CN114500303A - Temporary cloud resource usage charging method - Google Patents
Temporary cloud resource usage charging method Download PDFInfo
- Publication number
- CN114500303A CN114500303A CN202011256346.XA CN202011256346A CN114500303A CN 114500303 A CN114500303 A CN 114500303A CN 202011256346 A CN202011256346 A CN 202011256346A CN 114500303 A CN114500303 A CN 114500303A
- Authority
- CN
- China
- Prior art keywords
- temporary
- resource
- resources
- cloud
- data center
- 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
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000008569 process Effects 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 4
- 238000005265 energy consumption Methods 0.000 claims description 3
- 238000002955 isolation Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 abstract description 11
- 230000008859 change Effects 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 238000004088 simulation Methods 0.000 description 6
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000013486 operation strategy Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 101100498818 Arabidopsis thaliana DDR4 gene Proteins 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
- H04L41/5051—Service on demand, e.g. definition and deployment of services in real time
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/14—Charging, metering or billing arrangements for data wireline or wireless communications
- H04L12/1403—Architecture for metering, charging or billing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/14—Charging, metering or billing arrangements for data wireline or wireless communications
- H04L12/1485—Tariff-related aspects
- H04L12/1496—Tariff-related aspects involving discounts
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a temporary cloud resource usage charging method, which comprises the following steps: an original user submits idle resources in preset cloud resources to a cloud data center; the cloud data center combines the idle resources into a temporary resource queue, and places a resource instance application submitted by a temporary user into a resource application queue; matching the resource application queue with the temporary resource queue, if the matching is successful, allocating the temporary resources to the temporary user for use, and paying the temporary resource use cost by the temporary user; otherwise, putting the resource application which is not successfully matched into a waiting queue, and performing preferential matching; the original user puts forward a request for recovering the temporary resources, the cloud data center stores the application data running on the temporary resources, then the recovered temporary resources are returned to the original user, and expense discount or cash compensation is provided for the original user. Compared with the prior art, the method and the system can fully and flexibly use the temporary cloud resources and ensure the economic benefits of users and cloud providers.
Description
Technical Field
The invention relates to the technical field of cloud computing, in particular to a temporary cloud resource usage charging method.
Background
Cloud computing is a new business model that allows users to access various computing resources and software services of a cloud computing center, including servers, storage, networks, etc., in an on-demand, easily scalable manner over the internet. In the cloud computing era, an IT service mode is completely different from the personal computer era, enterprises, organizations or individuals do not need to purchase a large number of IT infrastructures, establish data centers of the enterprises, the organizations or the individuals, hire special IT personnel to conduct infrastructure operation and maintenance management and the like, and various Internet-based services and applications can be completed only by renting cloud resources. This allows businesses, institutions and individuals to better focus on their professional areas of expertise.
At present, the technical research on Cloud computing is relatively mature, and the Cloud computing is widely applied in the industry, such as Amazon EC2, S3, Google AppEng, Microsoft Azure, salesform CRM and the like internationally, and domestic airy Cloud, huahua cheng Cloud, tengcun Cloud, kyotong Cloud, U-Cloud and the like all provide Cloud service products and services in different forms. Although many companies and research and development organizations have made many breakthroughs in the cloud computing-related technical field and products, the pricing and charging of cloud computing resources have not been studied intensively.
At present, charging modes widely adopted and generally accepted in the cloud computing service and application fields are mainly two modes of charging according to needs and charging in auction. However, both of these modes have certain drawbacks and disadvantages, wherein the on-demand charging mode is very difficult for the user, and although cloud computing is a flexible computing mode, the cloud computing center can provide how much resource as much resource is needed by the user, in fact, the user, especially a non-professional user, cannot predict how much resource is needed for the application, which causes two situations: the resources selected by the user are too much or too little, and for the case of too much resources, the user obviously needs to pay more resource usage cost, while for the case of too little selection, the situation that the application cannot be completed or cannot be completed within a specific time may occur;
the price of the auction charging mode depends on 2 factors, one is the number of users requesting the resource and the other is the number of the resource. This is an unpredictable factor for the user, but the demand for IT resources is positive. The key is the amount of resources, and for a cloud resource provider, the cloud resource provider always can give the illusion of resource shortage to a user, so that the auction price of the resources is improved.
In order to overcome the defects of the two charging modes, research has been proposed to acquire cloud computing resources in a reservation mode. The method mainly aims at some users, the application of the users may need a large amount of cloud computing resources, and if the cloud computing resources are requested in an on-demand mode, timely response may not be obtained; if the cloud computing resources are used in a bidding mode, the price possibly paid is high. Then, with the reservation method, on one hand, the response problem of the resource can be solved, and on the other hand, the use cost can be compromised. This solution is a relatively ideal choice for cloud computing users. The service mode is similar to that of a restaurant in real life, and a guest can order food in advance through a telephone, so that the service mode is very convenient for the guest, and the guest can have food at any time when arriving at the restaurant, but the service mode is different for the boss of the restaurant, because once the guest receives the food order, the dining positions of the guest must be ensured, and even if more guests come subsequently, the ordered dining positions cannot play due value. Similarly, the cloud provider may also face the same problem, and after receiving the reservation of the cloud computing resource by the user, the cloud provider may consider to charge a certain reservation fee, but if the user does not use the reserved resource, the resource will be in an idle state, become a temporary cloud resource (i.e., a cloud computing resource that is reserved by the user but is not yet used), and cannot perform the function, thereby causing the waste of the resource. At this time, if the cloud resource provider wants to serve more customers, the investment must be increased, more hardware infrastructure must be purchased, and more resource capacity is provided, which is obviously a very uneconomical strategy.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a temporary cloud resource use charging method, which can enable the temporary cloud resources to be used fully and flexibly and ensure the economic benefits of users and cloud providers.
The purpose of the invention can be realized by the following technical scheme: a charging method for use of temporary cloud resources comprises the following steps:
s1, reserving cloud resources from the cloud data center by the original user, and submitting idle resources in the reserved cloud resources to the cloud data center;
s2, the cloud data center sequentially arranges and combines all idle resources into a temporary resource queue according to the types and the configuration of the idle resources;
s3, submitting resource instance applications to a cloud data center by a temporary user according to application requirements of the temporary user, and sequentially placing all resource instance applications into a resource application queue by the cloud data center;
s4, the cloud data center matches the resource application queue with the temporary resource queue, if the matching is successful, the step S6 is executed, otherwise, the step S5 is executed;
s5, the cloud data center puts the resource application which is not successfully matched into a waiting queue, preferentially matches the resource application in the waiting queue, if the matching is successful within a preset time range, the step S6 is executed, otherwise, the cloud data center directly allocates a new resource instance to the temporary user;
s6, the cloud data center allocates the successfully matched temporary resources to temporary users for use, and the temporary users pay temporary resource use fees to the cloud data center according to the resource use amount;
s7, the original user puts forward a request for recovering the temporary resources to the cloud data center according to the self requirement, if the requested temporary resources are not allocated to the temporary user for use, the cloud data center directly returns the temporary resources to the original user for use, and if the requested temporary resources are already allocated to the temporary user for use, the step S8 is executed;
s8, the cloud data center saves the program, data, intermediate results and operation environment which are operated on the temporary resources, then recovers the temporary resources, returns the recovered temporary resources to the original user, and provides expense discount or cash compensation for the original user.
Further, the step S2 specifically includes the following steps:
s21, the cloud data center carries out grouping processing on idle resources with the consistent types to obtain a plurality of temporary resource groups;
and S22, sequencing the idle resources in each temporary resource group by the cloud data center according to the configuration size of each idle resource to obtain a temporary resource queue comprising a plurality of temporary resources.
Further, the application of the temporary user needs to satisfy any time algorithm.
Further, the content of the resource instance application submitted by the temporary user comprises an instance platform type and a resource configuration.
Further, the resource configuration includes the number of CPU cores, the size of a memory, the size of a disk, and the size of a storage space.
Further, the specific process of step S4 is: and the cloud data center sequentially matches each resource instance application with each temporary resource in the temporary resource queue according to the arrangement sequence of the resource instance applications in the resource application queue and the arrangement sequence of the temporary resources in the temporary resource queue, if a preset matching condition is met, the matching is successful, and the step S6 is executed, otherwise, the step S5 is executed.
Further, the preset matching conditions are specifically as follows: the type of the temporary resource is the same as the type of the instance platform applied by the resource instance, and the configuration of the temporary resource is greater than or equal to the resource configuration applied by the resource instance.
Further, the step S6 specifically includes the following steps:
s61, if the successfully matched temporary resources contain the data of the original user, the cloud data center conducts data isolation processing on the successfully matched temporary resources, then the step S62 is executed, and otherwise, the step S62 is directly executed;
and S62, the cloud data center allocates the temporary resources to the temporary users for use, and the temporary users pay the temporary resource use fees to the cloud data center according to the resource use amount by constructing a temporary resource charging model.
Further, the step S62 specifically includes the following steps:
s621, the cloud data center allocates the temporary resources to the temporary users for use, and obtains the starting time and the ending time of the temporary resources used by the temporary users, the type and the quantity of the temporary resources requested by the temporary users, the quantity of the temporary users requesting the temporary resources, and the real-time price of the temporary resources when used as required, so as to determine the charging function of the temporary resources;
s622, based on equipment cost of temporary resources borne by the cloud provider, cost discount or cash compensation of the cloud provider to an original user, profit expectation of the cloud provider and social influence income of the temporary resources, determining a temporary resource return constraint of the cloud provider;
s623, a temporary resource charging model is constructed according to the temporary resource charging function and the temporary resource return constraint, so that a temporary user can pay the temporary resource use fee to the cloud data center according to the resource use amount.
Further, the temporary resource charging function is specifically:
xit∈N,xit≥2,ki∈[1,+∞),qi,ci∈(0,1]
wherein p (t) is the use cost of the temporary resource at the time t, m is the number of the temporary resource, pi(t) is the price of the ith temporary resource at time t, riNumber of temporary resources demanded of i-th kind for temporary users, qiDiscount coefficient for the ith temporary resource, Di(t) is the real-time price of the ith temporary resource when it is used as required at time t, kiPrice preference coefficient, x, for the ith temporary resourceitTemporary number of users requesting temporary resources of i-th kind at time t, ciA price preference index for the ith temporary resource;
the temporary resource return constraint specifically includes:
R(t)＝p(t)+E(z,t)-C(d,n,a,t)-B(z,t)≥pE(z,t)
wherein R (t) is the temporary resource return, E (z, t) is the social influence income of the temporary resource, C (d, n, a, t) is the depreciation cost of the equipment, the energy and power cost and the personnel management operation cost of the temporary resource carried by the cloud provider, B (z, t) is the cash compensation or the discount of the cost of the original user provided by the cloud provider, p (z, t)E(z, t) is profit expectation of a cloud provider, d is related factor of equipment depreciation, n is related factor of energy consumption, a is related factor of management operation, t is time parameter, and z is related temporary resource.
Compared with the prior art, the invention has the following advantages:
the method and the system have the advantages that the original user reservation cloud resources are used as the basis, the idle resources in the reservation cloud resources are classified and sequenced, so that the idle resources in the reservation cloud resources can be rented to temporary users for use again as temporary resources, the temporary resources can be timely recovered to the original users for reuse, the temporary users pay temporary resource use cost to the cloud data center according to the resource use amount in the process of using the temporary resources by the temporary users, on one hand, the resource use rights and interests of the original users can be guaranteed, on the other hand, the temporary resources can be fully and flexibly utilized, the use efficiency of cloud computing resources is improved, resource waste is avoided, the temporary users can timely finish self application, and the cloud provider can be guaranteed to increase benefits and serve more temporary users without more investment in hardware infrastructure.
The method and the system construct a temporary resource charging model, comprehensively consider the economic benefits of the original user, the cloud provider and the temporary user, dynamically change the price of the temporary resource in real time along with the number of the temporary users, ensure that the price of the temporary resource is always lower than the original charging price according to the requirement, enable the temporary user to complete the application of the temporary user in time, and ensure the economic benefits of the temporary user; by utilizing the temporary resource return constraint, the cloud provider can obtain a return which is not lower than the expected profit when the temporary resource is leased for use, so that the economic benefit of the cloud provider is ensured; when the original user reserves the temporary resources in the cloud resources to be leased for use, the cloud provider returns a discount on the cost or compensates the cash to the original user, so that the economic benefit of the original user is ensured.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a schematic diagram showing the relationship between the resource price and the number of temporary users;
FIG. 3 is a schematic diagram of the change of resource prices with time;
FIG. 4 is a diagram illustrating the number of users using different resources in an embodiment;
FIG. 5 is a diagram illustrating prices of different resources according to an embodiment;
FIG. 6 is a diagram illustrating the relationship between the prices of different resources and the users in the embodiment.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
Examples
Since the invention is a usage charging method based on a part of the reserved but not used cloud computing resources of the cloud computing resource provider, new concepts are introduced for this purpose:
original user (i.e., intended user): subscribing to a user of cloud computing resources from a cloud provider;
temporary resources: the cloud computing resources which are reserved by the original user but are not used yet are all or part of the resources in idle state at some time, but are different from other cloud computing resources because the resources can be recycled by the original user at any time;
temporary user (i.e., secondary user): users who rent temporary resources from a cloud resource provider to complete their applications.
Since the temporary cloud computing resources are different from the original resources, it relates to the benefits of cloud providers, original users and temporary users, and the characteristics of cloud computing applications, etc., therefore, when applying the temporary resources, the following must be fully considered:
(1) since the temporary user can use the original user's instance resources, it is necessary that the original user has first paid to reserve a large amount of long-term reserved cloud computing resources from the cloud provider in a reserved instance manner, and for some period of time, some or all of these reserved resources are idle and unused, while the original user also has to be willing to submit the resources in idle state to the cloud data center for another lease, so as to obtain the benefit of reducing the partial lease charge or cash compensation.
(2) Because the original user does not necessarily determine how much cloud computing resources are needed for subsequent applications when submitting idle cloud resources to a cloud provider for renting again, the original user has a priority usage right for the submitted resources, that is, the original user can recover the idle resources as needed at any time after submitting them.
(3) Because the free resources of the original user are temporarily leased by the cloud data center, the application of the temporary user should not occupy the leased resources for a long time, which requires that the application of the temporary user must conform to the computation of Anytime Algorithm (i.e. any time Algorithm, which can stop at any time and the computation is iterative), such as newton's iteration, wavelet transform, etc., and the payment mode is performed according to the resource usage amount.
(4) Because the temporary resources used by the temporary user are possibly recovered by the original user at any time, the application of the temporary user cannot guarantee that the ideal result is strictly obtained at a certain time period or time point, and in order to make up for the unfavorable factor, the temporary user rents the temporary resources in a cheaper mode than directly renting the resources from a cloud provider, namely, a temporary resource charging model is adopted, so that the temporary user can complete the application of the temporary user only by paying lower cost than any other economic model.
(5) One aspect of importance for charging of temporary resources comes from the cloud provider, since on the one hand it is considered necessary to compensate the original users, and on the other hand these resources have to be rented to the temporary users in a cheaper way.
(6) The security of the data. Because the temporary resources are temporarily leased by the cloud provider according to the resource use condition of the original user, and the resource application of the temporary resources may relate to private data or personal privacy and other information of the original user, when the cloud provider leases the resources to the temporary user, the integrity and security of the data of the original user must be ensured, so that a good data isolation mechanism is required, and any temporary user cannot access and destroy the data of the original user.
(7) Transparency of charging. Since the use of temporary resources involves the respective interests of the original user, the temporary user and the cloud provider, the use of these resources is not as clear and understandable as the usual monthly subscription fee. Therefore, the consumption of temporary resources must be recorded to ensure fairness in billing. In practical application, a 'cloud table' similar to a 'water meter' and an 'electric meter' can be added to the temporary user side to display the use condition of the temporary resources, a simple cloud table can be added to the original user side to display the use condition of the temporary resources from a certain moment to a certain moment, and the trust between the user and a cloud provider can be improved through the addition of the 'cloud table'.
Therefore, the invention provides a temporary cloud resource usage charging method, as shown in fig. 1, which includes the following steps:
s1, reserving cloud resources from the cloud data center by the original user, and submitting idle resources in the reserved cloud resources to the cloud data center;
s2, the cloud data center sequentially arranges and combines all idle resources into a temporary resource queue according to the types and the configuration of the idle resources;
s3, submitting resource instance applications to a cloud data center by a temporary user according to application requirements of the temporary user, and sequentially placing all resource instance applications into a resource application queue by the cloud data center;
s4, the cloud data center matches the resource application queue with the temporary resource queue, if the matching is successful, the step S6 is executed, otherwise, the step S5 is executed;
s5, the cloud data center puts the resource application which is not successfully matched into a waiting queue, preferentially matches the resource application in the waiting queue, if the matching is successful within a preset time range, the step S6 is executed, otherwise, the cloud data center directly allocates a new resource instance to the temporary user;
s6, the cloud data center allocates the successfully matched temporary resources to temporary users for use, and the temporary users pay temporary resource use fees to the cloud data center according to the resource use amount;
s7, the original user puts forward a request for recovering the temporary resources to the cloud data center according to the self requirement, if the requested temporary resources are not allocated to the temporary user for use, the cloud data center directly returns the temporary resources to the original user for use, and if the requested temporary resources are already allocated to the temporary user for use, the step S8 is executed;
s8, the cloud data center saves the program, data, intermediate results and operation environment which are operated on the temporary resources, then recovers the temporary resources, returns the recovered temporary resources to the original user, and provides expense discount or cash compensation for the original user.
The invention sets the temporary resource price to dynamically change in real time along with the number of temporary users, and particularly, the temporary resource price can be gradually increased along with the increase of the number of temporary users applying for the resource, but is always lower than the cost of renting the resource from other charging modes or providers directly.
In order to reflect the relationship between the price of the temporary resource and the number of the temporary users, the relationship can be described by establishing a suitable mathematical model, and the expression form of the specific mathematical model is shown as formula (1):
the formula (1) reflects that the number of temporary resources in the temporary users is xtPrice of time, wherein xtIs the number of temporary users at time t; d (t) represents the real-time price of the cloud computing resources used as required at the moment t; q is a positive number less than or equal to 1 and represents a resource discount coefficient; k is a constant greater than or equal to 1 and represents a price preference coefficient of the resource; and c is a positive number less than or equal to 1 and represents the price preference index of the cloud computing resource. As can be seen from formula (1), the larger the k value is, the less the price preference of the temporary resource is, and the higher the cost of the temporary user renting is; the smaller the q value and the c value are, the larger the price preference of the temporary resource is, and the lower the cost of the temporary user leasing is. The values of the three variables are determined by the operation strategy of the cloud provider, and may change at different times.
Considering the case where q is 1, k is 1, and d (t) is h, the formula (1) can be simplified to the formula (2):
fig. 2 gives a graphical example of equation (2), where the ordinate is the price of the temporary resource and the abscissa is the number of temporary users. It reflects the changing relationship between the price of the temporary resource and the number of temporary users, assuming that the price of the pay-on-demand is a constant h. As can be known from FIG. 2, the price of the temporary resource increases with the number of the user requests, i.e. the more the user demands the temporary resource, the higher the price of the resource will be; when the number of requests exceeds a certain number, the resource price under this model will be close to the price of the pay-on-demand resource, but will not exceed the price of the pay-on-demand resource.
The number of temporary users requesting a temporary resource is, in fact, random, and it changes dynamically over time, according to dynamic charging principles,when the resource utilization rate is high, the price should be increased; the price should be lowered when the resource usage is low, so the resource price is reflected as a fluctuating curve on the time axis. Thus, the price of a temporary resource can also be viewed as a function of time, i.e., P (x)t) Is denoted by p (t).
Fig. 3 is a schematic diagram showing the dynamic variation relationship of the resource price with time, and it can be seen from the diagram that the variation relationship between the resource price and time is completely random, which is also easily understood in terms of resource usage, because the demand of the user for the resource is completely random at different time, and the resource price is dynamically varied in time due to the close correlation of the user request number.
For temporary resources, a resource instance is a resource package that includes various cloud computing resources, such as cpu, memory, bandwidth, storage, etc. To accurately calculate resources
(where i is 1,2, 3, 4 may represent resources as cpu, memory, bandwidth, storage, etc., respectively.) the price, available at a certain time period
Is shown in which P isiAnd (t) is the price of the resource i at the time t, namely the formula (1) is converted into the formula (3).
Pi(t) reflects the price of a certain resource i which is only rented for one unit at the time t, and in the actual application process, the temporary user can rent more than one unit for each resource, and r is usediRepresenting the number of temporary leases for a particular resource i, which is an integer multiple of the basic leased unit. For example, if the standard of the external lease of the cloud provider is 1.0GB, the user may lease 2.0GB, 4.0GB orOther values, at this time Pi(t) is the price of 1.0GB of memory, and riThe amount of memory leased for the user may be 2, 4 or some other value.
From the above analysis, it can be seen that the temporary user needs to pay for the temporary resource rented by the user for the following:
xit∈N,xit≥2,ki∈[1,+∞),qi,ci∈(0,1] (4)
in the formula (4), m represents the number of the cloud computing resources, ts、teRespectively representing the time of the beginning and the end of the temporary user renting the leased resource, and because the change relation of the price of the resource along with the time is random, the discrete form can be applied in the process of applying the formula (4) to charge and use the temporary resource.
xit∈N,xit≥2,ki∈[1,+∞),qi,ci∈(0,1] (5)
In order to enable the formula (5) to reflect the change relationship of the price along with the time in a real way, the value of the time interval delta t can be made as small as possible, but in the actual use process, according to the actual situation of charging, the value of the delta t can be only taken in seconds.
The formula (4) gives a calculation function of charging temporary resources, and coefficients and indexes in the specific calculation function depend on operation strategies, management levels, expected levels of profits, fund compensation amounts for original users and the like of the cloud provider, if depreciation cost, energy and power cost and personnel management and operation cost of the equipment carrying the temporary resources of the cloud provider are C (d, n, a, t); the discount of the cash compensation or tariff given to the original user by the cloud provider is B (z, t); term of profit for cloud providerThe expected value is pE(z, t); the cloud provider provides temporary resource leasing services for temporary users, so that market and business expansion is brought, finally competitiveness is improved, and a mode of converting formed social influence into income is represented by E (z, t). If the development of certain services, or the development of certain services for certain specific groups or customers is important for the service expansion, market competition and customer attraction, the value of the social impact profit term E (z, t) can be set to a larger value; otherwise, if the effect is general, the value of E (z, t) can be set to a small value, even 0. Wherein C (d, n, a, t), B (z, t), pED in (z, t) and E (z, t) represents a factor related to equipment, n represents a factor related to energy consumption, a represents a factor related to management operation, t is a time parameter, and z is a related cloud resource.
Through the above analysis, it can be obtained that the cloud provider provides the reward r (t) that is available for the temporary resource lease service based on the temporary user:
R(t)＝p(t)+E(z,t)-C(d,n,a,t)-B(z,t) (6)
obviously, the return that the cloud provider needs to receive must be no less than its expected revenue, i.e., the return is
R(t)≥pE(z,t) (7)
Based on the above-mentioned reward constraints, the values of the parameters in the temporary resource charging function can be determined.
In summary, the specific working process of the method of the present invention can be described as follows:
(1) the original user and the temporary user respectively enter into a temporary resource leasing agreement with the cloud provider, provided that the original user {1,2 … … n } has resources idle and is willing to submit the resources to the cloud provider for fee discount or cash compensation; the application of the temporary user has a characteristic of satisfying Anytime Algorithm.
(2) The method comprises the steps that an original user submits idle resources to a cloud provider, the cloud provider arranges the idle resources in a leased resource queue according to the type and the configuration size of a temporary resource packet in a certain sequence, the temporary resource packet is firstly grouped according to the resource type, and then the temporary resource queue is obtained by arranging the temporary resource packet in each group according to the sequence of the configuration size of the resources.
(3) The temporary user {1,2 … … n } submits a resource instance application according to the characteristics and requirements of the application, and the required instance platform types such as win 864 bits, win 1064 bits, Ubuntu 16.04 and the like must be explicitly supported; the number of resources, such as the number of cores of the CPU, the size of the memory, the size of the disk, the size of the storage space, and the like. And the cloud data center places all resource instance applications in an application queue according to the requests of the users.
(4) The cloud data center matches the resource application queue of the temporary user with the temporary resource queue, the matching strategy adopts a front-to-back scanning mode, and as long as a resource instance in the temporary resource queue can meet the requirement of a user request (the system platform is the same, and the resource configuration of the temporary instance is the application instance configuration), the matching is regarded as successful; otherwise, the match is unsuccessful. If the matching is unsuccessful, the user request is placed in a waiting queue until the temporary resource queue has instance resources meeting the matching condition, and the instance resources are preferentially distributed. When the waiting time of the user request exceeds a certain threshold value and no proper leased resource still exists, the cloud data center can respond to the user request by allocating a new resource instance meeting the condition.
(5) And for the successfully matched temporary resources, allocating the temporary resources to the temporary users for use, and paying the fees according to the resource usage amount by the temporary users based on the temporary resource charging model.
(6) After the original users {1,2 … … n } submit their unused resources to the cloud provider, they can make requests to reclaim these instance resources at any time according to their needs, regardless of whether they have been allocated for temporary use.
(7) If the instance that was proposed to be reclaimed by the original user is not already allocated at the cloud computing center, it will be reclaimed directly; if the instance has been allocated for temporary user use, the cloud data center must save the program, data, intermediate results, and operating environment that is running on the portion of temporary resources and then reclaim it.
(8) And after the cloud data center finishes storing the operating environment of the temporary user, the temporary resources occupied by the temporary user are returned to the original user.
In order to verify the effectiveness of the method provided by the present invention, the present embodiment adopts a simulation experiment for verification. The experimental hardware adopts a quad-core i 57400 CPU desktop computer, the working frequency of a host computer is 3.0GHz, the capacity of a hard disk is 1TB, the rotating speed is 7200rpm, the internal memory is 8.0GB DDR4, and the operating system is 64-bit Ubuntu 16.04 single-machine version.
In the experimental process, the rented resources which are temporarily free by the original user are submitted to a resource pool of the cloud data center, the resource demand of the temporary user is matched with some resources in the resource queue and responded, and in addition, the application of the temporary user conforms to Anytime Algorithm and conforms to the principle of one-time payment.
Since newton's iteration is a typical Algorithm that conforms to Anytime Algorithm, the whole calculation process is iterative and can be stopped at any time. In the simulation experiment, the embodiment adopts a pre-charging strategy, that is, before the algorithm starts, the user needs to charge in advance, and the charged amount is determined by the user. In the calculation process, the consumed fee is deducted from the account of the user according to the resource condition used by the application each time and the proposed temporary resource charging model, and when the fee is insufficient, the system prompts the user to recharge. At the moment, the application is suspended, and the intermediate result is fed back to the user, the user determines whether the result meets the requirement according to the actual condition, and if the result meets the requirement, the application can be selected to quit; otherwise, the recharging can be selected to ensure that the application continues to be carried out until a result satisfied by the user is obtained. The simulation experiment uses the function f (x) x27x +9 as an example, solving for f (x) 0 (x) by using a Newton iteration method>0) And (6) obtaining the result. The iteration initial value is set to be 0, the initial recharging amount is 2000 (the specific unit can be set in the actual application process, and the user can recharge again when the amount is not enough). The resource situation is as follows: it is assumed that the CPU requests are in units of cores (the performance of a specific CPU core may be specified according to actual conditions during service operation), the memory is in units of 128MB, and the bandwidth is in units of 2 MB. In the process of resource application, the user can set the required resource quantity to be 1Any other number may be used, depending on the user. In this embodiment, it is assumed that the CPU, the memory, and the bandwidth used by the user are all 1 unit.
Since the prices of the three resources are closely related to the number of users using the resources, in order to make the simulation experiment closer to the actual application situation, the number of users using each resource is simulated by using a random number in the experiment process, and the specific situation is as shown in fig. 4. The number of persons using various resources in each step of the calculation is shown in the figure, wherein the time required for each step of the calculation is shown in each small square of the abscissa.
The experimental results of the simulation run are shown in table 1, wherein the first and fifth columns represent the number of iterative operations; the second and sixth columns represent the time required by each corresponding iteration, and the unit is microsecond; the third and seventh columns are results obtained after each iteration; the fourth and eighth columns are the cost required by each iterative operation, and the specific unit can be set according to the actual situation in the application process. As can be seen from the table, for this application, a total of 18 steps are required to obtain a satisfactory result for the user, and the time required in each iteration is different, and the corresponding costs are different, and the cost required in each iteration is the sum of the costs required for various resources.
TABLE 1
| Number of runs | Run time | Calculation results | Price | Number of runs | Run time | Calculation results | Price |
| 1 | 287 | 1.285714 | 1606.434 | 10 | 387 | 1.696739 | 2202.758 |
| 2 | 221 | 1.521866 | 1248.796 | 11 | 360 | 1.696989 | 1694.857 |
| 3 | 232 | 1.616582 | 1351.957 | 12 | 234 | 1.69711 | 1297.591 |
| 4 | 357 | 1.659048 | 2081.421 | 13 | 392 | 1.697169 | 2278.924 |
| 5 | 323 | 1.67892 | 1861.049 | 14 | 265 | 1.697198 | 1551.236 |
| 6 | 299 | 1.688396 | 1702.038 | 15 | 369 | 1.697211 | 2148.442 |
| 7 | 317 | 1.692955 | 1841.976 | 16 | 287 | 1.697218 | 1654.267 |
| 8 | 335 | 1.695156 | 1791.534 | 17 | 219 | 1.697221 | 1225.508 |
| 9 | 271 | 1.696222 | 1552.747 | 18 | 249 | 1.697223 | 1341.956 |
Fig. 5 shows the dynamic change of various resources in the using process, and the change depends on the unit price of various resources and the using time thereof. In fact, as can be seen from the figure, even if the time occupied by the resources is the same, the price change rules of the various resources are not completely the same, and in fig. 5, in step 11 of the operation, the operation time of the application is 360us, although the usage cost of each resource is a trend that is reduced, the reduction range of the memory usage cost is far beyond that of other resources, which indicates that the usage price of the memory is very low in this period. As can be seen from the analysis of the above equation (3), when the number of leased resources is constant, the unit price of the resource is related to the number of users using the resource. The unit price of the memory is low, which only indicates that the number of temporary users requesting to use the memory is small at this moment.
Fig. 6 shows unit price conversion of various resources in different user usage situations. Formula (1) indicates that the unit price of renting resources based on the temporary user is related to the real-time price of the cloud computing resources used as required, and for convenience of discussion in the experiment, the prices when the resources are used as required are assumed to be constant, wherein the price of the memory is 3, the price of the CPU is 2, and the price of the bandwidth is 1. As can be seen from the results of the simulation experiment, for the temporary resource billing model, the price of each resource is lower than that of the resource when used on demand, and they change sharply with the change of the number of temporary users. It can also be seen from the figure that in the 11 th operation stage, the number of users of the CPU is only 5, which results in that the unit price of the CPU in this period is very low, has fallen to 2.0 and is only 1.8, which is far lower than the price of the resource used on demand, and this result just explains the variation law of the CPU in fig. 4 in this period. Similarly, when the number of users of the CPU is small, 13 and 12, respectively, the unit price thereof also shows a distinct downward inflection point, which indicates that the unit price of the resource varies significantly with the number of users.
In summary, by using the method for charging the temporary cloud resource, the original user can rent some cloud resources in a predetermined manner, and the problem that the original user cannot complete the application of the original user because the original user cannot rent appropriate resources can be well avoided;
the temporary user can complete the application of the temporary user by using lower cost;
the method can ensure that the original user rents temporary users through the cloud provider when the reserved cloud resources are partially or completely idle, thereby obtaining cash compensation or expense reduction;
the cloud provider can serve more potential users, so that the use efficiency of cloud resources is improved, and meanwhile, on the premise of not increasing the investment of infrastructure, the service is expanded and the income is increased.
Claims (10)
1. A charging method for use of temporary cloud resources is characterized by comprising the following steps:
s1, reserving cloud resources from the cloud data center by the original user, and submitting idle resources in the reserved cloud resources to the cloud data center;
s2, the cloud data center sequentially arranges and combines all idle resources into a temporary resource queue according to the types and the configuration of the idle resources;
s3, submitting resource instance applications to a cloud data center by a temporary user according to application requirements of the temporary user, and sequentially placing all resource instance applications into a resource application queue by the cloud data center;
s4, the cloud data center matches the resource application queue with the temporary resource queue, if the matching is successful, the step S6 is executed, otherwise, the step S5 is executed;
s5, the cloud data center puts the resource application which is not successfully matched into a waiting queue, preferentially matches the resource application in the waiting queue, if the matching is successful within a preset time range, the step S6 is executed, otherwise, the cloud data center directly allocates a new resource instance to the temporary user;
s6, the cloud data center allocates the successfully matched temporary resources to temporary users for use, and the temporary users pay temporary resource use fees to the cloud data center according to the resource use amount;
s7, the original user puts forward a request for recovering the temporary resources to the cloud data center according to the self requirement, if the requested temporary resources are not allocated to the temporary user for use, the cloud data center directly returns the temporary resources to the original user for use, and if the requested temporary resources are already allocated to the temporary user for use, the step S8 is executed;
s8, the cloud data center saves the program, data, intermediate results and operation environment which are operated on the temporary resources, then recovers the temporary resources, returns the recovered temporary resources to the original user, and provides expense discount or cash compensation for the original user.
2. The method for charging for use of temporary cloud resources according to claim 1, wherein the step S2 specifically includes the following steps:
s21, the cloud data center carries out grouping processing on idle resources with the consistent types to obtain a plurality of temporary resource groups;
and S22, sequencing the idle resources in each temporary resource group by the cloud data center according to the configuration size of each idle resource to obtain a temporary resource queue comprising a plurality of temporary resources.
3. The method according to claim 2, wherein the application of the temporary user needs to satisfy any time algorithm.
4. The method for charging temporary cloud resource usage according to claim 3, wherein the content of the resource instance application submitted by the temporary user includes an instance platform type and a resource configuration.
5. The method according to claim 4, wherein the resource configuration includes the number of CPU cores, the size of a memory, the size of a disk, and the size of a storage space.
6. The method for charging for use of temporary cloud resources according to claim 4, wherein the specific process of step S4 is as follows: and the cloud data center sequentially matches each resource instance application with each temporary resource in the temporary resource queue according to the arrangement sequence of the resource instance applications in the resource application queue and the arrangement sequence of the temporary resources in the temporary resource queue, if a preset matching condition is met, the matching is successful, and the step S6 is executed, otherwise, the step S5 is executed.
7. The method for charging for use of temporary cloud resources according to claim 6, wherein the preset matching condition is specifically: the type of the temporary resource is the same as the type of the instance platform applied by the resource instance, and the configuration of the temporary resource is greater than or equal to the resource configuration applied by the resource instance.
8. The method for charging for use of temporary cloud resources according to claim 1, wherein the step S6 specifically includes the following steps:
s61, if the successfully matched temporary resources contain the data of the original user, the cloud data center conducts data isolation processing on the successfully matched temporary resources, then the step S62 is executed, and otherwise, the step S62 is directly executed;
and S62, the cloud data center allocates the temporary resources to the temporary users for use, and the temporary users pay the temporary resource use fee to the cloud data center according to the resource use amount by constructing a temporary resource charging model.
9. The method for charging for use of temporary cloud resources according to claim 8, wherein the step S62 specifically includes the following steps:
s621, the cloud data center allocates the temporary resources to the temporary users for use, and obtains the starting time and the ending time of the temporary resources used by the temporary users, the type and the quantity of the temporary resources requested by the temporary users, the quantity of the temporary users requesting the temporary resources, and the real-time price of the temporary resources when used as required, so as to determine the charging function of the temporary resources;
s622, based on equipment cost of temporary resources borne by the cloud provider, cost discount or cash compensation of the cloud provider to an original user, profit expectation of the cloud provider and social influence income of the temporary resources, determining a temporary resource return constraint of the cloud provider;
s623, a temporary resource charging model is constructed according to the temporary resource charging function and the temporary resource return constraint, so that a temporary user can pay the temporary resource use fee to the cloud data center according to the resource use amount.
10. The method for charging for use of temporary cloud resources according to claim 9, wherein the temporary resource charging function is specifically:
xit∈N,xit≥2,ki∈[1,+∞),qi,ci∈(0,1]
wherein p (t) is the use cost of the temporary resource at the time t, m is the number of the temporary resource, pi(t) is the price of the ith temporary resource at time t, riFor temporary useThe number of users' demands for the ith temporary resource, qiDiscount coefficient for i-th temporary resource, Di(t) is the real-time price of the ith temporary resource when it is used as required at time t, kiIs the price preference coefficient, x, of the ith temporary resourceitTemporary number of users requesting temporary resources of i-th kind at time t, ciA price preference index for the ith temporary resource;
the temporary resource return constraint specifically includes:
R(t)＝p(t)+E(z,t)-C(d,n,a,t)-B(z,t)≥pE(z,t)
wherein R (t) is the temporary resource return, E (z, t) is the social influence income of the temporary resource, C (d, n, a, t) is the depreciation cost of the equipment, the energy and power cost and the personnel management operation cost of the temporary resource carried by the cloud provider, B (z, t) is the cash compensation or the discount of the cost of the original user provided by the cloud provider, p (z, t)E(z, t) is profit expectation of a cloud provider, d is related factor of equipment depreciation, n is related factor of energy consumption, a is related factor of management operation, t is time parameter, and z is related temporary resource.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011256346.XA CN114500303A (en) | 2020-11-11 | 2020-11-11 | Temporary cloud resource usage charging method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011256346.XA CN114500303A (en) | 2020-11-11 | 2020-11-11 | Temporary cloud resource usage charging method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114500303A true CN114500303A (en) | 2022-05-13 |
Family
ID=81489700
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011256346.XA Pending CN114500303A (en) | 2020-11-11 | 2020-11-11 | Temporary cloud resource usage charging method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114500303A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116567454A (en) * | 2023-07-12 | 2023-08-08 | 江苏赛融科技股份有限公司 | Equipment full life cycle operation monitoring method and system based on Internet of things |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102117468A (en) * | 2011-04-07 | 2011-07-06 | 上海兆民云计算科技有限公司 | Method for leasing instance based on cloud computing |
| CN102184503A (en) * | 2011-04-07 | 2011-09-14 | 上海兆民云计算科技有限公司 | Charging method based on lease instance cloud computing |
| CN102739803A (en) * | 2012-07-09 | 2012-10-17 | 上海达希能源科技有限公司 | Cloud computing resource allocation method |
-
2020
- 2020-11-11 CN CN202011256346.XA patent/CN114500303A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102117468A (en) * | 2011-04-07 | 2011-07-06 | 上海兆民云计算科技有限公司 | Method for leasing instance based on cloud computing |
| CN102184503A (en) * | 2011-04-07 | 2011-09-14 | 上海兆民云计算科技有限公司 | Charging method based on lease instance cloud computing |
| CN102739803A (en) * | 2012-07-09 | 2012-10-17 | 上海达希能源科技有限公司 | Cloud computing resource allocation method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116567454A (en) * | 2023-07-12 | 2023-08-08 | 江苏赛融科技股份有限公司 | Equipment full life cycle operation monitoring method and system based on Internet of things |
| CN116567454B (en) * | 2023-07-12 | 2023-09-22 | 江苏赛融科技股份有限公司 | Equipment full life cycle operation monitoring method and system based on Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Towards optimal capacity segmentation with hybrid cloud pricing | |
| JP5680105B2 (en) | Sharing resources between clouds in a cloud computing environment | |
| Li et al. | Optimal resource provisioning for cloud computing environment | |
| Liu et al. | Cost-aware cloud service request scheduling for SaaS providers | |
| US20100057625A1 (en) | Negotiation of power rates based on dynamic workload distribution | |
| Bapna et al. | A market design for grid computing | |
| Hartline et al. | Envy, truth, and profit | |
| CN105721565A (en) | Game based cloud computation resource allocation method and system | |
| US20080301029A1 (en) | Application of brokering methods to recoverability characteristics | |
| CN102739803A (en) | Cloud computing resource allocation method | |
| CN102184503A (en) | Charging method based on lease instance cloud computing | |
| Li et al. | A price-incentive resource auction mechanism balancing the interests between users and cloud service provider | |
| CN106506607A (en) | Cloud computing resources distribution method based on fair credible two way auction mechanism | |
| Li et al. | A dynamic pricing reverse auction‐based resource allocation mechanism in cloud workflow systems | |
| CN114757710A (en) | Evaluation method for load aggregator bidding | |
| CN114500303A (en) | Temporary cloud resource usage charging method | |
| Ding et al. | Truthful online double auctions for on-demand integrated ride-sourcing platforms | |
| Alzhouri et al. | Dynamic pricing for maximizing cloud revenue: a column generation approach | |
| Courcoubetis et al. | Economic issues in shared infrastructures | |
| Tsai et al. | Bid-proportional auction for resource allocation in capacity-constrained clouds | |
| Xu et al. | Hierarchical combinatorial auction in computing resource allocation for mobile blockchain | |
| Valogianni et al. | Multiple Vickrey auctions for sustainable electric vehicle charging | |
| US20180165772A1 (en) | Tiered greening for large business operations with heavy power reliance | |
| Zhou et al. | An online emergency demand response mechanism for cloud computing | |
| CN109559024B (en) | Dynamic resource scheduling method under server-free architecture for power system |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220513 |