A network load balancing method and system for cloud servers

Abstract

The present invention relates to the field of network load technology, specifically to a network load balancing method and system for a cloud server, comprising the following steps: based on the operation record of the cloud server device, obtaining the network bandwidth load record of the cloud server in a target period, analyzing the change trend of the network bandwidth load of the cloud server, predicting the network bandwidth load of the cloud server in the future period, and obtaining the network load prediction result. The present invention analyzes the change trend of the network bandwidth load, predicts the network bandwidth load in the target period, and allocates the network bandwidth by analyzing and evaluating the network stability, thereby improving the adaptability and flexibility of the network, monitoring and adjusting the network bandwidth allocation, compensating for bandwidth shortage, ensuring the continuity and stability of network services, optimizing data processing efficiency, reducing resource waste, ensuring that the performance of the network is not affected under high load conditions, and improving user experience and service quality at the same time.

Description

Technical Field

[0001] The present invention relates to the field of network load technology, and in particular to a network load balancing method and system for a cloud server. Background Art

[0002] The field of network load technology involves a variety of methods and technologies aimed at optimizing data transmission and processing capabilities to improve the overall efficiency and performance of network systems. Key components of the network load field include the monitoring, management and distribution of data traffic, which involves the application of algorithms and hardware to ensure the effective routing and processing of data packets. Network load technology not only focuses on improving network processing speed and reducing latency, but also focuses on enhancing network reliability and scalability, and dynamically allocating resources to cope with different network load conditions, thereby ensuring the continuity and quality of network services.

[0003] Among them, the network load balancing method of cloud servers refers to the technology of effectively distributing network traffic among multiple network components. The core of the method is to automatically adjust the path and load of network traffic through algorithms to optimize resource utilization, avoid overloading of any single node, and ensure stable operation of the network. The network load balancing method of cloud servers is widely used in large websites, cloud computing services, and enterprise data centers. Through the network load balancing processing of cloud servers, it can ensure that the cloud server group remains stable and efficient when processing a large number of concurrent requests.

[0004] Traditional balancing methods have obvious limitations when dealing with unconventional or sudden network loads. For example, they fail to make full use of historical data and real-time network conditions to predict future network loads, resulting in the inability to effectively foresee and respond to changes in network loads in actual applications. Traditional methods use static or semi-static strategies for bandwidth allocation, which are difficult to adapt to rapid changes in network conditions, resulting in network congestion during peak hours, affecting overall network performance and user service experience. The evaluation of network stability is also mostly periodic or post-processing, lacking the ability to monitor in real time and dynamically adjust, affecting network operational efficiency and service reliability. This limits the overall performance of the network, especially in data centers and large-scale enterprise applications, where network instability or performance degradation can lead to significant economic losses and reduced user satisfaction. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a network load balancing method and system for a cloud server.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a network load balancing method for a cloud server, comprising the following steps:

[0007] S1: Based on the operation records of the cloud server equipment, obtain the network bandwidth load records of the cloud server in the target period, analyze the change trend of the network bandwidth load of the cloud server, predict the network bandwidth load of the cloud server in the future period, and obtain the network load prediction result;

[0008] S2: Based on the network status record, obtain the delay and packet loss records within a period of time, analyze the network fluctuation degree, evaluate the network stability, and obtain the network stability evaluation result;

[0009] S3: Based on the network load prediction result and the network stability evaluation result, combined with the preset cloud server priority, in the target period, the network bandwidth allocation of the differentiated cloud servers is adjusted to match the predicted network bandwidth load, and a network bandwidth balanced allocation result is obtained;

[0010] S4: Based on the network bandwidth balanced allocation result, monitor the network status of the cloud server, determine whether the cloud server has a bandwidth bottleneck, evaluate the degree of bandwidth shortage, calculate the bandwidth compensation amount, adjust the network bandwidth allocation, and obtain the network bandwidth compensation result.

[0011] The present invention is improved in that the method for analyzing the change trend of the network bandwidth load of the cloud server is:

[0012] S111: Based on the operation record of the cloud server device, obtain the network bandwidth load record of the cloud server in the target period to obtain network bandwidth load record data;

[0013] S112: Based on the network bandwidth load record data, the formula:

[0014]

[0015] Calculate bandwidth load change rate ,in, is the total number of data points, and is the weight parameter, and They are Tianhedi Bandwidth load data for the target period of the day, is the bandwidth load change rate;

[0016] S113: Based on the bandwidth load change rate ,according to The size and positive and negative values ​​are used to evaluate the changing trend of the network bandwidth load and obtain the changing trend analysis results of the network bandwidth load.

[0017] The present invention is improved in that the steps of obtaining the network load prediction result are:

[0018] S121: Based on the bandwidth load change rate , collect the latest bandwidth load data in the target period, and according to the prediction target, obtain the time length between the target prediction time and the latest bandwidth load data to obtain the prediction correlation data;

[0019] S122: Based on the predicted associated data, by formula:

[0020]

[0021] Calculate predicted network bandwidth load , and obtain the network load prediction result, where , and is the weight coefficient, The latest bandwidth load data for the target period. is the time length from the latest data point to the predicted time point, is the bandwidth load change rate, and are the maximum and minimum values ​​of bandwidth load data during the observation period, is the predicted network bandwidth load.

[0022] The present invention is improved in that the steps of obtaining the network stability evaluation result are:

[0023] S211: Based on the network status record, collect network status data for a period of time, where the network status data includes network delay and packet loss rate;

[0024] S212: Based on the network status data, by formula:

[0025]

[0026] Calculating the Network Stability Index ,in, is the average network delay, is the average packet loss rate of the network, and is the weight coefficient, is the stability constant, is the magnification factor, is the network stability index;

[0027] S213: Based on the network stability index ,according to The value is compared with the preset standard to evaluate the stability of the network and obtain the network stability evaluation result.

[0028] The present invention is improved in that the steps of obtaining the network bandwidth balanced allocation result are:

[0029] S311: Based on the network load prediction result, by formula:

[0030]

[0031] Calculate the ratio of the predicted bandwidth load of the cloud server to the total load ,in, For the The network bandwidth load predicted by each cloud server, is the total predicted load, For the The ratio of the predicted bandwidth load of each cloud server to the total load;

[0032] S312: Based on the ratio of the predicted bandwidth load of the cloud server to the total load And the network stability evaluation results, combined with the preset cloud server priority, through the formula:

[0033] Calculate the amount of bandwidth allocated to the cloud server ,in, To allocate to The bandwidth of each cloud server, For the The preset priority of each cloud server, is the scaling factor, For the The ratio of the predicted bandwidth load of each cloud server to the total load, is the total bandwidth available for allocation, is the network stability index, is the weight factor, is the adjustment factor;

[0034] S313: Based on the bandwidth allocated to the cloud server , bandwidth allocation is implemented for differentiated cloud servers, and the result of balanced network bandwidth allocation is obtained.

[0035] The present invention is improved in that the method for evaluating the degree of bandwidth shortage is:

[0036] S411: Collect the current bandwidth allocation of each cloud server , analyze the applications and services running on the cloud server, obtain the maximum total amount of data expected to be transmitted per unit time by the applications and services, and obtain operation-related data;

[0037] S412: Based on the operation-related data, by formula:

[0038]

[0039] Calculate the ideal bandwidth requirements of cloud servers ,in, is the maximum total amount of data that the service is expected to transmit per unit time, is the number of concurrently connected users, is the service quality adjustment factor, The bandwidth requirement of the cloud server under ideal conditions;

[0040] S413: Based on the bandwidth requirement of the cloud server in an ideal state And run the associated data, through the formula:

[0041]

[0042] Calculating bandwidth shortage , get the bandwidth missing information, where, is the current bandwidth allocation of the cloud server, For the degree of bandwidth shortage, This is the ideal bandwidth requirement for a cloud server.

[0043] The present invention is improved in that the steps of obtaining the network bandwidth compensation result are:

[0044] S421: Based on the network bandwidth balanced allocation result, collect the idle bandwidth available in the network to obtain idle bandwidth data;

[0045] S422: Based on the idle bandwidth data and the degree of bandwidth shortage , through the formula:

[0046]

[0047] Calculate bandwidth compensation for cloud servers ,in, is the amount of free bandwidth available in the network, For the The degree of bandwidth shortage of each cloud server, is the sum of the cloud server bandwidth shortage levels, For the The bandwidth compensation amount of each cloud server, is the scaling factor;

[0048] S423: Based on the bandwidth compensation amount of the cloud server , adjust the network bandwidth allocation and obtain the network bandwidth compensation result.

[0049] A network load balancing system for a cloud server, the network load balancing system for the cloud server is used to execute the network load balancing method for the cloud server, the system comprising:

[0050] The load trend prediction module obtains the network bandwidth load record of the cloud server in the target period based on the operation record of the cloud server equipment, analyzes the change trend of the network bandwidth load of the cloud server, and obtains the change trend analysis result of the network bandwidth load;

[0051] The load prediction module predicts the cloud server network bandwidth load in the future period based on the change trend analysis result of the network bandwidth load to obtain a network load prediction result;

[0052] The network status assessment module obtains the delay and packet loss records within a period of time based on the network status records, analyzes the network fluctuation degree, evaluates the network stability, and obtains the network stability assessment results;

[0053] The network bandwidth allocation module adjusts the network bandwidth allocation of the differentiated cloud servers based on the network load prediction result and the network stability evaluation result, combined with the preset cloud server priority, to match the predicted network bandwidth load, and obtain a network bandwidth balanced allocation result;

[0054] The network bandwidth compensation module monitors the network status of the cloud server based on the network bandwidth balanced allocation result, determines whether the cloud server has a bandwidth bottleneck, evaluates the degree of bandwidth shortage, calculates the bandwidth compensation amount, adjusts the network bandwidth allocation, and obtains the network bandwidth compensation result.

[0055] Compared with the prior art, the advantages and positive effects of the present invention are:

[0056] In the present invention, by analyzing the changing trend of network bandwidth load, predicting the network bandwidth load in the target time period, and by analyzing and evaluating the network stability, the network bandwidth is allocated, thereby improving the adaptability and flexibility of the network, monitoring and adjusting the network bandwidth allocation, compensating for bandwidth shortage, ensuring the continuity and stability of network services, optimizing data processing efficiency, reducing resource waste, ensuring that the performance of the network is not affected under high load conditions, and improving user experience and service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a flow chart of the method of the present invention;

[0058] Figure 2 A flow chart of analyzing the network bandwidth load change trend of a cloud server according to the present invention;

[0059] Figure 3 A flowchart of obtaining network load prediction results according to the present invention;

[0060] Figure 4 A flow chart of obtaining network stability evaluation results according to the present invention;

[0061] Figure 5 A flow chart of obtaining the network bandwidth balanced allocation result of the present invention;

[0062] Figure 6 A flow chart for evaluating the degree of bandwidth deficiency for the present invention;

[0063] Figure 7 The present invention is a flow chart of obtaining network bandwidth compensation results. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0065] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0066] See also Figure 1 The present invention provides a technical solution: a network load balancing method for a cloud server, comprising the following steps:

[0067] S1: Based on the operation records of the cloud server equipment, obtain the network bandwidth load records of the cloud server in the target period, analyze the change trend of the network bandwidth load of the cloud server, predict the network bandwidth load of the cloud server in the future period, and obtain the network load prediction result;

[0068] S2: Based on the network status record, obtain the delay and packet loss records within a period of time, analyze the network fluctuation degree, evaluate the network stability, and obtain the network stability evaluation result;

[0069] S3: Based on the network load prediction results and network stability assessment results, combined with the preset cloud server priorities, the network bandwidth allocation of differentiated cloud servers is adjusted during the target period to match the predicted network bandwidth load, and the network bandwidth balanced allocation result is obtained;

[0070] S4: Based on the network bandwidth balanced allocation result, monitor the network status of the cloud server, determine whether there is a bandwidth bottleneck in the cloud server, evaluate the degree of bandwidth shortage, calculate the bandwidth compensation amount, adjust the network bandwidth allocation, and obtain the network bandwidth compensation result.

[0071] The network load prediction results include the predicted average load, prediction error range and confidence interval; the network stability assessment results include the delay time standard deviation, average delay time and packet loss rate; the network bandwidth balancing allocation results include the adjusted bandwidth quota and corresponding traffic type allocation information; the network bandwidth compensation results include the identified bandwidth bottleneck nodes, the calculated bandwidth compensation amount and the compensated cloud server bandwidth configuration.

[0072] See also Figure 2 ,The method to analyze the network bandwidth load change trend of the cloud server is:

[0073] S111: Based on the operation record of the cloud server device, obtain the network bandwidth load record of the cloud server in the target period to obtain network bandwidth load record data;

[0074] S112: Record data based on network bandwidth load, using the formula:

[0075]

[0076] Calculate bandwidth load change rate ,in, is the total number of data points, and is the weight parameter, and They are Tianhedi Bandwidth load data for the target period of the day, is the bandwidth load change rate;

[0077] S113: Based on bandwidth load change rate ,according to The size and positive and negative values ​​are used to evaluate the changing trend of the network bandwidth load and obtain the changing trend analysis results of the network bandwidth load.

[0078] formula:

[0079]

[0080] Parameter details and how to obtain them:

[0081] : The parameter represents the total number of data points in the dataset, that is, the number of observation days. and :The parameters represent Tianhedi Network bandwidth data for a target period of the day. The data is automatically collected by the network monitoring system during a fixed period of time and then stored in the database. and :The weight parameter is used to adjust the influence of bandwidth data on different days to improve the accuracy of trend prediction. The parameter is determined to its optimal value through regression analysis or optimization algorithm to ensure the consistency of trend prediction with actual situation.

[0082] Calculation example:

[0083] Set up a 3-day bandwidth dataset , .

[0084] Setting weight parameters and .

[0085] calculate :

[0086]

[0087] Calculated It means that after considering the weight adjustment, the average daily network bandwidth load change rate for 3 days is Unit. It reflects that the network bandwidth of cloud servers showed a slight increase trend overall on the days considered. The results can help network administrators better understand the changes in network bandwidth demand in the short term and make appropriate network resource adjustments and planning.

[0088] See also Figure 3 , the steps to obtain the network load prediction results are:

[0089] S121: Based on bandwidth load change rate , collect the latest bandwidth load data in the target period, and according to the prediction target, obtain the time length between the target prediction time and the latest bandwidth load data to obtain the prediction correlation data;

[0090] S122: Based on the predicted associated data, the formula:

[0091]

[0092] Calculate predicted network bandwidth load , and obtain the network load prediction result, where , and is the weight coefficient, The latest bandwidth load data for the target period. is the time length from the latest data point to the predicted time point, is the bandwidth load change rate, and are the maximum and minimum values ​​of bandwidth load data during the observation period, is the predicted network bandwidth load.

[0093] formula:

[0094]

[0095] Parameter details and how to obtain them:

[0096] , and : Weight coefficients are set to adjust the importance of different variables on the prediction model. Parameters are usually obtained through optimization algorithms or regression analysis to ensure the accuracy and adaptability of the prediction model.

[0097] : Indicates the latest bandwidth load data during the target period, which is directly obtained from the network monitoring system to ensure the timeliness and accuracy of the data.

[0098] : It is the length of time from the latest data point to the prediction time point, which is manually set according to the business cycle that needs to be predicted.

[0099] : Bandwidth load change rate, calculated using the previous method.

[0100] and : are the maximum and minimum values ​​of the bandwidth load data during the observation period. The values ​​are obtained by finding the maximum and minimum records in the bandwidth dataset and are used to standardize the predicted growth rate.

[0101] Calculation example:

[0102] The parameters are set as follows:

[0103] , 0Mbps / day, sky, Mbps, , , , Mbps.

[0104] Calculation process:

[0105] Compute the logarithmic part:

[0106]

[0107] Calculating predicted bandwidth :

[0108]

[0109] Calculated prediction results Mbps, indicating that during the target period, the average network bandwidth load of the cloud server is expected to reach approximately Mbps. The forecast helps network administrators understand the possible trend of future network load and make decisions on the allocation of network resources and bandwidth management accordingly.

[0110] See also Figure 4 , the steps to obtain the network stability evaluation results are:

[0111] S211: Based on the network status record, collect network status data for a period of time, where the network status data includes network delay and packet loss rate;

[0112] S212: Based on the network status data, by formula:

[0113]

[0114] Calculating the Network Stability Index ,in, is the average network delay, is the average packet loss rate of the network, and is the weight coefficient, is the stability constant, is the magnification factor, is the network stability index;

[0115] S213: Based on the Network Stability Index ,according to The value is compared with the preset standard to evaluate the stability of the network and obtain the network stability evaluation result.

[0116] formula:

[0117]

[0118] Parameter details and how to obtain them:

[0119] : Indicates the average latency of the network, which is obtained by summing up the data of network response time measured within a specified period of time. The calculation method is to add up all the measured latency times and divide them by the number of measurements.

[0120] : Represents the average packet loss rate of the network. The parameter is calculated by recording the proportion of packet losses in the same time period, by dividing the total number of lost packets by the total number of sent packets.

[0121] and : Weight coefficient, used to adjust the influence of latency and packet loss rate when calculating the network stability index. The coefficient is set based on expert presets or network performance optimization needs.

[0122] : Stability constant, a constant used to ensure that the denominator is not zero and adjust the evaluation sensitivity. The parameters are set according to the network performance standard to ensure the applicability of the formula and the stability of the evaluation results.

[0123] : Amplification factor, used to amplify or adjust the final network stability score to better suit actual monitoring and evaluation needs. The choice of parameters depends on the expected evaluation scope and sensitivity requirements.

[0124] Calculation example:

[0125] Set the following data:

[0126] Average latency Milliseconds, average packet loss rate , weight , , , .

[0127] The calculation process is as follows:

[0128]

[0129] The network stability index obtained , indicating that after taking into account the impact of delay and packet loss rate, the stability of the network is in a good state. The higher the index, the more stable the network.

[0130] See also Figure 5 , the steps to obtain the network bandwidth balanced allocation result are:

[0131] S311: Based on the network load prediction result, the formula is:

[0132]

[0133] Calculate the ratio of the predicted bandwidth load of the cloud server to the total load ,in, For the The network bandwidth load predicted by each cloud server, is the total predicted load, For the The ratio of the predicted bandwidth load of each cloud server to the total load;

[0134] S312: The ratio of the predicted bandwidth load of the cloud server to the total load And the network stability evaluation results, combined with the preset cloud server priority, through the formula:

[0135]

[0136] Calculate the amount of bandwidth allocated to the cloud server ,in, To allocate to The bandwidth of each cloud server, For the The preset priority of each cloud server, is the scaling factor, For the The ratio of the predicted bandwidth load of each cloud server to the total load, is the total bandwidth available for allocation, is the network stability index, is the weight factor, is the adjustment factor;

[0137] S313: Based on the bandwidth allocated to the cloud server , bandwidth allocation is implemented for differentiated cloud servers, and the result of balanced network bandwidth allocation is obtained.

[0138] formula:

[0139]

[0140] Parameter details and how to obtain them:

[0141] :Indicates the The preset priority of a cloud server. The priority is usually set according to the importance of the cloud server, business needs, etc. The preset priority can be determined by historical data analysis, policy decision-making, etc. Its numerical range should be reasonably set under actual operating conditions to ensure that the priority has practical significance for network resource allocation.

[0142] : Scaling factor, which is used to adjust the impact of the priority logarithm in bandwidth allocation so as to maintain reasonable bandwidth allocation under larger or smaller priority differences. It can be determined through bandwidth usage, performance indicator analysis, etc.

[0143] : The total predicted load is obtained by summing the predicted network bandwidth load of each cloud server.

[0144] :The total amount of bandwidth available for allocation. The parameter is usually set by the total bandwidth capacity or network traffic control policy in the network bandwidth management system. It can be set by the network administrator based on the maximum carrying capacity of the equipment and network services.

[0145] : Network stability index, which indicates the overall stability of the network, is calculated by the previous steps.

[0146] : Weight factor, used to adjust the network stability index The strength of the impact in bandwidth allocation is usually set according to the network operation strategy. A higher value means that network stability has a greater weight in bandwidth allocation.

[0147] : The coefficient used to adjust the bandwidth allocation ratio in the numerator. The coefficient is usually used to enlarge or reduce the bandwidth allocation to meet different application needs. The specific value depends on the dynamic allocation requirements of network resources.

[0148] Calculation example:

[0149] Set the following parameters:

[0150] Network bandwidth load Mbps, Mbps, total predicted load Mbps, scaling factor , Network Stability Index , the total bandwidth that can be allocated by the system Mbps, preset priority , , weight factor , adjustment coefficient ;

[0151] Calculation example:

[0152] Calculate the load share of each cloud server:

[0153]

[0154] Calculate the logarithmic form of the priority:

[0155] For the first cloud server :

[0156]

[0157] For the second cloud server :

[0158]

[0159] Calculate the bandwidth allocation for each cloud server:

[0160] For the first cloud server :

[0161]

[0162] For the second cloud server :

[0163]

[0164] Calculation results show that the first cloud server will be allocated approximately 370Mbps bandwidth, and the second cloud server will be allocated approximately 196Mbps bandwidth. The allocation method is dynamically adjusted based on the priority of each cloud server, the predicted load share, and the overall network stability to ensure the rational use of bandwidth resources and the optimization of network performance.

[0165] See also Figure 6 , the method to evaluate the degree of bandwidth shortage is:

[0166] S411: Collect the current bandwidth allocation of each cloud server , analyze the applications and services running on the cloud server, obtain the maximum total amount of data expected to be transmitted per unit time by the applications and services, and obtain operation-related data;

[0167] S412: Based on the running correlation data, by formula:

[0168]

[0169] Calculate the ideal bandwidth requirements of cloud servers ,in, is the maximum total amount of data that the service is expected to transmit per unit time, is the number of concurrently connected users, is the service quality adjustment factor, The bandwidth requirement of the cloud server under ideal conditions;

[0170] S413: Bandwidth requirements based on ideal cloud server conditions And run the associated data, through the formula:

[0171]

[0172] Calculating bandwidth shortage , get the bandwidth missing information, where, is the current bandwidth allocation of the cloud server, For the degree of bandwidth shortage, This is the ideal bandwidth requirement for a cloud server.

[0173] formula:

[0174]

[0175] and

[0176]

[0177] Parameter details and how to obtain them:

[0178] : Indicates the current bandwidth allocation of the cloud server, which is obtained through the previous method.

[0179] : The total amount of data that is expected to be transmitted by the service in a unit of time. The parameter is estimated based on the types of applications and services running on the cloud server. For example, if the cloud server is mainly used for video streaming services, the estimate is based on the video bit rate and the number of users expected to be served.

[0180] : The number of concurrently connected users. The value is obtained based on the service's concurrent online user records, usually provided by the cloud server log or user session management system.

[0181] : Service quality adjustment factor. It is a coefficient manually set according to the bandwidth sensitivity of the service. For example, a real-time video service may be set to a higher value to ensure smoothness.

[0182] Calculation Example

[0183] Set the following parameters:

[0184] Mbps: The current bandwidth allocation of the cloud server.

[0185] Mbps: Maximum data transfer rate expected per user, suitable for high-definition video streaming.

[0186] : Number of concurrent users, assuming that the number of users connected to the cloud server at the same time is 6.

[0187] : Quality of service adjustment factor, required for a given high-quality video streaming service.

[0188] Calculating ideal bandwidth requirements for cloud servers :

[0189]

[0190] Calculating bandwidth shortage :

[0191]

[0192] Calculation results , indicating that the actual bandwidth usage of the current cloud server is less than the ideal demand. Mbps bandwidth requires a considerable amount of additional bandwidth resources.

[0193] See also Figure 7 , the steps to obtain the network bandwidth compensation result are:

[0194] S421: Based on the network bandwidth balanced allocation result, collect the idle bandwidth available in the network to obtain idle bandwidth data;

[0195] S422: Based on idle bandwidth data and bandwidth shortage degree , through the formula:

[0196]

[0197] Calculate bandwidth compensation for cloud servers ,in, is the amount of free bandwidth available in the network, For the The degree of bandwidth shortage of each cloud server, is the sum of the cloud server bandwidth shortage levels, For the The bandwidth compensation amount of each cloud server, is the scaling factor;

[0198] S423: Bandwidth compensation based on cloud server , adjust the network bandwidth allocation and obtain the network bandwidth compensation result.

[0199] formula:

[0200]

[0201] Parameter details and how to obtain them:

[0202] : is the amount of idle bandwidth available in the network. The parameter is monitored and updated in real time by the network management system, reflecting the bandwidth resources that are not being used in the network at any time.

[0203] : Indicates the bandwidth shortage of the cloud server, which is calculated by the previous method.

[0204] : The sum of the bandwidth shortage of all cloud servers. The value is accumulated.

[0205] : Scaling factor, used to adjust the effect of the square root term in the compensation formula. The factor is set based on the needs of network performance optimization to adjust the sensitivity of the compensation amount to small bandwidth loss values.

[0206] Calculation example:

[0207] Set the following parameters:

[0208] 40Mbps: The network has Mbps of free bandwidth is available.

[0209] Mbps: Cloud Server j currently lacks 30 Mbps of bandwidth.

[0210] Mbps: The total bandwidth shortage of all cloud servers in the network is Mbps.

[0211] : Set the zoom factor to 0.5.

[0212] Calculate the bandwidth compensation of cloud server j as follows:

[0213]

[0214] Calculation results , indicating that in order to compensate for the lack of bandwidth of cloud server j, approximately Mbps is given to the cloud server, and the compensation amount takes into account the proportion of the cloud server in the entire network bandwidth shortage, ensuring the fairness and efficiency of bandwidth compensation.

[0215] A network load balancing system for a cloud server, the network load balancing system for a cloud server is used to execute the network load balancing method for the cloud server, and the system comprises:

[0216] The load trend prediction module obtains the network bandwidth load record of the cloud server in the target period based on the operation record of the cloud server equipment, analyzes the change trend of the network bandwidth load of the cloud server, and obtains the change trend analysis result of the network bandwidth load;

[0217] The load prediction module predicts the cloud server network bandwidth load in the future period based on the change trend analysis results of the network bandwidth load, and obtains the network load prediction result;

[0218] The network status assessment module obtains the delay and packet loss records within a period of time based on the network status records, analyzes the network fluctuation degree, evaluates the network stability, and obtains the network stability assessment results;

[0219] The network bandwidth allocation module adjusts the network bandwidth allocation of differentiated cloud servers based on the network load prediction results and network stability assessment results, combined with the preset cloud server priority, to match the predicted network bandwidth load and obtain the network bandwidth balanced allocation result;

[0220] The network bandwidth compensation module monitors the network status of the cloud server based on the network bandwidth balanced allocation result, determines whether the cloud server has a bandwidth bottleneck, evaluates the degree of bandwidth shortage, calculates the bandwidth compensation amount, adjusts the network bandwidth allocation, and obtains the network bandwidth compensation result.

[0221] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A network load balancing method for a cloud server, characterized in that: The following steps are involved: Based on the operation records of the cloud server equipment, obtain the network bandwidth load records of the cloud server in the target period, analyze the change trend of the network bandwidth load of the cloud server, predict the network bandwidth load of the cloud server in the future period, and obtain the network load prediction result; Based on the network status records, obtain the delay and packet loss records within a period of time, analyze the network fluctuation degree, evaluate the network stability, and obtain the network stability evaluation results; Based on the network load prediction result and the network stability evaluation result, combined with the preset cloud server priority, in the target period, the network bandwidth allocation of the differentiated cloud servers is adjusted to match the predicted network bandwidth load, so as to obtain a network bandwidth balanced allocation result; Based on the network bandwidth balanced allocation result, the network status of the cloud server is monitored, whether the cloud server has a bandwidth bottleneck is determined, the degree of bandwidth shortage is evaluated, the bandwidth compensation amount is calculated, the network bandwidth allocation is adjusted, and the network bandwidth compensation result is obtained.

2. The network load balancing method of the cloud server according to claim 1, characterized in that: The method for analyzing the network bandwidth load change trend of the cloud server is: Based on the operation record of the cloud server device, obtain the network bandwidth load record of the cloud server in the target period, and obtain the network bandwidth load record data; Based on the network bandwidth load record data, through the formula: Calculate bandwidth load change rate ,in, is the total number of data points, and is the weight parameter, and They are Tianhedi Bandwidth load data for the target period of the day, is the bandwidth load change rate; Based on the bandwidth load change rate ,according to The size and positive and negative values ​​are used to evaluate the changing trend of the network bandwidth load and obtain the changing trend analysis results of the network bandwidth load.

3. The network load balancing method of the cloud server according to claim 2, characterized in that: The steps for obtaining the network load prediction result are: Based on the bandwidth load change rate , collect the latest bandwidth load data in the target period, and according to the prediction target, obtain the time length between the target prediction time and the latest bandwidth load data to obtain the prediction related data; Based on the predicted associated data, by the formula: Calculate predicted network bandwidth load , and obtain the network load prediction result, where , and is the weight coefficient, The latest bandwidth load data for the target period. is the time length from the latest data point to the predicted time point, is the bandwidth load change rate, and are the maximum and minimum values ​​of bandwidth load data during the observation period, is the predicted network bandwidth load.

4. The network load balancing method of the cloud server according to claim 1, characterized in that: The steps for obtaining the network stability evaluation result are: Based on the network status record, collect network status data within a period of time, including network delay and packet loss rate; Based on the network status data, by the formula: Calculating the Network Stability Index ,in, is the average network delay, is the average packet loss rate of the network, and is the weight coefficient, is the stability constant, is the magnification factor, is the network stability index; Based on the network stability index ,according to The value is compared with the preset standard to evaluate the stability of the network and obtain the network stability evaluation result.

5. The network load balancing method of the cloud server according to claim 1, characterized in that: The steps for obtaining the network bandwidth balanced allocation result are: Based on the network load prediction result, the formula is: Calculate the ratio of the predicted bandwidth load of the cloud server to the total load ,in, For the The network bandwidth load predicted by each cloud server, is the total predicted load, For the The ratio of the predicted bandwidth load of each cloud server to the total load; Based on the proportion of the predicted bandwidth load of the cloud server to the total load And the network stability evaluation results, combined with the preset cloud server priority, through the formula: Calculate the amount of bandwidth allocated to the cloud server ,in, To allocate to The bandwidth of each cloud server, For the The preset priority of each cloud server, is the scaling factor, For the The ratio of the predicted bandwidth load of each cloud server to the total load, is the total bandwidth available for allocation, is the network stability index, is the weight factor, is the adjustment factor; Based on the amount of bandwidth allocated to the cloud server , bandwidth allocation is implemented for differentiated cloud servers, and the result of balanced network bandwidth allocation is obtained.

6. The network load balancing method of the cloud server according to claim 1, characterized in that: The method for evaluating the degree of bandwidth shortage is: Collect the current bandwidth allocation of each cloud server , analyze the applications and services running on the cloud server, obtain the maximum total amount of data expected to be transmitted per unit time by the applications and services, and obtain operation-related data; Based on the running correlation data, by the formula: Calculate the ideal bandwidth requirements of cloud servers ,in, is the maximum total amount of data that the service is expected to transmit per unit time, is the number of concurrently connected users, is the service quality adjustment factor, The bandwidth requirement of the cloud server under ideal conditions; Based on the bandwidth requirements of the cloud server under ideal conditions And run the associated data, through the formula: Calculating bandwidth shortage , get the bandwidth missing information, where, is the current bandwidth allocation of the cloud server, For the degree of bandwidth shortage, This is the ideal bandwidth requirement for a cloud server.

7. The network load balancing method of the cloud server according to claim 6, characterized in that: The steps for obtaining the network bandwidth compensation result are: Based on the network bandwidth balanced allocation result, the amount of idle bandwidth available in the network is collected to obtain idle bandwidth data; Based on the idle bandwidth data and the degree of bandwidth shortage , through the formula: Calculate bandwidth compensation for cloud servers ,in, is the amount of free bandwidth available in the network, For the The degree of bandwidth shortage of each cloud server, is the sum of the cloud server bandwidth shortage levels, For the The bandwidth compensation amount of each cloud server, is the scaling factor; Based on the bandwidth compensation amount of the cloud server , adjust the network bandwidth allocation and obtain the network bandwidth compensation result.

8. A network load balancing system for a cloud server, characterized in that: According to the network load balancing method of the cloud server according to any one of claims 1 to 7, the system comprises: The load trend prediction module obtains the network bandwidth load record of the cloud server in the target period based on the operation record of the cloud server equipment, analyzes the change trend of the network bandwidth load of the cloud server, and obtains the change trend analysis result of the network bandwidth load; The load prediction module predicts the cloud server network bandwidth load in the future period based on the change trend analysis result of the network bandwidth load to obtain a network load prediction result; The network status assessment module obtains the delay and packet loss records within a period of time based on the network status records, analyzes the network fluctuation degree, evaluates the network stability, and obtains the network stability assessment results; The network bandwidth allocation module adjusts the network bandwidth allocation of the differentiated cloud servers based on the network load prediction result and the network stability evaluation result, combined with the preset cloud server priority, to match the predicted network bandwidth load, and obtain a network bandwidth balanced allocation result; The network bandwidth compensation module monitors the network status of the cloud server based on the network bandwidth balanced allocation result, determines whether the cloud server has a bandwidth bottleneck, evaluates the degree of bandwidth shortage, calculates the bandwidth compensation amount, adjusts the network bandwidth allocation, and obtains the network bandwidth compensation result.

Images