load balancing management method based on big data analysis according to claim 3 is characterized in that: The performance status coefficient x j and the preset performance status threshold interval [x jth , x jch The process of comparative analysis is as follows: If x j ∈[x jth , x jch ], it is judged that the performance status of the current push service node for the current push task is good; like It is determined that the performance status of the current push service node for the current push task is poor.

5. The push service node load balancing management method based on big data analysis according to claim 4 is characterized in that: The process of generating a load balancing strategy based on the comparison and analysis results in S2 is as follows: S31, analyzing and comparing the current push task with each push service node; S32, remove push service nodes with poor performance; S33, after the remaining push service nodes are arranged in descending order according to the performance status coefficient, weights X are set for the push service nodes from high to low. j .

6. The push service node load balancing management method based on big data analysis according to claim 1 or 5, characterized in that: The process of dynamically adjusting the load balancing strategy according to the changing situation is as follows: S41. Obtain a curve x showing the performance status coefficient of each push service node changing over time within a statistical period. j (t); S42, based on big data, obtain the parameter curve x of the performance status coefficient of each push service node changing over time j0 (t); S43. Acquisition The corresponding time points take two adjacent time points as a time period; S44, according to x j (t) and x j0 (t) The deviation coefficient of each push service node is obtained by comparing the situation analysis; S45. According to the comparison result between the deviation coefficient and the deviation threshold, determine whether to adjust the load balancing strategy of the current push service node.

7. The push service node load balancing management method based on big data analysis according to claim 6 is characterized in that: The process of obtaining the deviation coefficient of each push service node is: By formula: Calculate the deviation coefficient θ of each push service node j ; Among them, β y is the preset proportional coefficient corresponding to y time periods, t i ,t i+1 are the starting time point and the ending time point of the time period respectively, and N is the total number of time periods.

8. The push service node load balancing management method based on big data analysis according to claim 7 is characterized in that: The working process of S45 is as follows: The deviation coefficient θ j With the deviation threshold θ j0 Compare; If θ j ≥θ j0 , it is judged that the performance status of the recommended service node fluctuates greatly, and the load balancing strategy of the recommended service node needs to be adjusted, and the adjustment process is: By formula: Calculate the weight of the recommended service node after adjustment γ j is the conversion coefficient; If θ j <θ j0 , it is determined that the performance status of the service node fluctuates slightly and no load balancing strategy adjustment is performed.

Citation Information

Patent Citations

  • Data pushing method and system

    CN107835246A

  • Dynamic priority-based load balancing strategy in cloud environment

    CN107948330A

  • Micro-service dynamic adaptive client load balancing method and system

    CN117155942A

  • Method and system for cluster dynamic balance expansion

    CN117950858A

  • Cell load balancing method, cell load measuring method, and devices thereof

    US20110171952A1

Cited By

  • Key management system load balancing method based on AI

    CN119166370A

  • A load balancing method for key management system based on AI

    CN119166370B