Optimized cloud resource charging method, system and equipment based on prediction of cloud network resources

Abstract

The invention relates to an optimized cloud resource charging method, system and device based on prediction of cloud network resources. The method comprises the following steps: obtaining use data of each use device using the cloud network resources and charging information of the cloud network resources; constructing a charging model according to the charging information and all the use data, outputting bill information corresponding to all the use data by the charging model, and determining index data related to cloud network resources from the bill information; acquiring historical sequence data of the cloud network resources, and predicting the use data by adopting an LSTM-ARMA model according to the historical sequence data to obtain predicted use data; and optimizing a charging rule of the charging model according to the predicted use data to obtain a cloud resource charging model. According to the method, the charging model is optimized by predicting the predicted use data used by the cloud network resources in the future to obtain the optimized cloud resource charging model, and the charging rule can be adjusted according to the actual demand, so that the calculation of the use cost of the cloud network resources is high in accuracy and reasonable.

Description

Technical Field

[0001] The present application relates to the field of resource management technology, and in particular to a method, system and device for optimizing cloud resource billing based on predicting cloud network resources. Background Art

[0002] Cloud billing is the fee charged by cloud service providers to customers for using their cloud platform and cloud resources. Traditional cloud resource billing is based on the billing method of initial resource purchase. As the usage changes in the later period, the billing method of cloud resources remains unchanged, resulting in cloud billing billing methods that do not adapt to market changes, making cloud billing unreasonable. Secondly, the traditional cloud resource billing model is issued by the cloud platform, which will not be associated with and analyzed by the business, and cannot form a business cloud resource usage portrait. Therefore, in the existing hybrid cloud service environment, enterprises often find it difficult to accurately predict future resource needs, and there may be over-purchase or insufficient resources, resulting in low resource utilization or high cloud service costs. Summary of the invention

[0003] The present application provides a method, system and device for optimizing cloud resource billing based on predicting cloud network resources, which is used to solve the technical problem that traditional cloud resource billing methods cannot predict future required resources and have unreasonable cloud billing.

[0004] In order to achieve the above objectives, this application provides the following technical solutions:

[0005] On the one hand, a method for optimizing cloud resource billing based on predicting cloud network resources is provided, comprising the following steps:

[0006] Obtaining usage data of each device using cloud network resources and charging information of the cloud network resources, wherein the charging information includes billing items, billing rules and service information;

[0007] Building a charging model based on the charging information and all the usage data, the charging model outputs billing information corresponding to all the usage data, and determining indicator data related to the cloud network resources from the billing information;

[0008] Acquire historical sequence data of the cloud network resources corresponding to the indicator data, and use the LSTM-ARMA model to predict usage data according to the historical sequence data to obtain predicted usage data;

[0009] The billing rules of the charging model are optimized according to the predicted usage data to obtain an optimized cloud resource billing model.

[0010] Preferably, the usage data is predicted using the LSTM-ARMA model according to the historical sequence data, and the predicted usage data obtained includes:

[0011] Preprocessing the historical sequence data to obtain stable time historical sequence data; and obtaining the predicted duration;

[0012] The usage data is predicted using the LSTM-ARMA model according to the predicted duration and the time history series data to obtain predicted usage data.

[0013] Preferably, the usage data is predicted using an LSTM-ARMA model according to the predicted duration and the time history series data, and the content of the predicted usage data includes:

[0014] Using the ARMA model of the LSTM-ARMA model to perform a stationarity test on the time history series data to obtain a test result;

[0015] If the detection result shows that the time history series data is stable, the parameters of the ARMA model are estimated by using maximum likelihood estimation to obtain estimated parameters; the ARMA model is used to perform prediction based on the estimated parameters and the time history series data to obtain residual error history series data;

[0016] According to the prediction duration and the residual error historical sequence data, the LSTM model of the LSTM-ARMA model is used to perform sequence nonlinear capture prediction to obtain predicted usage data.

[0017] Preferably, the billing rules of the charging model are optimized according to the predicted usage data, and the optimized cloud resource billing model includes:

[0018] Obtaining the unit price of each billing mode of the billing rule, and calculating according to the unit price of each billing mode to obtain optimization judgment data;

[0019] Comparing the predicted usage data with the optimization judgment data to obtain a comparison result;

[0020] An updated billing model corresponding to the cloud network resources is determined according to the comparison result, and the updated billing model is used to replace the billing model of the billing rules in the charging model to obtain an optimized cloud resource billing model.

[0021] Preferably, the indicator data includes resource utilization rate, resource utilization flow and resource utilization capacity; if the predicted utilization data is predicted utilization flow, then according to the predicted utilization flow being greater than the optimization judgment data, the updated billing mode is selected as a bandwidth monthly billing mode; or according to the predicted utilization flow being less than the optimization judgment data, the updated billing mode is selected as a bandwidth on-demand billing mode; or according to the predicted utilization flow being equal to the optimization judgment data, the updated billing mode is selected as a bandwidth on-demand billing mode or a bandwidth monthly billing mode.

[0022] In yet another aspect, there is provided an optimized cloud resource billing system based on predicted cloud network resources, comprising a data acquisition module, a model building module, a predicted data module, and a model optimization module;

[0023] The data acquisition module is used to acquire usage data of each device using cloud network resources and charging information of the cloud network resources, wherein the charging information includes billing items, billing rules and service information;

[0024] The model building module is used to build a charging model based on the charging information and all the usage data, the charging model outputs billing information corresponding to all the usage data, and determines the indicator data related to the cloud network resources from the billing information;

[0025] The prediction data module is used to obtain the historical sequence data of the cloud network resources corresponding to the indicator data, and use the LSTM-ARMA model to predict the usage data according to the historical sequence data to obtain the predicted usage data;

[0026] The model optimization module is used to optimize the billing rules of the charging model according to the predicted usage data to obtain an optimized cloud resource billing model.

[0027] Preferably, the prediction data module includes a data processing submodule and a prediction submodule;

[0028] The data processing submodule is used to preprocess the historical sequence data to obtain stable time historical sequence data; and to obtain the predicted duration;

[0029] The prediction submodule is used to use the ARMA model of the LSTM-ARMA model to perform a stationarity test on the time history series data to obtain a test result; based on the test result that the time history series data is stationary, the parameters of the ARMA model are estimated by maximum likelihood estimation to obtain estimated parameters; based on the estimated parameters and the time history series data, the ARMA model is used to perform prediction to obtain residual error history series data; based on the prediction duration and the residual error history series data, the LSTM model of the LSTM-ARMA model is used to perform sequence nonlinear capture prediction to obtain predicted usage data.

[0030] Preferably, the model optimization module includes an optimization data acquisition submodule, a comparison submodule and an optimization submodule;

[0031] The optimization data acquisition submodule is used to obtain the unit price of each billing mode of the billing rule, and obtain the optimization judgment data according to the calculation of each billing mode unit price;

[0032] The comparison submodule is used to compare the predicted usage data with the optimization judgment data to obtain a comparison result;

[0033] The optimization submodule is used to determine an updated billing model corresponding to the cloud network resources according to the comparison result, and use the updated billing model to replace the billing model of the billing rules in the charging model to obtain an optimized cloud resource billing model.

[0034] Preferably, the indicator data includes resource utilization rate, resource utilization flow and resource utilization capacity; if the predicted utilization data is predicted utilization flow, then according to the predicted utilization flow being greater than the optimization judgment data, the updated billing mode is selected as a bandwidth monthly billing mode; or according to the predicted utilization flow being less than the optimization judgment data, the updated billing mode is selected as a bandwidth on-demand billing mode; or according to the predicted utilization flow being equal to the optimization judgment data, the updated billing mode is selected as a bandwidth on-demand billing mode or a bandwidth monthly billing mode.

[0035] In another aspect, a terminal device is provided, comprising a processor and a memory;

[0036] The memory is used to store program codes and transmit the program codes to the processor;

[0037] The processor is used to execute the above-mentioned optimized cloud resource billing method based on predicted cloud network resources according to the instructions in the program code.

[0038] The optimized cloud resource billing method, system and device based on predicting cloud network resources include obtaining usage data of each device using the cloud network resources and charging information of the cloud network resources, wherein the charging information includes billing items, billing rules and business information; building a charging model based on the charging information and all usage data, wherein the charging model outputs billing information corresponding to all usage data, and determines indicator data related to the cloud network resources from the billing information; obtaining historical sequence data of the cloud network resources corresponding to the indicator data, and using the LSTM-ARMA model to predict the usage data based on the historical sequence data to obtain predicted usage data; optimizing the billing rules of the charging model based on the predicted usage data to obtain an optimized cloud resource billing model.

[0039] It can be seen from the above technical scheme that the present application has the following advantages: the optimized cloud resource billing method based on predicted cloud network resources first constructs a charging model based on the actual usage data and charging information of the cloud network resources, and then optimizes the charging model based on the predicted usage data of the future use of the cloud network resources predicted by the indicator data, thereby obtaining an optimized cloud resource billing model. The cloud resource billing model can adjust the billing rules according to actual needs, so that the calculation of the cloud network resource usage fee is accurate and reasonable, which solves the technical problem that the traditional cloud resource billing method cannot predict the resources required in the future and has unreasonable cloud billing.

[0040] The optimized cloud resource billing system based on predicted cloud network resources cooperates with each other through the data acquisition module, model construction module, prediction data module and model optimization module to achieve the setting of a reasonable billing model based on the actual usage of resources and the prediction of future usage of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 This is a flowchart of the steps of the method for optimizing cloud resource billing based on predicting cloud network resources described in an embodiment of the present application;

[0043] Figure 2 A schematic diagram of the framework of the optimized cloud resource billing system based on predicted cloud network resources described in an embodiment of the present application;

[0044] Figure 3 A schematic diagram of a terminal device described in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, features, and advantages of the invention of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the embodiments described below are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] In the description of the embodiments of the present application, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0047] In the embodiments of the present application, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0048] In the application embodiment, the ‌‌ARMA model (autoregressive moving average model) is a basic model for time series analysis, which is used to fit stationary time series. The ARMA model is the most commonly used model for fitting stationary series, which can be subdivided into three categories: AR model (autoregressive model), MA model (moving average model) and ARMA model (autoregressive moving average model). The ARMA model is a classic linear statistical model that describes the correlation and average value of the time series through autoregressive (AR) and moving average (MA) parts. The modeling steps of the ‌‌ARMA model include:

[0049] Step 1: Perform zero mean stabilization on the time series. If the time series is a non-stationary series with a trend, it needs to be stabilized, i.e. zero mean and stabilization.

[0050] Step 2: gradually increase the model order and fit the ARMA (n, n-1) model, that is, increase the model order by one order and one order, and use the nonlinear least squares method to estimate the model parameters. The specific algorithm uses the steepest descent method, for example, the model corresponding to the minimum variance of the residual sequence is selected as the preliminary model;

[0051] Step 3: Test the adaptability of the preliminary model, for example, by using the correlation function method;

[0052] Step 4: Find the optimal model; check whether the preliminary model contains small parameters. If so, use the F test to determine whether it can be deleted and a lower-order model can be fitted to obtain the optimal model in the system sense.

[0053] Step 5: Use the optimal model to predict deformation time series.

[0054] In the application embodiment, the LSTM (Long Short-Term Memory) model is a special recurrent neural network (RNN) designed to solve the gradient vanishing and gradient exploding problems of traditional RNN when processing long historical sequence data. LSTM effectively alleviates these problems by introducing a gating mechanism, including an input gate, a forget gate, and an output gate, to control the flow of information. The memory unit of LSTM can store information for a long time, allowing the model to capture long-distance dependencies in the sequence. The structure and working mechanism of the LSTM model: The core of the LSTM model is its special unit structure, each unit contains three gates: an input gate, a forget gate, and an output gate. The input gate controls how much new information is stored in the unit state, the forget gate controls how much information is retained in the current unit state, and the output gate determines how much information is output from the unit state. These gates allow LSTM to maintain and update memory over a longer period of time by controlling the flow of information.

[0055] The embodiments of the present application provide a method, system and device for optimizing cloud resource billing based on predicting cloud network resources, which solves the technical problem that the traditional cloud resource billing method cannot predict the resources required in the future and has unreasonable cloud billing. The method, system and device for optimizing cloud resource billing based on predicting cloud network resources can be applied to the resource management field focusing on cloud computing and multi-cloud platforms, such as resource optimization, cloud resource capacity planning, resource usage evaluation and cloud service cost management in hybrid multi-cloud environments.

[0056] Embodiment 1:

[0057] Figure 1 This is a flowchart of the steps of the method for optimizing cloud resource billing based on predicted cloud network resources described in an embodiment of the present application.

[0058] like Figure 1 As shown, the embodiment of the present application provides a method for optimizing cloud resource billing based on predicted cloud network resources, comprising the following steps:

[0059] S1. Obtain usage data of each device that uses cloud network resources and charging information of cloud network resources, where the charging information includes billing items, billing rules and business information.

[0060] It should be noted that in step S1, the usage data of each device using cloud network resources and the charging information of cloud network resources are obtained to provide data for subsequent steps. In this embodiment, the charging information also includes the direct contract between the enterprise using cloud network resources and the enterprise using cloud network resources. The device used can be a host, a storage device (such as a hard disk, an optical disk, etc.) and a bandwidth device (such as a gateway, etc.). The usage data is the usage of cloud network resources. Business information includes cloud computing services, cloud storage services, cloud network security services, and other cloud-related services. The billing items include service fees, storage fees, etc. The billing rules include billing modes such as pay-as-you-go, pay-as-you-go, annual and monthly packages, resource packages, and usage time. The usage situation obtained in step S1 is to monitor and measure the usage data of cloud network resources in real time by making full use of cloud computing and network virtualization technologies, ensuring that users only pay for the resources they actually use, avoiding resource waste and unfair billing.

[0061] In an embodiment of the present application, the optimized cloud resource billing method based on predicted cloud network resources detects usage data of each device using cloud network resources through detection equipment and metering tools.

[0062] It should be noted that the detection equipment and measurement tools may be a network tester, a bandwidth analyzer, a throughput tester and / or an 800G network tester, etc.

[0063] S2. A charging model is constructed based on the charging information and all usage data. The charging model outputs billing information corresponding to all usage data, and indicator data related to cloud network resources is determined from the billing information. The indicator data includes resource utilization rate, resource usage flow, and resource usage capacity.

[0064] It should be noted that in step S2, firstly, a charging model is constructed based on the data obtained in step S1, and secondly, bill information corresponding to all usage data is output based on the charging model. The bill information generated by the charging model is accurate, and the resources consumed by each device and the corresponding fees can be clearly understood, thereby improving the cost awareness of users of cloud network resources and forming a profile of the use of each resource.

[0065] S3. Obtain the historical sequence data of the cloud network resources corresponding to the indicator data, and use the LSTM-ARMA model to predict the usage data based on the historical sequence data to obtain the predicted usage data.

[0066] It should be noted that in step S3, first, the historical sequence data of cloud network resources is obtained according to the indicator data determined in step S2, and second, the LSTM-ARMA model is used to predict the historical sequence data to obtain the predicted usage data, which provides data for the subsequent optimization of the charging model. In this embodiment, the prediction of the optimized cloud resource billing method based on the prediction of cloud network resources introduces a combination of deep learning and machine learning models (ARMA-LSTM model) to more accurately capture the future usage of cloud network resources. Among them, the ARMA model is used to predict the historical sequence data of cloud network resource related indicators, and the error in the historical sequence data is predicted using the LSTM model. The ARMA model is combined with the LSTM model to obtain the predicted usage data, which reduces the error of a single model and improves the accuracy of the prediction results.

[0067] S4. Optimize the billing rules of the charging model according to the predicted usage data to obtain an optimized cloud resource billing model.

[0068] It should be noted that in step S4, the charging model is adjusted or optimized according to the predicted usage data obtained in step S3 to obtain an optimized cloud resource billing model, which has the characteristics of elastic adjustment and optimization, and can flexibly adjust and improve the charging rules according to user feedback and changes in demand to adapt to the ever-changing market environment. In this embodiment, the optimized cloud resource billing method based on predicted cloud network resources can better understand and predict its cloud network resource needs, thereby avoiding the problem of insufficient resources or waste of resources, improving the utilization efficiency of cloud network resources, and reducing cloud service costs.

[0069] The present application provides an optimized cloud resource billing method based on predicted cloud network resources, including obtaining usage data of each device using the cloud network resources and charging information of the cloud network resources, the charging information including billing items, billing rules and business information; building a charging model based on the charging information and all usage data, the charging model outputs bill information corresponding to all usage data, and determines indicator data related to the cloud network resources from the bill information; obtaining historical sequence data of the cloud network resources corresponding to the indicator data, using an LSTM-ARMA model to predict the usage data based on the historical sequence data to obtain predicted usage data; optimizing the billing rules of the charging model based on the predicted usage data to obtain an optimized cloud resource billing model. The optimized cloud resource billing method based on predicted cloud network resources first builds a charging model according to the actual usage data and charging information of the cloud network resources, and then optimizes the charging model based on the predicted usage data of the future usage of the cloud network resources predicted by the indicator data, so as to obtain the optimized cloud resource billing model. The cloud resource billing model can adjust the billing rules according to the actual needs, so that the calculation of the cloud network resource usage fee is accurate and reasonable, which solves the technical problem that the traditional cloud resource billing method cannot predict the resources required in the future and has unreasonable cloud billing.

[0070] It should be noted that the optimized cloud resource billing method based on predicted cloud network resources can be applied in cloud computing environments to help users monitor and analyze cloud resource usage in real time and predict future cloud resource demand trends, which is very important for users in hybrid multi-cloud environments and can help users plan cloud resources more effectively and avoid emergency purchases and unnecessary cost increases due to insufficient resources. The optimized cloud resource billing method based on predicted cloud network resources can be used to optimize the use of cloud resources and help users reduce cloud computing costs through machine learning methods and integrated LSTM-ARMA model predictions, which is very valuable to users because users can optimize resource allocation based on prediction results (such as predicted usage data) to ensure efficient operation of the business environment and avoid resource waste, thereby achieving cost optimization and resource management.

[0071] In one embodiment of the present application, the usage data is predicted using the LSTM-ARMA model according to the historical sequence data, and the predicted usage data obtained includes:

[0072] Preprocess the historical sequence data to obtain stable time historical sequence data; and obtain the prediction duration;

[0073] According to the prediction duration and time history series data, the LSTM-ARMA model is used to predict the usage data to obtain the predicted usage data;

[0074] Among them, the LSTM-ARMA model is used to predict the usage data according to the prediction duration and time history series data, and the contents of the predicted usage data include:

[0075] The ARMA model of the LSTM-ARMA model is used to perform a stationarity test on the time history series data and obtain the test results;

[0076] If the test result shows that the time history series data is stable, the parameters of the ARMA model are estimated by using the maximum likelihood estimation to obtain the estimated parameters; the ARMA model is used to make predictions based on the estimated parameters and the time history series data to obtain the residual error history series data;

[0077] According to the prediction duration and the historical sequence data of the residual error, the LSTM model of the LSTM-ARMA model is used to perform sequence nonlinear capture prediction to obtain the predicted usage data.

[0078] It should be noted that the historical sequence data includes historical sequence data such as resource utilization rate, resource usage flow and resource usage capacity. In this embodiment, the abnormal data affected by factors such as trend, seasonality and noise are removed from the historical sequence data to obtain stable historical sequence data. The ARMA model of the LSTM-ARMA model first selects the order, and by observing the autocorrelation diagram (ACF) and partial autocorrelation diagram (PACF) formed by the time historical sequence data, it is determined whether the autoregressive order (p) and the sliding average order (q) are consistent with the selected order to achieve stationarity detection and obtain the detection result. The ARMA model is diagnosed by estimating parameters to predict the time historical sequence data based on the time historical sequence data, and whether the residual sequence in the prediction process conforms to the white noise assumed by the ARMA model is detected, thereby obtaining the residual error historical sequence data of nonlinear and non-stationary characteristics that the ARMA model cannot capture. After that, the residual error historical sequence data is input into the LSTM-ARMA model for sequence nonlinear capture prediction to obtain predicted usage data. Among them, because LSTM has memory units and gating mechanisms, it can effectively handle long-term dependencies and capture nonlinear patterns in sequences. The optimized cloud resource billing method based on predicted cloud network resources analyzes user usage patterns and market development trends, and continuously improves and optimizes the billing rules of the charging model according to predicted usage data to provide a more flexible and competitive charging plan or billing model.

[0079] In one embodiment of the present application, the billing rules of the charging model are optimized according to the predicted usage data, and the optimized cloud resource billing model includes:

[0080] Obtain the unit price of each billing mode of the billing rules, calculate according to the unit price of each billing mode, and obtain the optimization judgment data;

[0081] Compare the predicted usage data with the optimization judgment data to obtain a comparison result;

[0082] Determine the updated billing model corresponding to the cloud network resources based on the comparison results, and use the updated billing model to replace the billing model of the billing rules in the charging model to obtain an optimized cloud resource billing model;

[0083] Among them, if the predicted usage data is the predicted usage traffic, then based on the predicted usage traffic being greater than the optimization judgment data, the billing mode is updated to the bandwidth monthly billing mode; or based on the predicted usage traffic being less than the optimization judgment data, the billing mode is updated to the bandwidth on-demand billing mode; or based on the predicted usage traffic being equal to the optimization judgment data, the billing mode is updated to the bandwidth on-demand billing mode or the bandwidth monthly billing mode.

[0084] It should be noted that the units of each billing mode in the billing rules corresponding to the predicted usage data are first determined, and the judgment data is optimized based on the unit price calculation of all billing modes. Then, the updated billing mode is determined based on the comparison of the optimized judgment data and the predicted usage data. The charging model is optimized by updating the billing mode to achieve reasonable billing calculation for the use of cloud network resources, so as to reasonably allocate resources according to the cloud network resource billing of different indicator data and reduce the cost of resource use. In this embodiment, the predicted usage data is used as a case to illustrate the predicted usage traffic, and the bandwidth is flexibly configured according to the monthly billing mode and the bandwidth on-demand billing mode based on the predicted usage data. If a is the unit price of the on-demand billing mode and b is the unit price of the monthly billing mode, if the predicted usage traffic> a / b times the bandwidth, the monthly billing mode will be used. If the predicted usage traffic

[0085] In an embodiment of the present application, the optimized cloud resource billing method based on predicted cloud network resources can also be applied to count the number and details of newly added and unsubscribed machines in the month according to product attributes; for example, automatically define the annual budget of each department and product, output the machine usage sequence on a monthly basis, issue alarms for departments that exceed the sequence, manage new resource applications and cost allocation, and form a resource usage profile for each business.​

[0086] Embodiment 2:

[0087] Figure 2 This is a schematic diagram of the framework of the optimized cloud resource billing system based on predicted cloud network resources described in an embodiment of the present application.

[0088] like Figure 2 As shown, the embodiment of the present application provides an optimized cloud resource billing system based on predicted cloud network resources, including a data acquisition module 100, a model building module 200, a prediction data module 300 and a model optimization module 400;

[0089] The data acquisition module 100 is used to acquire the usage data of each device using the cloud network resources and the charging information of the cloud network resources, wherein the charging information includes billing items, billing rules and service information;

[0090] A model building module 200 is used to build a charging model based on the charging information and all the usage data, the charging model outputs billing information corresponding to all the usage data, and determines the indicator data related to the cloud network resources from the billing information;

[0091] The prediction data module 300 is used to obtain the historical sequence data of the cloud network resources corresponding to the indicator data, and use the LSTM-ARMA model to predict the usage data according to the historical sequence data to obtain the predicted usage data;

[0092] The model optimization module 400 is used to optimize the billing rules of the charging model according to the predicted usage data to obtain an optimized cloud resource billing model.

[0093] It should be noted that the content of the module in the optimized cloud resource billing system based on predicted cloud network resources corresponds to the step content of the optimized cloud resource billing method based on predicted cloud network resources, and the step content of the optimized cloud resource billing method based on predicted cloud network resources has been described in the first embodiment, and the content of the module in the optimized cloud resource billing system based on predicted cloud network resources will not be repeated in this embodiment. The optimized cloud resource billing system based on predicted cloud network resources cooperates with each other through the data acquisition module, the model construction module, the prediction data module and the model optimization module to achieve the setting of a reasonable billing mode based on the actual use of resources and the prediction of future use of resources.

[0094] In one embodiment of the present application, the prediction data module 300 includes a data processing submodule and a prediction submodule;

[0095] The data processing submodule is used to preprocess the historical sequence data to obtain stable time historical sequence data and obtain the predicted duration;

[0096] The prediction submodule is used to use the ARMA model of the LSTM-ARMA model to perform a stationarity test on the time history series data to obtain the test results; based on the test results that the time history series data is stationary, the parameters of the ARMA model are estimated by maximum likelihood estimation to obtain the estimated parameters; based on the estimated parameters and the time history series data, the ARMA model is used to perform prediction to obtain the residual error history series data; based on the prediction duration and the residual error history series data, the LSTM model of the LSTM-ARMA model is used to perform sequence nonlinear capture prediction to obtain the predicted usage data.

[0097] In one embodiment of the present application, the model optimization module 400 includes an optimization data acquisition submodule, a comparison submodule, and an optimization submodule;

[0098] The optimization data acquisition submodule is used to obtain the unit price of each billing mode of the billing rules, and obtain the optimization judgment data according to the unit price of each billing mode;

[0099] A comparison submodule is used to compare the predicted usage data with the optimization judgment data to obtain a comparison result;

[0100] The optimization submodule is used to determine the updated billing mode corresponding to the cloud network resources according to the comparison results, and use the updated billing mode to replace the billing mode of the billing rules in the charging model to obtain the optimized cloud resource billing model.

[0101] In one embodiment of the present application, the indicator data includes resource utilization rate, resource utilization flow and resource utilization capacity; if the predicted utilization data is the predicted utilization flow, then based on the predicted utilization flow being greater than the optimization judgment data, the billing mode is updated to a bandwidth monthly billing mode; or based on the predicted utilization flow being less than the optimization judgment data, the billing mode is updated to a bandwidth on-demand billing mode; or based on the predicted utilization flow being equal to the optimization judgment data, the billing mode is updated to a bandwidth on-demand billing mode or a bandwidth monthly billing mode.

[0102] Embodiment three:

[0103] Figure 3 A schematic diagram of a terminal device described in an embodiment of the present application.

[0104] like Figure 3 As shown, an embodiment of the present application provides a terminal device, including a processor and a memory;

[0105] A memory, used for storing program codes and transmitting the program codes to a processor;

[0106] A processor is used to execute the above-mentioned optimized cloud resource billing method based on predicted cloud network resources according to instructions in the program code.

[0107] It should be noted that the processor is used to execute the steps in the above-mentioned embodiment of the method for optimizing cloud resource billing based on predicting cloud network resources according to the instructions in the program code. Alternatively, the processor implements the functions of each module / unit in the above-mentioned system / device embodiments when executing the computer program.

[0108] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in a memory and executed by a processor to complete the present application. One or more modules / units may be a series of computer program instruction segments capable of completing a specific function, and the instruction segments are used to describe the execution process of the computer program in a terminal device.

[0109] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that this does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.

[0110] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (dSIC), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0111] The memory may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The memory may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device. Furthermore, the memory may include both an internal storage unit of the terminal device and an external storage device. The memory is used to store computer programs and other programs and data required by the terminal device. The memory may also be used to temporarily store data that has been output or is to be output.

[0112] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0113] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0114] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0115] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0116] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0117] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for optimizing cloud resource billing based on predicting cloud network resources, characterized in that: The following steps are involved: Obtaining usage data of each device using cloud network resources and charging information of the cloud network resources, wherein the charging information includes billing items, billing rules and service information; Building a charging model based on the charging information and all the usage data, the charging model outputs billing information corresponding to all the usage data, and determining indicator data related to the cloud network resources from the billing information; Acquire historical sequence data of the cloud network resources corresponding to the indicator data, and use the LSTM-ARMA model to predict usage data according to the historical sequence data to obtain predicted usage data; The billing rules of the charging model are optimized according to the predicted usage data to obtain an optimized cloud resource billing model.

2. The method for optimizing cloud resource billing based on predicting cloud network resources according to claim 1, characterized in that: The usage data is predicted using the LSTM-ARMA model based on the historical sequence data, and the predicted usage data includes: Preprocessing the historical sequence data to obtain stable time historical sequence data; and obtaining the predicted duration; The usage data is predicted using the LSTM-ARMA model according to the predicted duration and the time history series data to obtain predicted usage data.

3. The method for optimizing cloud resource billing based on predicting cloud network resources according to claim 2, characterized in that: The usage data is predicted using the LSTM-ARMA model according to the predicted duration and the time history series data, and the content of the predicted usage data obtained includes: Using the ARMA model of the LSTM-ARMA model to perform a stationarity test on the time history series data to obtain a test result; If the detection result shows that the time history series data is stable, the parameters of the ARMA model are estimated by using maximum likelihood estimation to obtain estimated parameters; the ARMA model is used to perform prediction based on the estimated parameters and the time history series data to obtain residual error history series data; According to the prediction duration and the residual error historical sequence data, the LSTM model of the LSTM-ARMA model is used to perform sequence nonlinear capture prediction to obtain predicted usage data.

4. The method for optimizing cloud resource billing based on predicted cloud network resources according to claim 1, characterized in that: The charging rules of the charging model are optimized according to the predicted usage data, and the optimized cloud resource charging model includes: Obtaining the unit price of each billing mode of the billing rule, and calculating according to the unit price of each billing mode to obtain optimization judgment data; Comparing the predicted usage data with the optimization judgment data to obtain a comparison result; An updated billing model corresponding to the cloud network resources is determined according to the comparison result, and the updated billing model is used to replace the billing model of the billing rules in the charging model to obtain an optimized cloud resource billing model.

5. The method for optimizing cloud resource billing based on predicting cloud network resources according to claim 4, characterized in that: The indicator data includes resource utilization rate, resource utilization flow and resource utilization capacity; if the predicted utilization data is predicted utilization flow, then according to the predicted utilization flow being greater than the optimization judgment data, the updated billing mode is selected as the bandwidth monthly billing mode; or according to the predicted utilization flow being less than the optimization judgment data, the updated billing mode is selected as the bandwidth on-demand billing mode; or according to the predicted utilization flow being equal to the optimization judgment data, the updated billing mode is selected as the bandwidth on-demand billing mode or the bandwidth monthly billing mode.

6. An optimized cloud resource billing system based on predicted cloud network resources, characterized in that: It includes data acquisition module, model building module, data prediction module and model optimization module; The data acquisition module is used to acquire usage data of each device using cloud network resources and charging information of the cloud network resources, wherein the charging information includes billing items, billing rules and service information; The model building module is used to build a charging model based on the charging information and all the usage data, the charging model outputs billing information corresponding to all the usage data, and determines the indicator data related to the cloud network resources from the billing information; The prediction data module is used to obtain the historical sequence data of the cloud network resources corresponding to the indicator data, and use the LSTM-ARMA model to predict the usage data according to the historical sequence data to obtain the predicted usage data; The model optimization module is used to optimize the billing rules of the charging model according to the predicted usage data to obtain an optimized cloud resource billing model.

7. The optimized cloud resource billing system based on predicted cloud network resources according to claim 6, characterized in that: The prediction data module includes a data processing submodule and a prediction submodule; The data processing submodule is used to preprocess the historical sequence data to obtain stable time historical sequence data; and to obtain the predicted duration; The prediction submodule is used to use the ARMA model of the LSTM-ARMA model to perform a stationarity test on the time history series data to obtain a test result; according to the test result that the time history series data is stationary, the parameters of the ARMA model are estimated by maximum likelihood estimation to obtain estimated parameters; According to the estimated parameters and the time history series data, the ARMA model is used to perform prediction to obtain the residual error history series data; according to the prediction duration and the residual error history series data, the LSTM model of the LSTM-ARMA model is used to perform sequence nonlinear capture prediction to obtain predicted usage data.

8. The optimized cloud resource billing system based on predicted cloud network resources according to claim 6, characterized in that: The model optimization module includes an optimization data acquisition submodule, a comparison submodule and an optimization submodule; The optimization data acquisition submodule is used to obtain the unit price of each billing mode of the billing rule, and obtain the optimization judgment data according to the calculation of each billing mode unit price; The comparison submodule is used to compare the predicted usage data with the optimization judgment data to obtain a comparison result; The optimization submodule is used to determine an updated billing model corresponding to the cloud network resources according to the comparison result, and use the updated billing model to replace the billing model of the billing rules in the charging model to obtain an optimized cloud resource billing model.

9. The optimized cloud resource billing system based on predicted cloud network resources according to claim 8, characterized in that: The indicator data includes resource utilization rate, resource utilization flow and resource utilization capacity; if the predicted utilization data is predicted utilization flow, then according to the predicted utilization flow being greater than the optimization judgment data, the updated billing mode is selected as the bandwidth monthly billing mode; or according to the predicted utilization flow being less than the optimization judgment data, the updated billing mode is selected as the bandwidth on-demand billing mode; or according to the predicted utilization flow being equal to the optimization judgment data, the updated billing mode is selected as the bandwidth on-demand billing mode or the bandwidth monthly billing mode.

10. A terminal device, characterized in that: including a processor and a memory; The memory is used to store program codes and transmit the program codes to the processor; The processor is used to execute the optimized cloud resource billing method based on predicted cloud network resources as described in any one of claims 1-5 according to the instructions in the program code.

Images