CN116266323A - Intelligent charging method and device in cloud scene - Google Patents
Intelligent charging method and device in cloud scene Download PDFInfo
- Publication number
- CN116266323A CN116266323A CN202211398994.8A CN202211398994A CN116266323A CN 116266323 A CN116266323 A CN 116266323A CN 202211398994 A CN202211398994 A CN 202211398994A CN 116266323 A CN116266323 A CN 116266323A
- Authority
- CN
- China
- Prior art keywords
- charging
- factor
- influence
- association
- factors
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0283—Price estimation or determination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent charging method and device in a cloud scene, wherein the method comprises the following steps: acquiring a training sample, and fitting according to the training sample to obtain a nonlinear fitting relation between a charging result of the cloud resource and a charging influence factor; wherein, the charging influence factor is an influence factor taking user resource preference as influence weight; selecting part or all of training samples, analyzing a first association influence factor in the charging influence factors, and providing the charging influence factors corresponding to the first association factors as options for users; and receiving at least one selected option of a user, obtaining a corresponding factor set according to the at least one selected option, and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set. By the method, the relationship between the user resource preference and the recommended cloud resource and charging thereof is accurately matched, and the user viscosity is improved.
Description
Technical Field
The invention relates to the technical field of cloud resource charging, in particular to an intelligent charging method, an intelligent charging device, a computing device and a computer readable storage medium in a cloud scene.
Background
Edge clouds are cloud computing platforms built on top of an edge infrastructure based on the core and edge computing capabilities of cloud computing technology. The edge cloud can reduce response time delay, cloud pressure and bandwidth cost by sinking network forwarding, storage, calculation and other resources to the user side, and can provide cloud services such as whole network scheduling, calculation and distribution. If the user experience is to be improved, the user viscosity is improved, the continuous profit of the edge cloud computing service provider is ensured, the robustness of cloud computing resource service which can be provided by the manufacturer is to be improved, the rationality of matching the user experience preference with the resource charge requirement is considered, and a novel intelligent charging method aiming at the edge cloud is to be researched.
Fig. 1 shows an edge cloud resource charging scheme architecture in the prior art, as shown in fig. 1, an overall flow of edge cloud resource charging of a cloud vendor in the prior art is as follows: the user selects the resource specification and the resource area, the resource scheduling module performs resource pre-scheduling according to the specification and the area selected by the user, and then the charging module performs charging according to the corresponding price of the resource specification and feeds back the charging result to the user. The charging modes of the edge cloud service mainly comprise the following steps: 1) Flat charging: namely, charging the package duration, and charging once according to the ordering duration, wherein the charging is irrelevant to the use amount of the user, and the difficulty is in the pricing link of the manufacturer and the rationality and fairness of the resource allocation when the user orders (the resources purchased by different users with the same price can be inconsistent in user experience due to the limitations of the resource scheduling of the manufacturer or network service and the like); 2) Metering according to the amount: according to the real-time charging of the user usage amount, the method meets the visual requirements and charging rules of the user, brings certain difficulty to cloud manufacturer charging and real-time dynamic resource allocation, and simultaneously lacks the rationality and fairness of resource allocation; 3) Package custom charging: the manufacturer can deeply customize different packages according to different resource and service requirements of users, resource slice management is realized among different packages, cloud manufacturer resource scheduling pressure can be relieved to a certain extent, the requirements of the users can be met, and the users can purchase according to the needs. However, with flat billing, metering and package-customized billing, there are the following disadvantages:
Flat charging is a charging mode requiring a user to pay in advance or post-pay for a fixed period of time, and has a problem of low flexibility. Different from a central cloud centralized resource deployment mode, the edge cloud resources are distributed on the edge side in a relatively distributed mode, the resource stock of each machine room node is relatively less, the machine room nodes are more, and when a user subscribes to the resources, the manufacturer preferentially schedules the resources of the nearest edge nodes of the user. When the nearest side resource is insufficient, the resources of the nodes of the far side edge machine room are scheduled, and the problem that the user purchases the resources and service experiences are inconsistent under the same expense possibly exists, even if the manufacturer charging logic has a differential charging mechanism aiming at the situation that the resources are not matched with the user requirements, the phenomena of resource waste and unreasonable charging exist for the user when the original purpose of the user resource requirements is overcome. Compared with the charging mode of the packet duration, the charging mode of the per-volume charging improves certain flexibility, relieves the phenomenon of user resource waste, and still has the phenomenon of lacking resource allocation rationality and fairness. Compared with the two charging modes, the package customized charging is characterized in that differentiated charging is carried out on the user to a certain extent on the premise of considering the purchase intention of the user, but the quantifiable charging factors considered by the mode are on one side, the flexibility is low, the phenomena of additional product expenditure and the like of the user exist, and the problems of mismatching of the consumption tendency of the user, the resource use preference and the resource allocation and charging of cloud manufacturers still exist.
Disclosure of Invention
The present invention has been made in view of the above problems, and provides an intelligent charging method, apparatus, computing device, and computer-readable storage medium in a cloud scenario that overcomes or at least partially solves the above problems.
According to one aspect of the present invention, there is provided an intelligent charging method in a cloud scenario, the method comprising:
acquiring a training sample, and fitting according to the training sample to obtain a nonlinear fitting relation between a charging result of the cloud resource and a charging influence factor; wherein, the charging influence factor is an influence factor taking user resource preference as influence weight;
selecting part or all of training samples, analyzing a first association influence factor in the charging influence factors, and providing the charging influence factors corresponding to the first association factors as options for users;
and receiving at least one selected option of a user, obtaining a corresponding factor set according to the at least one selected option, and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
Optionally, the selecting part or all of the training samples, and analyzing the first association influence factor in the charging influence factors further includes:
Selecting part or all of the training samples;
according to the selected training sample and likelihood formula, calculating the average likelihood value corresponding to the selected training sample, and calculating the strong correlation coefficient corresponding to different charging influence factor combinations;
and according to the strong association coefficients corresponding to the different charging influence factor combinations, analyzing a first association influence factor in the charging influence factors.
Optionally, the selecting part or all of the training samples, and analyzing the first association influence factor in the charging influence factors further includes:
step S1, t training samples are selected, wherein t is less than or equal to k, and k is the total number of the training samples; assigning n as m, wherein m is the total number of charging influence factors;
s2, acquiring a charging influence factor combination, and calculating average likelihood values corresponding to t training samples according to the t training samples and a likelihood formula;
step S3, calculating strong association coefficients corresponding to the n charging influence factors;
step S4, extracting n-1 charging influence factors from n charging influence factors without repetition, wherein each n-1 charging influence factors are 1 group of charging influence factor combinations, obtaining average likelihood values under the n groups of charging influence factor combinations according to the step S2, sequencing the n average likelihood values, selecting the charging influence factor combination corresponding to the largest average likelihood value, and assigning n as n-1; iteratively executing the steps S2 to S4 until the iteration times meet the conditions and are finished;
And S5, selecting a charging influence factor combination corresponding to the minimum value of the strong association coefficient from the calculation result as a first association influence factor.
Optionally, after the analyzing the first associated influence factor of the charging influence factors, the method further comprises:
acquiring a second association influence factor except the first association influence factor in the charging influence factors;
calculating a correlation coefficient between the second correlation influencing factor and the first correlation influencing factor;
and dividing the second association influence factor into an association hidden factor and a non-association hidden factor according to the correlation coefficient.
Optionally, the calculating a correlation coefficient between the second correlation influencing factor and the first correlation influencing factor specifically includes: a Pearson correlation coefficient between the second correlation influencing factor and the first correlation influencing factor is calculated.
Optionally, the classifying the second association influencing factor into the association hiding factor and the non-association hiding factor according to the correlation coefficient specifically includes:
and taking the second association influence factor with the correlation coefficient larger than the preset value as an association hiding factor, and taking the second association influence factor with the correlation coefficient smaller than or equal to the preset value as an unassociated hiding factor.
Optionally, the receiving at least one option selected by the user, obtaining a corresponding factor set according to the selected at least one option, and outputting the charging scheme for recommending to the user by using the nonlinear fitting relation and the factor set further includes:
receiving at least one selected option selected by a user, acquiring a first associated influence factor and an influence weight thereof, associated hiding factors and influence weights thereof corresponding to the selected at least one option, and combining unassociated hiding factors to obtain a factor set;
and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
According to another aspect of the present invention, there is provided an intelligent charging apparatus in a cloud scenario, the apparatus including:
the training module is used for acquiring training samples, and obtaining a nonlinear fitting relation between a charging result of the cloud resource and a charging influence factor according to fitting of the training samples; wherein, the charging influence factor is an influence factor taking user resource preference as influence weight;
the analysis module is used for selecting part or all of the training samples, analyzing a first association influence factor in the charging influence factors, and providing the charging influence factors corresponding to the first association factors as options for users;
And the charging module is used for receiving at least one selected option selected by a user, obtaining a corresponding factor set according to the selected at least one selected option, and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
Optionally, the analysis module is further configured to:
selecting part or all of the training samples;
according to the selected training sample and likelihood formula, calculating the average likelihood value corresponding to the selected training sample, and calculating the strong correlation coefficient corresponding to different charging influence factor combinations;
and according to the strong association coefficients corresponding to the different charging influence factor combinations, analyzing a first association influence factor in the charging influence factors.
Optionally, the analyzing module is further configured to analyze a first association influence factor of the charging influence factors by:
step S1, t training samples are selected, wherein t is less than or equal to k, and k is the total number of the training samples; assigning n as m, wherein m is the total number of charging influence factors;
s2, acquiring a charging influence factor combination, and calculating average likelihood values corresponding to t training samples according to the t training samples and a likelihood formula;
Step S3, calculating strong association coefficients corresponding to the n charging influence factors;
step S4, extracting n-1 charging influence factors from n charging influence factors without repetition, wherein each n-1 charging influence factors are 1 group of charging influence factor combinations, obtaining average likelihood values under the n groups of charging influence factor combinations according to the step S2, sequencing the n average likelihood values, selecting the charging influence factor combination corresponding to the largest average likelihood value, and assigning n as n-1; iteratively executing the steps S2 to S4 until the iteration times meet the conditions and are finished;
and S5, selecting a charging influence factor combination corresponding to the minimum value of the strong association coefficient from the calculation result as a first association influence factor.
Optionally, the analysis module is further configured to:
acquiring a second association influence factor except the first association influence factor in the charging influence factors;
calculating a correlation coefficient between the second correlation influencing factor and the first correlation influencing factor;
and dividing the second association influence factor into an association hidden factor and a non-association hidden factor according to the correlation coefficient.
Optionally, the analysis module is specifically configured to: a Pearson correlation coefficient between the second correlation influencing factor and the first correlation influencing factor is calculated.
Optionally, the analysis module is specifically configured to: and taking the second association influence factor with the correlation coefficient larger than the preset value as an association hiding factor, and taking the second association influence factor with the correlation coefficient smaller than or equal to the preset value as an unassociated hiding factor.
Optionally, the charging module is further configured to:
receiving at least one selected option selected by a user, acquiring a first associated influence factor and an influence weight thereof, associated hiding factors and influence weights thereof corresponding to the selected at least one option, and combining unassociated hiding factors to obtain a factor set;
and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
According to yet another aspect of the present invention, there is provided a computing device comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the intelligent charging method in the cloud scene.
According to still another aspect of the present invention, there is provided a computer storage medium having at least one executable instruction stored therein, the executable instruction causing a processor to perform operations corresponding to the intelligent charging method in the cloud scenario described above.
According to the scheme provided by the invention, the relation between the factors influencing the resource pricing and the cloud node and the resource charging in the cloud scene is quantized, and the user selected resource and reasonable charging can be accurately recommended by utilizing the quantization result influencing the resource pricing and the fitted nonlinear fitting relation between the user resource preference and the cloud resource node and the cloud resource charging. The invention takes the user resource preference as the weight coefficient for influencing the resource charging, and further carries out the resource charging calculation according to the analyzed user selectable trend factor, thereby being capable of further accurately matching the relationship between the user resource preference and the recommended cloud resource and the charging thereof and being beneficial to improving the user viscosity.
The foregoing description is only an overview of the present invention, and is intended to be implemented in accordance with the teachings of the present invention in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present invention more readily apparent.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
FIG. 1 illustrates a prior art edge cloud resource billing scheme architecture;
FIG. 2 illustrates a flow chart of an intelligent charging method in a cloud scenario of one embodiment of the present invention;
FIG. 3 is a flow chart illustrating a method of acquiring a first associated influence factor according to one embodiment of the invention;
fig. 4 shows a schematic diagram of an intelligent charging method in a cloud scenario according to another embodiment of the present invention;
fig. 5 is a schematic structural diagram of an intelligent charging apparatus in a cloud scenario according to another embodiment of the present invention;
FIG. 6 illustrates a schematic diagram of a computing device embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The inventor of the present application finds that the charging mode of the edge cloud service in the prior art has a common problem: the user resource demand preference and the paid corresponding resource cost are not matched with the cloud manufacturer resource allocation. The cloud manufacturer has the dominant right in the aspects of resource charging and resource allocation, and the problems of consumption and resource use tendency of users are weakened, so that an intelligent charging mode in an edge cloud scene is necessary to be researched so as to balance the relation between the user resource use preference and the edge cloud manufacturer resource allocation. The intelligent charging method in the cloud scene is particularly suitable for the edge cloud scene by taking mismatching among user resource use requirements, consumption tendency and resource scheduling of edge cloud manufacturers and resource charging as an access point.
Fig. 2 shows a flow chart of an intelligent charging method in a cloud scenario according to an embodiment of the present invention. As shown in fig. 2, the method comprises the steps of:
step 210: acquiring a training sample, and fitting according to the training sample to obtain a nonlinear fitting relation between a charging result of the cloud resource and a charging influence factor; wherein the charging influence factor is an influence factor taking user resource preference as influence weight.
In the embodiment of the invention, the charging influence factor influencing the resource charging in the edge cloud scene is thinned, the charging influence factor is the influence factor taking the user resource preference as the influence weight, and the nonlinear fitting relation between the charging result of the cloud resource and the charging influence factor is obtained through training fitting, so that the user will, the cloud resource allocation and the charging are formulated, and the problem that the vendor resource allocation and the charging in the edge cloud scene are not matched with the user preference is solved. The following describes an example of a cloud host in an edge cloud scenario.
Specifically, define g i Pricing for the ith cloud host resource (sample) under a plurality of charging influence factors, wherein i is more than or equal to 1 and less than or equal to k, and k is the total number of samples of training samples, namely the total number of cloud hosts; defining cost factor vectors such as rent or electricity charge of the place where the edge node machine room is located as 
Yun Anshang cloud host CPU has a computing performance of +.>
The memory read-write speed of the cloud host is +.>
The magnetic disk read/write (I/O) capability is +.>
The average network environment between the user and the cloud host resource due to the distance and the like is +.>
And other refineable factors affecting cloud host resource pricing +.>
Namely, the charging influence factor influencing the pricing of the cloud host resources is +.>
The total factor is m, then it is:
marketing pricing is carried out on cloud host resources according to m charging influence factors (marketing pricing of different types of cloud hosts under influence of all factors can be the same), standard vectors of all training samples are represented by G, and G= (G) 1 ,g 2 ,...,g i ,...,g k ) Where k is the number of samples and the corresponding original billing impact factor vector is
In the embodiment of the invention, in order to balance the resource allocation relation between the user resource use preference and the edge cloud manufacturer, a grade vector epsilon= { epsilon capable of reflecting the user resource preference to a certain extent is further introduced 1 ,ε 2 ，ε 3 ,...,ε m And (3) performing the following calculation to obtain a charging influence factor A taking the user resource preference as influence weight:
and using G and A as training samples, and fitting a nonlinear mapping relation between tariffs G and charging influence factors a under each resource pool by using a nonlinear fitting function and a system identification method, namely:
g＝θ T a＝θ 1 a 1 +θ 12 a 2 +...+θ m a m (3)
Step 220: and selecting part or all of the training samples, analyzing a first association influence factor in the charging influence factors, and providing the charging influence factors corresponding to the first association factors as options for users.
In order to quantify user resource preference and obtain the influence of user subscription intention on final charging, the embodiment of the invention analyzes a strong correlation factor which influences charging, namely a first correlation influence factor, and provides the charging influence factor corresponding to the first correlation factor as a selection item for the user to select the intention when the user subscribes.
In an alternative embodiment, step 220 further comprises: selecting part or all of the training samples; according to the selected training sample and likelihood formula, calculating the average likelihood value corresponding to the selected training sample, and calculating the strong correlation coefficient corresponding to different charging influence factor combinations; and according to the strong association coefficients corresponding to the different charging influence factor combinations, analyzing a first association influence factor in the charging influence factors.
To reduce the amount of computation, a portion of the training samples may be selected from the training samples described above for the computation of step 220. For example, t training samples G are selected from G and A * And A * Referring to fig. 3, the following method is performed:
step S1, t training samples are selected, wherein t is less than or equal to k, and k is the total number of the training samples; and (3) assigning n as m, wherein m is the total number of the charging influence factors.
And S2, acquiring a charging influence factor combination, and calculating average likelihood values corresponding to t training samples according to the t training samples and the likelihood formula.
Set the selected training sample (G * ,A * ) The likelihood formula used is as follows:
based on the obtained combinations of the billing influence factors (the number of factors in the combinations decreases as the number of iterations increases), a training sample (G * ,A * ) And the likelihood values under t training samples are calculated according to the formula (4) and are marked as (L) 1 ,L 2 ,...,L t ) Further deriving the average likelihood values of t training samples:
and S3, calculating strong correlation coefficients corresponding to the n charging influence factors.
L calculated according to step S2 average Calculating strong association coefficients AIC corresponding to n charging influence factors n The calculation formula is as follows:
AIC n ＝2n-2ln(L average ) (6)
step S4, extracting n-1 charging influence factors from n charging influence factors without repetition, wherein each n-1 charging influence factors are 1 group of charging influence factor combinations, obtaining average likelihood values under the n groups of charging influence factor combinations according to the step S2, sequencing the n average likelihood values, selecting the charging influence factor combination corresponding to the largest average likelihood value, and assigning n as n-1; and iteratively executing the steps S2 to S4 until the iteration times meet the conditions.
Specifically, n-1 charging influence factors are not repeatedly extracted from n charging influence factors, each n-1 charging influence factors is 1 group of charging influence factor combination, and n groups of charging influence factor combinations can be obtained by extracting n times. And (2) repeatedly executing the step (S2) to obtain average likelihood values under n groups of charging influence factor combinations, namely obtaining n average likelihood values, sequencing the n average likelihood values, selecting a charging influence factor combination corresponding to the maximum average likelihood value, and taking the charging influence factor combination corresponding to the maximum average value as a parameter set for the next iteration execution.
And iteratively executing the steps S2 to S4 until the iteration times meet the conditions. Setting the threshold value as beta, performing iteration (m-beta) times, and ending the iteration process.
And S5, selecting a charging influence factor combination corresponding to the minimum value of the strong association coefficient from the calculation result as a first association influence factor.
Selecting the minimum AIC from the m-beta times of calculation results n Charging influence factor combination A corresponding to value β As a first association influencing factor, i.e. a set of strong association factors influencing charging. And providing the charging influence factors corresponding to the strong association factor set as selection items for the user to select when the user subscribes.
Further, a second association influence factor except the first association influence factor in the charging influence factors is obtained; calculating a correlation coefficient between the second correlation influencing factor and the first correlation influencing factor; and classifying the second association influence factor into an association hidden factor and a non-association hidden factor according to the correlation coefficient.
In order to strengthen the influence of the user resource preference weight on the charging result, a strong association hidden factor set of the strong association factor set needs to be obtained. Specifically, definition of the corpus A does not belong to A β Complement C of (C) A A β (second correlation influencing factor), calculating C by pearson correlation coefficient method A A β Factor A of (2) β The pearson correlation coefficient R between the factors in (1) is defined, a preset value of 0.8 is defined, and the factors with the phase relation number larger than 0.8 are defined as the correlation hiding factors, so that a strong correlation hiding factor set A is obtained R Note that the strong association attribute pair is (a) x ,a y ) { x, y=1, 2, 3..m, x+.y }, number e. A factor with a correlation coefficient less than or equal to 0.8 is defined as an uncorrelated concealment factor, i.e. the other (m- β -e) influencing factors are taken as fixed concealment terms.
Step 230: and receiving at least one selected option of a user, obtaining a corresponding factor set according to the at least one selected option, and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
Receiving at least one selected option selected by a user, acquiring a first associated influence factor and an influence weight thereof, associated hiding factors and influence weights thereof corresponding to the selected at least one option, and combining unassociated hiding factors to obtain a factor set; and outputting the charging scheme recommended to the user by using the nonlinear fitting relation and the factor set.
The user selects the selection item of the resource preference, the grade vector epsilon is obtained according to the selection of the user, and the grade vector epsilon is used as a strong association factor set corresponding to the resource use preference selected by the user and a corresponding strong association hiding factor set weight coefficient, and the nonlinear fitting relation g=theta is adopted in combination with other hiding factor sets T and a, recommending a cloud resource charging scheme which accords with the value level of the resource selected by the user, and carrying out resource pre-scheduling and distribution for the user by a cloud manufacturer according to the user selection resource preference and charging result.
Fig. 4 shows a schematic diagram of an intelligent charging method in a cloud scenario according to another embodiment of the present invention. As shown in fig. 4, compared with fig. 1, different from the edge cloud resource charging method in the prior art, the method is that firstly, according to the resource preference selected by the user and other non-selectable invisible requirements, the best charging result is obtained through the fitting relation between the influence factors and the resource charging, the cloud manufacturer performs edge cloud node selection and resource scheduling according to the charging result and the user requirement trend, and the cloud resource charging is performed before the cloud resource scheduling. Similarly, the charging method can be applied to intelligent charging scenes of other cloud resources such as an edge cloud network, edge cloud storage and the like.
In the embodiment of the invention, the relation between the factors influencing the resource pricing and the edge cloud nodes and the resource charging in the edge cloud scene is quantized, and the user selected resources and reasonable charging can be accurately recommended by utilizing the quantization result influencing the resource pricing and the nonlinear fitting relation between the fitted user resource preference and the edge cloud resource nodes and the cloud resource charging. The above embodiment of the present invention uses the user resource preference as the weight coefficient for influencing the resource charging, and further obtains the user selectable trend factor A according to the above step 220 β And non-optional stealth factor A R And the resource charging calculation is carried out, so that the relationship between the user resource preference and the recommended edge cloud resource and the charging of the edge cloud resource can be further accurately matched, and the user viscosity can be improved.
Compared with the scheme that the existing edge cloud manufacturer simply performs resource pre-scheduling and distribution for users according to cloud resource types selected by the users, and then performs flattening, quantitative or package customized charging on the resources, the scheme is dominant by the user side, recommends the resources and charging results thereof for the users according to the user side selected resource use preference and consumption tendency, and then the cloud manufacturer performs resource pre-scheduling and distribution according to user requirements and charging results, so that a certain selection right and a certain dominant right are given to the users to a certain extent.
Compared with the existing edge cloud resource charging method, the intelligent charging method provided by the embodiment of the invention has the advantages of higher flexibility and accuracy of resource charging, and is fit with the user resource use requirement and charging rationality under the user perception. The method is smoother to modify on the basis of the existing charging method, namely, the direct selection of the user is changed into the resource charging and resource recommendation result calculated by the intelligent charging and utilizing the subscription selection item of the user.
The embodiment of the invention quantifies the factors affecting the resource charging in the edge cloud, and can control the cloud resource cost and profit more finely. The method not only realizes the fine control of cost and profit in manufacturers to a certain extent, but also ensures fairness between user resource use experience and actual charging, and is also beneficial to improving user viscosity of cloud manufacturers.
Fig. 5 shows a schematic structural diagram of an intelligent charging apparatus in a cloud scenario according to another embodiment of the present invention. As shown in fig. 5, the apparatus includes:
the training module 510 is configured to obtain a training sample, and obtain a nonlinear fitting relationship between a charging result of the cloud resource and a charging influence factor according to fitting of the training sample; wherein, the charging influence factor is an influence factor taking user resource preference as influence weight;
The analysis module 520 is configured to select part or all of the training samples, analyze a first association influence factor of the charging influence factors, and provide the charging influence factor corresponding to the first association factor as an option to the user;
and the charging module 530 is configured to receive at least one option selected by a user, obtain a corresponding factor set according to the at least one selected option, and output a charging scheme for recommending to the user by using the nonlinear fitting relationship and the factor set.
Optionally, the analysis module is further configured to:
selecting part or all of the training samples;
according to the selected training sample and likelihood formula, calculating the average likelihood value corresponding to the selected training sample, and calculating the strong correlation coefficient corresponding to different charging influence factor combinations;
and according to the strong association coefficients corresponding to the different charging influence factor combinations, analyzing a first association influence factor in the charging influence factors.
Optionally, the analyzing module is further configured to analyze a first association influence factor of the charging influence factors by:
step S1, t training samples are selected, wherein t is less than or equal to k, and k is the total number of the training samples; assigning n as m, wherein m is the total number of charging influence factors;
S2, acquiring a charging influence factor combination, and calculating average likelihood values corresponding to t training samples according to the t training samples and a likelihood formula;
step S3, calculating strong association coefficients corresponding to the n charging influence factors;
step S4, extracting n-1 charging influence factors from n charging influence factors without repetition, wherein each n-1 charging influence factors are 1 group of charging influence factor combinations, obtaining average likelihood values under the n groups of charging influence factor combinations according to the step S2, sequencing the n average likelihood values, selecting the charging influence factor combination corresponding to the largest average likelihood value, and assigning n as n-1; iteratively executing the steps S2 to S4 until the iteration times meet the conditions and are finished;
and S5, selecting a charging influence factor combination corresponding to the minimum value of the strong association coefficient from the calculation result as a first association influence factor.
Optionally, the analysis module is further configured to:
acquiring a second association influence factor except the first association influence factor in the charging influence factors;
calculating a correlation coefficient between the second correlation influencing factor and the first correlation influencing factor;
and dividing the second association influence factor into an association hidden factor and a non-association hidden factor according to the correlation coefficient.
Optionally, the analysis module is specifically configured to: a Pearson correlation coefficient between the second correlation influencing factor and the first correlation influencing factor is calculated.
Optionally, the analysis module is specifically configured to: and taking the second association influence factor with the correlation coefficient larger than the preset value as an association hiding factor, and taking the second association influence factor with the correlation coefficient smaller than or equal to the preset value as an unassociated hiding factor.
Optionally, the charging module is further configured to:
receiving at least one selected option selected by a user, acquiring a first associated influence factor and an influence weight thereof, associated hiding factors and influence weights thereof corresponding to the selected at least one option, and combining unassociated hiding factors to obtain a factor set;
and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
According to the device provided by the invention, the relation between the factors influencing the resource pricing and the cloud node and the resource charging in the cloud scene is quantized, and the user selected resource and reasonable charging can be accurately recommended by utilizing the quantization result influencing the resource pricing and the fitted nonlinear fitting relation between the user resource preference, the edge cloud resource node and the cloud resource charging. The device of the invention uses the user resource preference as the weight coefficient for influencing the resource charging, and further carries out the resource charging calculation according to the analyzed user selectable trend factor, thereby being capable of further accurately matching the relationship between the user resource preference and the recommended edge cloud resource and the charging thereof, and being beneficial to improving the user viscosity.
The embodiment of the invention provides a non-volatile computer storage medium, which stores at least one executable instruction, and the computer executable instruction can execute the intelligent charging method in the cloud scene in any method embodiment.
FIG. 6 illustrates a schematic diagram of an embodiment of a computing device of the present invention, and the embodiments of the present invention are not limited to a particular implementation of the computing device.
As shown in fig. 6, the computing device may include: a processor 602, a communication interface (Communications Interface), a memory 606, and a communication bus 608.
Wherein: processor 602, communication interface 604, and memory 606 perform communication with each other via communication bus 608. Communication interface 604 is used to communicate with network elements of other devices, such as clients or other servers. The processor 602 is configured to execute the program 610, and may specifically perform relevant steps in the foregoing embodiment of the intelligent charging method in a cloud scenario for a computing device.
In particular, program 610 may include program code including computer-operating instructions.
The processor 602 may be a central processing unit CPU or a specific integrated circuit ASIC (Application Specific Integrated Circuit) or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included by the computing device may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.
A memory 606 for storing a program 610. The memory 606 may comprise high-speed RAM memory or may further comprise non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 610 is for causing the processor 602 to:
acquiring a training sample, and fitting according to the training sample to obtain a nonlinear fitting relation between a charging result of the cloud resource and a charging influence factor; wherein, the charging influence factor is an influence factor taking user resource preference as influence weight;
selecting part or all of training samples, analyzing a first association influence factor in the charging influence factors, and providing the charging influence factors corresponding to the first association factors as options for users;
and receiving at least one selected option of a user, obtaining a corresponding factor set according to the at least one selected option, and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
In an alternative, the program 610 causes the processor to:
selecting part or all of the training samples;
according to the selected training sample and likelihood formula, calculating the average likelihood value corresponding to the selected training sample, and calculating the strong correlation coefficient corresponding to different charging influence factor combinations;
And according to the strong association coefficients corresponding to the different charging influence factor combinations, analyzing a first association influence factor in the charging influence factors.
In an alternative, the program 610 causes the processor to:
step S1, t training samples are selected, wherein t is less than or equal to k, and k is the total number of the training samples; assigning n as m, wherein m is the total number of charging influence factors;
s2, acquiring a charging influence factor combination, and calculating average likelihood values corresponding to t training samples according to the t training samples and a likelihood formula;
step S3, calculating strong association coefficients corresponding to the n charging influence factors;
step S4, extracting n-1 charging influence factors from n charging influence factors without repetition, wherein each n-1 charging influence factors are 1 group of charging influence factor combinations, obtaining average likelihood values under the n groups of charging influence factor combinations according to the step S2, sequencing the n average likelihood values, selecting the charging influence factor combination corresponding to the largest average likelihood value, and assigning n as n-1; iteratively executing the steps S2 to S4 until the iteration times meet the conditions and are finished;
and S5, selecting a charging influence factor combination corresponding to the minimum value of the strong association coefficient from the calculation result as a first association influence factor.
In an alternative, the program 610 causes the processor to:
acquiring a second association influence factor except the first association influence factor in the charging influence factors;
calculating a correlation coefficient between the second correlation influencing factor and the first correlation influencing factor;
and dividing the second association influence factor into an association hidden factor and a non-association hidden factor according to the correlation coefficient.
In an alternative, the program 610 causes the processor to:
a Pearson correlation coefficient between the second correlation influencing factor and the first correlation influencing factor is calculated.
In an alternative, the program 610 causes the processor to:
and taking the second association influence factor with the correlation coefficient larger than the preset value as an association hiding factor, and taking the second association influence factor with the correlation coefficient smaller than or equal to the preset value as an unassociated hiding factor.
In an alternative, the program 610 causes the processor to:
receiving at least one selected option selected by a user, acquiring a first associated influence factor and an influence weight thereof, associated hiding factors and influence weights thereof corresponding to the selected at least one option, and combining unassociated hiding factors to obtain a factor set;
And outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general-purpose systems may also be used with the teachings herein. The required structure for a construction of such a system is apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It will be appreciated that the teachings of the present invention described herein may be implemented in a variety of programming languages, and the above description of specific languages is provided for disclosure of enablement and best mode of the present invention.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the above description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be construed as reflecting the intention that: i.e., the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component and, furthermore, they may be divided into a plurality of sub-modules or sub-units or sub-components. Any combination of all features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or units of any method or apparatus so disclosed, may be used in combination, except insofar as at least some of such features and/or processes or units are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features but not others included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.
Various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of the functionality of some or all of the components according to embodiments of the present invention may be implemented in practice using a microprocessor or Digital Signal Processor (DSP). The present invention can also be implemented as an apparatus or device program (e.g., a computer program and a computer program product) for performing a portion or all of the methods described herein. Such a program embodying the present invention may be stored on a computer readable medium, or may have the form of one or more signals. Such signals may be downloaded from an internet website, provided on a carrier signal, or provided in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specifically stated.
Claims (10)
1. An intelligent charging method in a cloud scene is characterized by comprising the following steps:
acquiring a training sample, and fitting according to the training sample to obtain a nonlinear fitting relation between a charging result of the cloud resource and a charging influence factor; wherein, the charging influence factor is an influence factor taking user resource preference as influence weight;
selecting part or all of training samples, analyzing a first association influence factor in the charging influence factors, and providing the charging influence factors corresponding to the first association factors as options for users;
and receiving at least one selected option of a user, obtaining a corresponding factor set according to the at least one selected option, and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
2. The method of claim 1, wherein selecting some or all of the training samples, analyzing a first associated one of the billing impact factors further comprises:
selecting part or all of the training samples;
according to the selected training sample and likelihood formula, calculating the average likelihood value corresponding to the selected training sample, and calculating the strong correlation coefficient corresponding to different charging influence factor combinations;
And according to the strong association coefficients corresponding to the different charging influence factor combinations, analyzing a first association influence factor in the charging influence factors.
3. The method according to claim 1 or 2, wherein selecting part or all of the training samples, analyzing a first associated one of the billing impact factors further comprises:
step S1, t training samples are selected, wherein t is less than or equal to k, and k is the total number of the training samples; assigning n as m, wherein m is the total number of charging influence factors;
s2, acquiring a charging influence factor combination, and calculating average likelihood values corresponding to t training samples according to the t training samples and a likelihood formula;
step S3, calculating strong association coefficients corresponding to the n charging influence factors;
step S4, extracting n-1 charging influence factors from n charging influence factors without repetition, wherein each n-1 charging influence factors are 1 group of charging influence factor combinations, obtaining average likelihood values under the n groups of charging influence factor combinations according to the step S2, sequencing the n average likelihood values, selecting the charging influence factor combination corresponding to the largest average likelihood value, and assigning n as n-1; iteratively executing the steps S2 to S4 until the iteration times meet the conditions and are finished;
And S5, selecting a charging influence factor combination corresponding to the minimum value of the strong association coefficient from the calculation result as a first association influence factor.
4. The method according to claim 1, wherein after said analyzing a first associated one of said charging influencing factors, the method further comprises:
acquiring a second association influence factor except the first association influence factor in the charging influence factors;
calculating a correlation coefficient between the second correlation influencing factor and the first correlation influencing factor;
and dividing the second association influence factor into an association hidden factor and a non-association hidden factor according to the correlation coefficient.
5. The method according to claim 4, wherein calculating the correlation coefficient between the second correlation influencing factor and the first correlation influencing factor is specifically: a Pearson correlation coefficient between the second correlation influencing factor and the first correlation influencing factor is calculated.
6. The method according to claim 4, wherein the classifying the second association influencing factor into the association hiding factor and the non-association hiding factor according to the correlation coefficient is specifically:
and taking the second association influence factor with the correlation coefficient larger than the preset value as an association hiding factor, and taking the second association influence factor with the correlation coefficient smaller than or equal to the preset value as an unassociated hiding factor.
7. The method of claim 4, wherein the receiving the at least one user-selected option, deriving a corresponding set of factors from the at least one user-selected option, and outputting a billing plan for recommendation to the user using the nonlinear fit relationship and the set of factors further comprises:
receiving at least one selected option selected by a user, acquiring a first associated influence factor and an influence weight thereof, associated hiding factors and influence weights thereof corresponding to the selected at least one option, and combining unassociated hiding factors to obtain a factor set;
and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
8. An intelligent charging device under cloud scene, which is characterized in that the device comprises:
the training module is used for acquiring training samples, and obtaining a nonlinear fitting relation between a charging result of the cloud resource and a charging influence factor according to fitting of the training samples; wherein, the charging influence factor is an influence factor taking user resource preference as influence weight;
the analysis module is used for selecting part or all of the training samples, analyzing a first association influence factor in the charging influence factors, and providing the charging influence factors corresponding to the first association factors as options for users;
And the charging module is used for receiving at least one selected option selected by a user, obtaining a corresponding factor set according to the selected at least one selected option, and outputting a charging scheme recommended to the user by utilizing the nonlinear fitting relation and the factor set.
9. A computing device, comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;
the memory is configured to store at least one executable instruction, where the executable instruction causes the processor to perform an operation corresponding to the intelligent charging method in the cloud scenario according to any one of claims 1-7.
10. A computer storage medium having stored therein at least one executable instruction that causes a processor to perform operations corresponding to the intelligent charging method in a cloud scenario according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211398994.8A CN116266323A (en) | 2022-11-09 | 2022-11-09 | Intelligent charging method and device in cloud scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211398994.8A CN116266323A (en) | 2022-11-09 | 2022-11-09 | Intelligent charging method and device in cloud scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116266323A true CN116266323A (en) | 2023-06-20 |
Family
ID=86744213
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211398994.8A Pending CN116266323A (en) | 2022-11-09 | 2022-11-09 | Intelligent charging method and device in cloud scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116266323A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117499313A (en) * | 2024-01-02 | 2024-02-02 | 中移(苏州)软件技术有限公司 | Request control method, device, storage medium and electronic equipment |
-
2022
- 2022-11-09 CN CN202211398994.8A patent/CN116266323A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117499313A (en) * | 2024-01-02 | 2024-02-02 | 中移(苏州)软件技术有限公司 | Request control method, device, storage medium and electronic equipment |
| CN117499313B (en) * | 2024-01-02 | 2024-05-03 | 中移(苏州)软件技术有限公司 | Request control method, device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Samimi et al. | A combinatorial double auction resource allocation model in cloud computing | |
| Toosi et al. | Revenue maximization with optimal capacity control in infrastructure as a service cloud markets | |
| CN109714797B (en) | Mobile edge network resource allocation method based on auction theory | |
| CN111506849A (en) | Page generation method and device | |
| CN111626767B (en) | Resource data issuing method, device and equipment | |
| US8832694B2 (en) | Method and system for the dynamic allocation of resources based on a multi-phase negotiation mechanism | |
| CN110111189B (en) | Online combined resource allocation and payment method based on double-sided auction | |
| Lee | Pricing schemes and profit-maximizing pricing for cloud services | |
| CN110796513A (en) | Multitask learning method and device, electronic equipment and storage medium | |
| CN103971256A (en) | Information push method and device | |
| CN116266323A (en) | Intelligent charging method and device in cloud scene | |
| CN112017042A (en) | Resource quota determining method and device based on tweed distribution and electronic equipment | |
| CN105119751A (en) | Service evaluation and selection method based on environment real-time perceiving | |
| CN113159213A (en) | Service distribution method, device and equipment | |
| Kumar et al. | Fair mechanisms for combinatorial reverse auction-based cloud market | |
| CN106157138A (en) | Group belonging to a kind of bank transaction data determines method and device | |
| CN109858756A (en) | A kind of service quality defect diagnostic method and device | |
| CN110992056B (en) | Object level service method, device, electronic equipment and storage medium | |
| Zhou et al. | Maximizing the profits of cloud service providers via dynamic virtual resource renting approach | |
| Vinothiyalakshmi et al. | Enhanced multi-attribute combinative double auction (emcda) for resource allocation in cloud computing | |
| Kwon et al. | Royalties vs. upfront lump-sum fees in data communication environments | |
| CN113128597B (en) | Method and device for extracting user behavior characteristics and classifying and predicting user behavior characteristics | |
| Liu et al. | Research on multi-task assignment model based on task similarity in crowdsensing | |
| Liu et al. | An online combinatorial auction based resource allocation and pricing mechanism for network slicing in 5G | |
| CN113822704A (en) | Method and device for calculating discount cost, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |