A network resource allocation method and system based on network load

Abstract

The application discloses a network resource deployment method and system based on network load, relates to the technical field of network resource deployment, and comprises the following steps: step 1, determining service demand satisfaction; step 2, calculating network load balancing degree; step 3, predicting future service demand data; step 4, defining constraint conditions and maximizing service demand satisfaction and network load balancing degree, and formulating a network resource deployment strategy. The application can effectively improve the prior art, which, when evaluating service demand, often only focuses on the overall average level and ignores the situation that individual key indicators are seriously out of standard.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of network resource allocation, and particularly relates to a network resource allocation method and system based on network load. BACKGROUND

[0002] In today's digital era, network services are increasingly diverse, covering core critical services such as financial transaction systems and medical data transmission, secondary important services such as enterprise daily office applications, and general services such as ordinary web browsing. Different services have different demands for network resources and different sensitivities to network performance. The degree of satisfaction of service demand is related to the quality of service operation, which involves the consideration of resource requests and actual allocation, as well as a number of quantitative indicators such as delay time and packet loss rate, which reflect the actual performance of the network when carrying services. The degree of network load balancing is a key indicator of whether the load distribution of each node in the network is reasonable, and is closely related to load data such as CPU usage, memory occupancy and bandwidth utilization. Accurate prediction of future service demand data can help the network to plan resources in advance and improve resource utilization efficiency.

[0003] However, the prior art has many deficiencies in network resource allocation. When evaluating whether the resource allocation meets the service demand, the prior art may mainly focus on whether the total amount of resource allocation meets the service request, such as only looking at whether enough bandwidth resources are allocated. If the amount of bandwidth allocation meets the bandwidth value of the service request, it is considered that the resource allocation meets the demand. However, there may be problems such as too long delay time and high packet loss rate. For example, in a video conference service, although the allocated bandwidth can ensure that the video stream is basically not stuck, the high delay will cause the audio and video to be out of sync, seriously affecting the user experience, and this key indicator not meeting the condition will be covered up by the better indicator of sufficient bandwidth allocation. SUMMARY

[0004] The present application aims to provide a network resource allocation method and system based on network load to solve the problems in the prior art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme: a network resource allocation method based on network load, the method comprising the following steps:

[0006] Step 1, determining the degree of satisfaction of service demand;

[0007] Step 2, calculating the degree of network load balancing;

[0008] Step 3, predicting future service demand data;

[0009] Step 4, define constraints and maximize service demand satisfaction and network load balancing, formulate network resource allocation strategy.

[0010] In step 1, the specific steps to determine the service demand satisfaction are as follows:

[0011] Step 1.1, define different types of services and their corresponding priorities;

[0012] Service types include but are not limited to core critical services, secondary important services and general services;

[0013] Priority includes but is not limited to the highest priority, high priority, medium priority and low priority;

[0014] Step 1.2, collect resource request data and actual allocation of different service types;

[0015] Resource request data includes but is not limited to required bandwidth, computing resources and storage resources;

[0016] Step 1.3, select quantitative indicators and establish evaluation criteria according to different service types;

[0017] Quantitative indicators include but are not limited to delay time, packet loss rate, average processing time and resource utilization;

[0018] Set different quantitative indicator conditions for different service types based on different priorities;

[0019] Step 1.4, compare the resource request data and actual allocation of each service type, and calculate the service demand satisfaction according to the corresponding quantitative indicator conditions in the evaluation criteria;

[0020] For a certain actual allocation or quantitative indicator of a certain service type:

[0021] When the actual allocation Re meets the resource request, set the satisfaction degree to 1;

[0022] When the actual allocation Re does not meet the resource request condition threshold Ru, set the satisfaction degree to 1-|Re-Ru| / Ru; wherein |Re-Ru| represents the absolute value of the difference between the actual allocation Re and the resource request condition threshold Ru;

[0023] When the quantitative indicator T meets the quantitative indicator condition, set the satisfaction degree to 1;

[0024] When the quantitative indicator T does not meet the quantitative indicator condition threshold Th, set the satisfaction degree to 1-|T-Th| / Th; wherein |T-Th| represents the absolute value of the difference between the quantitative indicator T and the quantitative indicator threshold Th;

[0025] Based on the short board effect, the weight allocation method is as follows:

[0026] The satisfaction set is represented as S: S={s1, s2,...,sn}; where n is a positive integer, representing the number of quantitative indicators and requested resources; s1~sn respectively represent the satisfaction of the first to n quantitative indicators or requested resources; s1, s2,...,sn∈[0,1];

[0027] The corresponding weight set W: W={w1, w2,...,wn}; where w1~wn respectively represent the weights corresponding to s1~sn; w1, w2,...,wn≥0;

[0028] Exponential function is used to amplify the influence of low value items: wi=e -si / (e -s1 +e -s2 +…+e -sn ); where i∈{1,2,…,n}, si represents the sequence of quantitative indicators or requested resource satisfaction, wi represents the weight corresponding to si;

[0029] The total satisfaction of the service: St=s1×w1+s2×w2+…+sn×wn;

[0030] Calculate the total satisfaction of each service, and calculate the average of the total satisfaction of all services as the total satisfaction of this scheduling.

[0031] In step 2, the specific steps for calculating the network load balancing degree are as follows:

[0032] Step 2.1, deploy monitoring tools in each node in the network to collect node load data in real time;

[0033] Load data includes but is not limited to CPU usage, memory occupancy and bandwidth utilization;

[0034] Step 2.2, set the best load value for different types of load data;

[0035] Step 2.3, construct the probability density function of Beta distribution for the corresponding different types of load data according to the different best load values, and calculate the load balancing degree;

[0036] For a type of load data, its best load value is represented as A, which means that the load balancing degree of this type of load data is the highest at A; 0

[0037] Construct the probability density function f(x,α,β) of Beta distribution: combine the best load value A of this type of load data and the mean μ or variance σ 2 of the load data in a period of time to determine α and β; where x represents the actual value of the load data, and α and β are the parameters of the probability density function of Beta distribution; α>1 and β>1;

[0038] Take the probability density function value of the Beta distribution corresponding to the type of actual load data as the load balancing degree value of the type of load data;

[0039] For a node, calculate the average value of the sum of the load balancing degree values of all types of load data contained in the node as the load balancing degree value of the node;

[0040] Step 2.4, calculate the total network load balancing degree;

[0041] Take the average value of the sum of the load balancing degree values of each node as the total network load balancing degree.

[0042] In step 3, the specific steps of predicting future business demand data are:

[0043] Step 3.1, collect historical business demand data;

[0044] Collect past business demand data from business system logs, databases and other data sources, including business traffic, transaction volume, user access volume and other specific indicators; record the corresponding timestamp of the data to ensure that the data has the characteristics of time series;

[0045] Step 3.2, build and train LSTM time series analysis model;

[0046] Divide the data into training set and test set;

[0047] Build LSTM model structure: staff determine LSTM model hyperparameters according to the complexity of data and the difficulty of prediction task; The number of neurons in each layer can be adjusted through experiments to find the optimal model structure; At the same time, add input layer and output layer to convert input data into a format suitable for LSTM processing and output prediction results;

[0048] Input the training set data into the built LSTM model, use the optimization algorithm Adam and the mean square error loss function for training; In the training process, the parameters of the model are constantly adjusted to minimize the value of the loss function, so as to improve the fitting ability of the model to historical data; The training process can be iterated for multiple rounds until the performance of the model no longer improves significantly;

[0049] Step 3.3, use the trained model for prediction;

[0050] Input the test set data into the trained LSTM model, and the model will output the predicted value of future business demand according to the learned rules and trends.

[0051] In step 4, define the constraints and maximize the satisfaction of business requirements and the degree of network load balancing. The specific implementation steps for formulating the network resource allocation strategy are as follows:

[0052] Step 4.1: Define the optimization goal.

[0053] Assign different weights to the business demand satisfaction and network load balancing degree, and perform weighted summation to form a new optimization goal;

[0054] Step 4.2: Define constraints.

[0055] Business resource allocation constraints:

[0056] Total resource constraint: The total amount of resources actually allocated to all businesses must not exceed the total available amount of the resource;

[0057] Network load balancing constraints:

[0058] Load adjustment strategy: When the load of a node is greater than the optimal load value, reduce the tasks assigned to it and migrate some tasks to nodes with lower load; if the load of a node is too low, increase the tasks assigned to it;

[0059] Step 4.3: Use the constrained optimization algorithm to solve the specified network resource allocation strategy;

[0060] Use multi-objective optimization algorithms and penalty function methods for iterative optimization, continuously adjust demand distribution and load to maximize the optimization goal; formulate network resource allocation strategies based on the maximized optimization goal.

[0061] A network resource allocation system based on network load, the system includes a business demand satisfaction calculation module, a network load balancing evaluation module, a future demand prediction module and a resource allocation strategy formulation module;

[0062] The business requirement satisfaction calculation module is used to determine the business requirement satisfaction;

[0063] The network load balancing evaluation module is used to calculate the degree of network load balancing;

[0064] The future demand forecasting module is used to forecast future business demand data;

[0065] The resource allocation strategy formulation module is used to define constraints and maximize business demand satisfaction and network load balancing, and formulate network resource allocation strategies;

[0066] The output end of the service demand satisfaction calculation module is connected with the input end of the resource allocation strategy formulation module; the output end of the network load balancing evaluation module is connected with the input end of the resource allocation strategy formulation module; and the output end of the future demand prediction module is connected with the input end of the resource allocation strategy formulation module.

[0067] The service demand satisfaction calculation module comprises a service classification and priority definition unit, a resource request and allocation data collection unit, and a satisfaction calculation and evaluation unit.

[0068] The service classification and priority definition unit is used for defining service types and their priorities to form classification standards; the resource request and allocation data collection unit is used for collecting resource request data and actual allocation results of each service type; and the satisfaction calculation and evaluation unit is used for calculating single-service satisfaction according to resource allocation comparison and quantitative indicators, and calculating total satisfaction through a short-board effect weight allocation method.

[0069] The output end of the service classification and priority definition unit is connected with the input end of the resource request and allocation data collection unit; the output end of the resource request and allocation data collection unit is connected with the input end of the satisfaction calculation and evaluation unit; and the output end of the satisfaction calculation and evaluation unit is connected with the input end of the resource allocation strategy formulation module.

[0070] The network load balancing evaluation module comprises a load data real-time collection unit, an optimal load value setting unit, and a load balancing degree calculation unit.

[0071] The load data real-time collection unit is used for collecting real-time load data of each node through a monitoring tool; the optimal load value setting unit is used for setting optimal load values for different types of load data; and the load balancing degree calculation unit is used for calculating single-node load balancing degree based on Beta distribution, and aggregating network-wide load balancing degree through node average values.

[0072] The output end of the load data real-time collection unit is connected with the input end of the optimal load value setting unit; the output end of the optimal load value setting unit is connected with the input end of the load balancing degree calculation unit; and the output end of the load balancing degree calculation unit is connected with the input end of the resource allocation strategy formulation module.

[0073] The future demand prediction module comprises a historical data collection unit, an LSTM model training unit, and a demand prediction generation unit.

[0074] The historical data collection unit is used to extract historical business demand data and perform time series formatting; the LSTM model training unit is used to build an LSTM model, train the model through an Adam optimizer and a mean square error loss function, and adjust hyperparameters; the demand prediction generation unit is used to predict future business demand using the trained LSTM model;

[0075] The output end of the historical data collection unit is connected with the input end of the LSTM model training unit; the output end of the LSTM model training unit is connected with the input end of the demand prediction generation unit; and the output end of the demand prediction generation unit is connected with the input end of the resource allocation strategy formulation module.

[0076] The resource allocation strategy formulation module comprises an optimization target definition unit, a constraint condition setting unit, and a strategy solving and generation unit.

[0077] The optimization target definition unit is used to weight and sum the business demand satisfaction degree and the network load balancing degree to form an optimization target function; the constraint condition setting unit is used to define resource total quantity constraints and load adjustment strategies; and the strategy solving and generation unit is used to solve an optimal resource allocation strategy using a multi-objective optimization algorithm and a penalty function method, and output a specific allocation scheme.

[0078] The output end of the optimization target definition unit is connected with the input end of the constraint condition setting unit; and the output end of the constraint condition setting unit is connected with the input end of the strategy solving and generation unit.

[0079] Compared with the prior art, the present application has the beneficial effects that: the weight distribution based on the short board effect can highlight the key factors affecting the business, avoid the situation that some key indicators are not satisfied and are covered by other better indicators, and make the evaluation of the business demand satisfaction degree more accurate and reasonable; the present application can more accurately measure the distribution of load data around the optimal load value by constructing a Beta distribution probability density function, thereby more accurately evaluating the load balancing degree of the network and providing a more reliable basis for network resource allocation. BRIEF DESCRIPTION OF DRAWINGS

[0080] Fig. 1 It is a step schematic diagram of the network resource allocation method based on network load of the present application;

[0081] Fig. 2 It is a flow schematic diagram of the network resource allocation system based on network load of the present application. DETAILED DESCRIPTION

[0082] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0083] Embodiment: As shown in the accompanying drawings, the present application provides a technical solution, a network resource allocation method based on network load, which comprises the following steps: Figs. 1-2

[0084] Step 1, determine the service demand satisfaction degree;

[0085] Step 2, calculate the network load balancing degree;

[0086] Step 3, predict future service demand data;

[0087] Step 4, define the constraint condition and maximize the service demand satisfaction degree and the network load balancing degree, and formulate the network resource allocation strategy.

[0088] In step 1, the specific steps for determining the service demand satisfaction degree are as follows:

[0089] Step 1.1, define the types of different services and their corresponding priorities;

[0090] The service types include but are not limited to core key services, secondary important services and general services;

[0091] The priority includes but is not limited to the highest priority, high priority, medium priority and low priority;

[0092] Step 1.2, collect resource request data and actual allocation of different service types;

[0093] The resource request data includes but is not limited to required bandwidth, computing resources and storage resources;

[0094] Step 1.3, select quantitative indicators and establish evaluation criteria according to different service types;

[0095] The quantitative indicators include but are not limited to delay time, packet loss rate, average processing time and resource utilization rate;

[0096] Different quantitative indicator conditions are set for different service types based on different priorities;

[0097] Step 1.4, compare the resource request data and the actual allocation of each service type, and calculate the service demand satisfaction degree combined with the corresponding quantitative indicator conditions in the evaluation criteria;

[0098] ​For a certain actual allocation or quantitative index of a certain service type:

[0099] When the actual allocation Re meets the resource request, the satisfaction degree is set to 1;

[0100] When the actual allocation Re does not meet the resource request condition threshold Ru, the satisfaction degree is set to 1- |Re-Ru| / Ru; wherein |Re-Ru| represents the absolute value of the difference between the actual allocation Re and the resource request condition threshold Ru;

[0101] When the quantitative index T meets the quantitative index condition, the satisfaction degree is set to 1;

[0102] When the quantitative index T does not meet the quantitative index condition threshold Th, the satisfaction degree is set to 1- |T-Th| / Th; wherein |T-Th| represents the absolute value of the difference between the quantitative index T and the quantitative index threshold Th;

[0103] Weight allocation method based on short board effect:

[0104] The satisfaction degree set is represented as S: S={s1, s2,..., sn}; wherein n is a positive integer, representing the number of quantitative indexes and requested resources; s1~sn respectively represent the satisfaction degrees of the 1st~nth quantitative index or requested resource; s1, s2,..., sn∈[0, 1];

[0105] The corresponding weight set W: W={w1, w2,..., wn}; wherein w1~wn respectively represent the weights corresponding to s1~sn; w1, w2,..., wn≥0;

[0106] Exponential function is used to amplify the influence of low value items: wi=e -si / (e -s1 +e -s2 +…+e -sn ); wherein i∈{1, 2, …, n}, si represents the quantitative index or requested resource satisfaction degree sequence, and wi represents the weight corresponding to si;

[0107] The total satisfaction degree of the service: St=s1×w1+s2×w2+…+sn×wn;

[0108] Calculate the total satisfaction degree of each service, and calculate the average of the total satisfaction degrees of all services as the total satisfaction degree of this scheduling.

[0109] In step 2, the specific steps for calculating the network load balancing degree are as follows:

[0110] Step 2.1, deploy monitoring tools in each node in the network to collect node load data in real time;

[0111] Load data includes but is not limited to CPU usage, memory occupancy and bandwidth utilization;

[0112] Step 2.2, set the optimal load value for different types of load data;

[0113] Step 2.3, according to the different optimal load value for the corresponding different types of load data to construct the probability density function of Beta distribution, calculate the load balancing degree;

[0114] For a type of load data: its optimal load value is represented as A, which means that the load balancing degree of this type of load data is the highest at A; 0 < A < 1;

[0115] Construct the probability density function f(x, a, b) of Beta distribution: combine the optimal load value A of this type of load data and the mean value μ or variance σ of the load data in a period of time 2 Determine a and b; where x represents the actual value of the load data, a and b are the parameters of the probability density function of Beta distribution; a > 1 and b > 1;

[0116] Take the probability density function value of the Beta distribution corresponding to the actual load data of this type as the load balancing degree value of this type of load data;

[0117] For a node, calculate the average value of the sum of the load balancing degree values of all types of load data contained in the node as the load balancing degree value of the node;

[0118] Step 2.4, calculate the total network load balancing degree;

[0119] Take the average value of the sum of the load balancing degree values of each node as the total network load balancing degree.

[0120] In step 3, the specific steps of predicting future business demand data are:

[0121] Step 3.1, collect historical business demand data;

[0122] Collect past business demand data from business system logs, databases and other data sources, including business traffic, transaction volume, user access volume and other specific indicators; record the timestamp corresponding to the data to ensure that the data has the characteristics of time series;

[0123] Step 3.2, construct and train LSTM time series analysis model;

[0124] Divide the data into training set and test set;

[0125] Build the LSTM model structure: Workers determine the LSTM model hyperparameters based on the complexity of the data and the difficulty of the prediction task. The number of neurons in each layer can be adjusted through experimentation to find the optimal model structure. At the same time, input and output layers are added to convert the input data into a format suitable for LSTM processing and output the prediction results.

[0126] The training set data is input into the constructed LSTM model and trained using the Adam optimization algorithm and the mean squared error loss function. During the training process, the model parameters are continuously adjusted to minimize the value of the loss function, thereby improving the model's ability to fit the historical data. The training process can be repeated for multiple rounds until the model performance no longer improves significantly.

[0127] Step 3.3: Use the trained model to make predictions.

[0128] Input the test set data into the trained LSTM model, and the model will output predicted values ​​for future business needs based on the learned patterns and trends.

[0129] In step 4, define the constraints and maximize the satisfaction of business requirements and the degree of network load balancing. The specific implementation steps for formulating the network resource allocation strategy are as follows:

[0130] Step 4.1: Define the optimization goal.

[0131] Assign different weights to the business demand satisfaction and network load balancing degree, and perform weighted summation to form a new optimization goal;

[0132] Step 4.2: Define constraints.

[0133] Business resource allocation constraints:

[0134] Total resource constraint: The total amount of resources actually allocated to all businesses must not exceed the total available amount of the resource;

[0135] Network load balancing constraints:

[0136] Load adjustment strategy: When the load of a node is greater than the optimal load value, reduce the tasks assigned to it and migrate some tasks to nodes with lower load; if the load of a node is too low, increase the tasks assigned to it;

[0137] Step 4.3: Use the constrained optimization algorithm to solve the specified network resource allocation strategy;

[0138] Use multi-objective optimization algorithms and penalty function methods for iterative optimization, continuously adjust demand distribution and load to maximize the optimization goal; formulate network resource allocation strategies based on the maximized optimization goal.

[0139] A network resource allocation system based on network load, the system comprising a service demand satisfaction degree calculation module, a network load balance evaluation module, a future demand prediction module and a resource allocation strategy formulation module;

[0140] The service demand satisfaction degree calculation module is used for determining service demand satisfaction degree;

[0141] The network load balance evaluation module is used for calculating network load balance degree;

[0142] The future demand prediction module is used for predicting future service demand data;

[0143] The resource allocation strategy formulation module is used for defining constraint conditions and maximizing service demand satisfaction degree and network load balance degree, and formulating network resource allocation strategy;

[0144] The output end of the service demand satisfaction degree calculation module is connected with the input end of the resource allocation strategy formulation module; the output end of the network load balance evaluation module is connected with the input end of the resource allocation strategy formulation module; and the output end of the future demand prediction module is connected with the input end of the resource allocation strategy formulation module.

[0145] The service demand satisfaction degree calculation module comprises a service classification and priority definition unit, a resource request and allocation data collection unit and a satisfaction degree calculation and evaluation unit;

[0146] The service classification and priority definition unit is used for defining service types and their priorities to form classification standards; the resource request and allocation data collection unit is used for collecting resource request data and actual allocation results of each service type; and the satisfaction degree calculation and evaluation unit is used for calculating single service satisfaction degree according to resource allocation comparison and quantitative indexes, and calculating total satisfaction degree through a short board effect weight distribution method;

[0147] The output end of the service classification and priority definition unit is connected with the input end of the resource request and allocation data collection unit; the output end of the resource request and allocation data collection unit is connected with the input end of the satisfaction degree calculation and evaluation unit; and the output end of the satisfaction degree calculation and evaluation unit is connected with the input end of the resource allocation strategy formulation module.

[0148] The network load balance evaluation module comprises a load data real-time collection unit, an optimal load value setting unit and a load balance degree calculation unit;

[0149] The load data real-time acquisition unit is configured to acquire real-time load data of each node through a monitoring tool; the optimal load value setting unit is configured to set optimal load values for different types of load data; and the load balancing degree calculation unit is configured to calculate a single-node load balancing degree based on a Beta distribution and aggregate a whole-network load balancing degree through a node average value.

[0150] An output end of the load data real-time acquisition unit is connected to an input end of the optimal load value setting unit; an output end of the optimal load value setting unit is connected to an input end of the load balancing degree calculation unit; and an output end of the load balancing degree calculation unit is connected to an input end of the resource allocation strategy formulation module.

[0151] The future demand prediction module includes a historical data collection unit, an LSTM model training unit, and a demand prediction generation unit.

[0152] The historical data collection unit is configured to extract historical business demand data and perform time series formatting; the LSTM model training unit is configured to construct an LSTM model, train the model through an Adam optimizer and a mean square error loss function, and adjust hyperparameters; and the demand prediction generation unit is configured to predict future business demand using the trained LSTM model.

[0153] An output end of the historical data collection unit is connected to an input end of the LSTM model training unit; an output end of the LSTM model training unit is connected to an input end of the demand prediction generation unit; and an output end of the demand prediction generation unit is connected to an input end of the resource allocation strategy formulation module.

[0154] The resource allocation strategy formulation module includes an optimization objective definition unit, a constraint condition setting unit, and a strategy solving and generation unit.

[0155] The optimization objective definition unit is configured to weight-sum a business demand satisfaction degree and a network load balancing degree to form an optimization objective function; the constraint condition setting unit is configured to define resource total quantity constraints and load adjustment strategies; and the strategy solving and generation unit is configured to solve an optimal resource allocation strategy using a multi-objective optimization algorithm and a penalty function method and output a specific allocation scheme.

[0156] An output end of the optimization objective definition unit is connected to an input end of the constraint condition setting unit; and an output end of the constraint condition setting unit is connected to an input end of the strategy solving and generation unit.

[0157] In this embodiment, an e-commerce company is about to usher in the annual large-scale promotion activities, and the business volume is expected to increase significantly during the activities. The company's network system carries core critical business (such as commodity transaction, payment processing), less important business (such as user information query, order status tracking) and general business (such as advertisement display, internal letter push). In order to ensure the stable operation of the network during the activities, reasonable network resource allocation is needed.

[0158] Step 1, determine the degree of satisfaction of business requirements;

[0159] Step 1.1: Define business type and priority:

[0160] Core critical business: commodity transaction, payment processing, priority is the highest priority.

[0161] Less important business: user information query, order status tracking, priority is high priority.

[0162] General business: advertisement display, internal letter push, priority is medium priority.

[0163] Step 1.2: Collect resource request data and actual allocation:

[0164] Core critical business: required bandwidth is 100 Mbps, computing resources are 50 CPU cores, and storage resources are 100 GB. The actual allocated bandwidth is 80 Mbps, CPU cores are 40, and storage resources are 80 GB.

[0165] Less important business: required bandwidth 30 Mbps, computing resources 20 CPU cores, storage resources 50 GB. Actual allocated bandwidth 25 Mbps, CPU cores 15, storage resources 40 GB.

[0166] General business: required bandwidth 10 Mbps, computing resources 5 CPU cores, storage resources 20 GB. Actual allocated bandwidth 8 Mbps, CPU cores 4, storage resources 15 GB.

[0167] Step 1.3: Select quantitative indicators and establish evaluation criteria:

[0168] Delay time: core critical business threshold is 50 ms, less important business is 100 ms, general business is 200 ms.

[0169] Packet loss rate: core critical business threshold is 0.1%, less important business is 0.5%, general business is 1%.

[0170] Average processing time: core critical business threshold is 100 ms, less important business is 200 ms, general business is 500 ms.

[0171] Resource utilization: 80% for core critical business, 70% for secondary important business, and 60% for general business.

[0172] Step 1.4: Calculate the business demand satisfaction degree:

[0173] Core critical business:

[0174] Bandwidth satisfaction degree: actual allocation 80Mbps, request 100Mbps, satisfaction degree 1-|80-100| / 100=0.8.

[0175] Compute resource satisfaction degree: actual allocation 40 CPU cores, request 50, satisfaction degree 1-|40-50| / 50=0.8.

[0176] Storage resource satisfaction degree: actual allocation 80GB, request 100GB, satisfaction degree 1-|80-100| / 100=0.8.

[0177] Latency time satisfaction degree: assume actual latency time is 60ms, satisfaction degree 1-|60-50| / 50=0.8.

[0178] Packet loss rate satisfaction degree: assume actual packet loss rate is 0.2%, satisfaction degree 1-|0.2%-0.1%| / 0.1%=0.

[0179] Average processing time satisfaction degree: assume actual average processing time is 120ms, satisfaction degree 1-|120-100| / 100=0.8.

[0180] Resource utilization satisfaction degree: assume actual resource utilization is 70%, satisfaction degree 1-|70%-80%| / 80%=0.875.

[0181] Satisfaction degree set S={0.8,0.8,0.8,0.8,0,0.8,0.875}.

[0182] Weight set W:

[0183] w1=e^(-0.8) / (e^(-0.8)+e^(-0.8)+e^(-0.8)+e^(-0.8)+e^(0)+e^(-0.8)+e^(-0.875))≈0.12.

[0184] Similarly, calculate other weights.

[0185] Total satisfaction degree St=0.8×0.12+0.8×0.12+0.8×0.12+0.8×0.12+0×0.12+0.8×0.12+0.875×0.12≈0.67.

[0186] The same method is used to calculate the secondary business and general business satisfaction, which is 0.75 and 0.8 respectively.

[0187] The average value of total satisfaction is (0.67+0.75+0.8) / 3≈0.74.

[0188] Step 2, calculate the network load balancing degree;

[0189] Step 2.1: Deploy monitoring tools in 5 nodes in the network to collect load data in real time. For example, the CPU usage of node 1 is 70%, the memory usage is 60%, and the bandwidth utilization is 50%.

[0190] Step 2.2: Set the optimal load value:

[0191] The optimal load value of CPU usage A1=0.6.

[0192] The optimal load value of memory usage A2=0.5.

[0193] The optimal load value of bandwidth utilization A3=0.4.

[0194] Step 2.3: Construct the probability density function of Beta distribution:

[0195] For CPU usage, the mean value μ=0.65 and the variance σ²=0.01 in a period of time. Through the formula, α=3 and β=2 are determined.

[0196] For the CPU usage of node 1, the actual value x=0.7, and the load balancing degree value is f(0.7,3,2).

[0197] The same method is used to calculate the load balancing degree values of memory usage and bandwidth utilization, which are 0.75 and 0.85 respectively.

[0198] The load balancing degree value of node 1 is (0.8+0.75+0.85) / 3=0.8.

[0199] Calculate the load balancing degree values of other nodes, which are 0.78, 0.82, 0.75, and 0.8 respectively.

[0200] Step 2.4: The total network load balancing degree is (0.8+0.78+0.82+0.75+0.8) / 5=0.79.

[0201] Step 3, predict future business demand data;

[0202] Step 3.1: Collect business demand data in the past year from business system logs and databases, such as daily commodity transaction quantity, user access volume, etc., and record the timestamp.

[0203] Step 3.2: Building and training LSTM time series analysis model: Divide the data into 70% training set and 30% test set. Build LSTM model structure: assume 3 layers of LSTM layers with 64, 32, and 16 neurons respectively. Add input layer and output layer. Use optimization algorithm Adam and mean square error loss function for training, and after 50 iterations, the model performance no longer improves significantly.

[0204] Step 3.3: Input test set data into the trained model to predict the business demand data for the next week.

[0205] Step 4, define constraints and maximize business demand satisfaction and network load balancing, develop network resource allocation strategy;

[0206] Step 4.1: Define optimization goal: assume business demand satisfaction weight is 0.6, network load balancing weight is 0.4. Optimization goal = 0.6 x business demand satisfaction + 0.4 x network load balancing.

[0207] Step 4.2: Define constraints: business resource allocation constraints: assume total bandwidth is 200Mbps, total CPU cores is 100, total storage resources is 200GB. Actual allocation to all business resources cannot exceed these total amounts.

[0208] Network load balancing constraints: when node load is greater than optimal load value, such as CPU usage of node 1 is greater than 0.6, reduce its allocated tasks and migrate part of the tasks to nodes with lower load. When node load is too low, increase the allocation of tasks.

[0209] Step 4.3: Solve using constraint optimization algorithm: use multi-objective optimization algorithm and penalty function method for iterative optimization. For example, after multiple iterations, it is found that increasing the bandwidth allocation of core key business to 90Mbps and adjusting the task allocation of some nodes can maximize the optimization goal.

[0210] Develop network resource allocation strategy according to the maximized optimization goal: increase core key business bandwidth to 90Mbps, compute resources to 45 CPU cores, and storage resources to 90GB.

[0211] Adjust the resource allocation of secondary important business and general business to ensure that the total resource does not exceed the limit.

[0212] Migrate tasks to network nodes to make node load closer to the optimal load value.

[0213] Through the above examples, the data processing process and specific implementation steps of the network resource allocation method based on network load in the actual e-commerce business scenario are demonstrated, which can effectively improve the rational use of network resources and the stability of business.

[0214] It will be apparent to those skilled in the art that the application is not limited to the details of the above-exemplified embodiments and that the present application can be implemented in other particular forms without departing from the spirit or essential characteristics of the present application. The presently disclosed embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. No reference herein to any prior art is to be taken as an admission that the present application is not entitled to antedate such prior art by virtue of prior application. No reference to any prior art in this specification is intended to be, nor should it be taken as, an acknowledgement or any admission that the prior art forms a part of the common general knowledge in the relevant art.

Claims

1. A network resource allocation method based on network load, characterized by: The method comprises the following steps: Step 1: Determine the degree of satisfaction of business needs; Step 2: Calculate the degree of network load balancing; Step 3: Forecast future business demand data; Step 4: Define constraints and maximize service demand satisfaction and network load balancing, and formulate a network resource allocation strategy; In step 1, the specific steps to determine the satisfaction of business requirements are: Step 1.1: Define different service types and their corresponding priorities. Step 1.2: Collect resource request data and actual allocation for different business types; Step 1.3: Select quantitative indicators and establish evaluation criteria according to different business types; Set different quantitative indicator conditions for different business types based on different priorities; Step 1.4: Compare the resource request data and actual allocation for each business type, and calculate the satisfaction of business requirements based on the corresponding quantitative indicators in the evaluation criteria. In step 2, the specific steps for calculating the network load balancing degree are as follows: Step 2.1: Deploy monitoring tools at each node in the network to collect node load data in real time. Step 2.2, set the optimal load value for different types of load data; Step 2.3: Construct a Beta distribution probability density function for different types of load data based on different optimal load values, and calculate the load balancing degree; Step 2.4, calculate the total network load balancing degree; In step 4, define the constraints and maximize the satisfaction of business requirements and the degree of network load balancing. The specific implementation steps for formulating the network resource allocation strategy are as follows: Step 4.1: Define the optimization goal. Assign different weights to the business demand satisfaction and network load balancing degree, and perform weighted summation to form a new optimization goal; Step 4.2: Define constraints. Business resource allocation constraints: Total resource constraint: The total amount of resources actually allocated to all businesses must not exceed the total available amount of the resource; Network load balancing constraints: Load adjustment strategy: When the load of a node is greater than the optimal load value, the assigned tasks are reduced and some tasks are migrated to nodes with lower load; when the load of a node is less than the optimal load value, the assigned tasks are increased; Step 4.3: Use the constrained optimization algorithm to solve the specified network resource allocation strategy; Use multi-objective optimization algorithms and penalty function methods for iterative optimization, continuously adjust demand distribution and load to maximize the optimization goal; formulate network resource allocation strategies based on the maximized optimization goal.

2. The network resource allocation method based on network load according to claim 1, characterized in that: In step 1, the following steps are specifically included: For a certain actual distribution or quantitative indicator of a certain business type: When the actual allocation situation Re satisfies the resource request, the satisfaction degree is set to 1; When the actual allocation situation Re does not meet the resource request condition threshold Ru, the satisfaction degree is set to 1-|Re-Ru| / Ru; where Ru represents the lower limit or upper limit of the resource request; |Re-Ru| represents the absolute value of the difference between the actual allocation situation Re and the resource request condition threshold Ru; When the quantitative index T meets the quantitative index condition, the satisfaction degree is set to 1; When the quantitative index T does not meet the quantitative index condition threshold Th, the satisfaction degree is set to 1-|T-Th| / Th; where Th represents the lower limit or upper limit of the quantitative index; |T-Th| represents the absolute value of the difference between the quantitative index T and the quantitative index threshold Th; Weight allocation method based on the short board effect: The satisfaction set is represented as S: S={s1,s2,...,sn}; where n is a positive integer, representing the number of quantitative indicators and requested resources; s1~sn represent the satisfaction of the 1st to nth quantitative indicators or requested resources respectively; s1,s2,...,sn∈[0,1]; The corresponding weight set W: W={w1,w2,...,wn}; where w1~wn represent the weights corresponding to s1~sn respectively; w1,w2,...,wn≥0; Use the exponential function to amplify the influence of low-value terms: wi=e -si / (e -s1 +e -s2 +…+e -sn ); where i∈{1,2,…,n}, si represents a quantitative indicator or a sequence of requested resource satisfaction, wi represents the weight corresponding to si; e represents a natural constant; The total satisfaction of the service: St = s1×w1+s2×w2+…+sn×wn; Calculate the total satisfaction of each business and the average of the total satisfaction of all businesses as the total satisfaction of this scheduling.

3. The network resource allocation method based on network load according to claim 2, characterized in that: In step 2, the following steps are specifically included: For a certain type of load data: its optimal load value is represented by A, indicating that the load balancing degree of this type of load data is the highest at A; 0<A<1; Construct the probability density function f(x,α,β) of the Beta distribution: combine the optimal load value A of this type of load data and the mean μ or variance σ of the load data over a period of time 2 , determine α and β; where x represents the actual value of the load data, α and β are the probability density function parameters of the Beta distribution; α>1 and β>1; The probability density function value of the Beta distribution corresponding to the actual load data of this type is taken as the load balancing degree value of this type of load data; For a node, the average value of the sum of the load balancing degree values ​​of all types of load data contained in it is calculated as the load balancing degree value of the node; The average of the sum of the load balancing degree values ​​of each node is taken as the total network load balancing degree.

4. The network resource allocation method based on network load according to claim 3, characterized in that: In step 3, the specific steps for predicting future business demand data are: Step 3.1: Collect historical business demand data; Collect past business demand data; record the timestamps corresponding to the data to ensure that the data has time series characteristics; Step 3.2: Build and train the LSTM time series analysis model; Divide the data into training and test sets; build the LSTM model structure; add input and output layers; input the training set data into the constructed LSTM model and train it using the Adam optimization algorithm and the mean square error loss function; During the training process, the model parameters are continuously adjusted and multiple rounds of iterations are performed to minimize the value of the loss function; Step 3.3: Use the trained model to make predictions. The test set data is input into the trained LSTM model, and the model outputs the predicted value of future business demand.

5. A network resource allocation system based on network load, applied to the network resource allocation method based on network load according to any one of claims 1 to 4, characterized in that: The system includes a business demand satisfaction calculation module, a network load balancing evaluation module, a future demand prediction module, and a resource allocation strategy formulation module; The business requirement satisfaction calculation module is used to determine the business requirement satisfaction; The network load balancing evaluation module is used to calculate the degree of network load balancing; The future demand forecasting module is used to forecast future business demand data; The resource allocation strategy formulation module is used to define constraints and maximize business demand satisfaction and network load balancing, and formulate network resource allocation strategies; The output end of the business demand satisfaction calculation module is connected to the input end of the resource allocation strategy formulation module; the output end of the network load balancing evaluation module is connected to the input end of the resource allocation strategy formulation module; the output end of the future demand prediction module is connected to the input end of the resource allocation strategy formulation module.

6. The network resource allocation system based on network load according to claim 5, characterized in that: The business demand satisfaction calculation module includes a business classification and priority definition unit, a resource request and allocation data collection unit, and a satisfaction calculation and evaluation unit; The service classification and priority definition unit is used to define service types and their priorities to form classification standards; the resource request and allocation data collection unit is used to collect resource request data and actual allocation results of each service type; the satisfaction calculation and evaluation unit is used to calculate the satisfaction of each service based on resource allocation comparison and quantitative indicators, and calculate the total satisfaction through the weak link effect weight allocation method; The output end of the business classification and priority definition unit is connected to the input end of the resource request and allocation data collection unit; the output end of the resource request and allocation data collection unit is connected to the input end of the satisfaction calculation and evaluation unit; the output end of the satisfaction calculation and evaluation unit is connected to the input end of the resource allocation strategy formulation module.

7. The network resource allocation system based on network load according to claim 6, characterized in that: The network load balancing evaluation module includes a load data real-time acquisition unit, an optimal load value setting unit, and a load balancing degree calculation unit; The load data real-time collection unit is used to collect real-time load data of each node through a monitoring tool; the optimal load value setting unit is used to set the optimal load value for different types of load data; the load balancing degree calculation unit is used to calculate the load balancing degree of a single node based on Beta distribution, and summarize the load balancing degree of the entire network through the node average value; The output end of the load data real-time acquisition unit is connected to the input end of the optimal load value setting unit; the output end of the optimal load value setting unit is connected to the input end of the load balancing degree calculation unit; the output end of the load balancing degree calculation unit is connected to the input end of the resource allocation strategy formulation module.

8. The network resource allocation system based on network load according to claim 7, characterized in that: The future demand forecasting module includes a historical data collection unit, an LSTM model training unit, and a demand forecast generation unit; The historical data collection unit is used to extract historical business demand data and format it into time series; the LSTM model training unit is used to build an LSTM model, train the model using the Adam optimizer and the mean square error loss function, and adjust hyperparameters; the demand forecast generation unit is used to use the trained LSTM model to predict future business demand; The output end of the historical data collection unit is connected to the input end of the LSTM model training unit; the output end of the LSTM model training unit is connected to the input end of the demand forecast generation unit; the output end of the demand forecast generation unit is connected to the input end of the resource allocation strategy formulation module.

9. The network resource allocation system based on network load according to claim 8, characterized in that: The resource allocation strategy formulation module includes an optimization target definition unit, a constraint condition setting unit, and a strategy solving and generating unit; The optimization target definition unit is used to perform weighted summation of the service demand satisfaction and the network load balancing degree to form an optimization target function; the constraint condition setting unit is used to define the total resource constraint and the load adjustment strategy; the strategy solving and generating unit is used to use the multi-objective optimization algorithm and the penalty function method to solve the optimal resource allocation strategy and output a specific deployment plan; The output end of the optimization target definition unit is connected to the input end of the constraint condition setting unit; the output end of the constraint condition setting unit is connected to the input end of the strategy solving and generating unit.

Images