Cloud-terminal secondary market resource collaborative purchasing method for edge computing

Abstract

The invention discloses a cloud-end secondary market resource collaborative purchasing method for edge computing. The method comprises the following steps: transmitting a user computing task to an edge server by a user with a resource demand, and purchasing computing resources from a mobile terminal device or a cloud server when the computing resources of the edge server cannot process the user computing task; the number of computing resources ordered to the mobile terminal equipment and the number of computing resources applied to the cloud server for reservation in each time slot, a fine value generated by unit computing resources lacked by the edge server and the number of computing resources capable of being provided by the mobile terminal equipment are obtained. According to the method and the device, the technical problem that the cost of the edge server is greatly increased due to the waste of idle resources caused by the fact that the edge server receives a large number of calculation demands in the resource demand peak period and deploys the calculation resources on the edge server according to the peak value of the calculation demands in the non-calculation demand peak period is solved.

Description

Technical Field

[0001] The present invention relates to the field of mobile group technology, and in particular to a cloud-end secondary market resource collaborative procurement method for edge computing. Background Art

[0002] The computing demand on the edge server side is sometimes high and sometimes low, and the peaks and troughs of computing demand appear with certain regularity. Based on relevant reference content and data, we can know the general distribution pattern of computing demand within a day. Figure 1 As shown in the figure, it is found that the amount of computing demand data on the edge server is closely coupled with the normal work and rest patterns of humans. For example, there will be relatively high traffic during the day and relatively low traffic during the midnight. There may be two peaks every day: one around noon and the other around 10 pm. At the same time, the computing demand within a week may have a similar pattern: that is, the computing demand data has certain regularities and characteristics, and it shows a periodic pattern, such as the amount of computing demand data on weekends is less than the traffic on weekdays. However, the computing resources on the edge server side are usually limited, and may not be able to meet user needs during the peak period of resource demand.

[0003] Edge servers receive a large number of computing requests during peak resource demand periods, and the local computing resources of edge servers are usually limited, so it is necessary to consider how to meet the computing resource requirements of user tasks. If computing resources on edge servers are deployed according to the peak computing demand, a large number of resources will be idle during non-peak computing demand periods. This waste will significantly increase the cost of edge servers.

[0004] At the same time, with the continuous development of mobile terminal devices, the computing power of mobile terminal devices is constantly increasing. Therefore, in order to solve this problem, the following solutions are considered: first, according to some historical data, the resources on the edge server are deployed according to the median value of computing resource requirements; then, during the peak period of resource demand, the edge server is allowed to expand its own resource pool by means of resource recruitment. In other words, it can be understood that the edge server has its own deployed local virtual machines. When the incoming virtual machine demand requests (computing resource requests) exceed the capacity of the edge node, the edge node needs to rent virtual machines from other nodes to meet these resource requests.

[0005] As the computing power of mobile terminal devices is constantly increasing, some mechanisms can be used to motivate mobile terminals. The advantage of mobile terminals is that they are closer to the edge of the network and can better guarantee QoS (Quality of Service). Treat it as a network infrastructure and use its computing resources to provide effective computing and services. This can reduce the huge economic and time costs of deploying large-scale edge servers to meet the growing computing needs of various delay-sensitive applications. However, although mobile terminal devices can provide certain computing resources, they are extremely unstable in themselves, and with the subjective will of mobile terminal devices, their mobility causes their resource availability time to be unstable.

[0006] For cloud servers, the computing resources on them are relatively sufficient, and some resources can be purchased from cloud servers to supplement the resource pool of edge servers. The available time of its computing resources is relatively stable, it can provide more reliable services, and its computing power is also stronger. Even though the cloud resources will cause a long delay due to the remoteness, it may not be able to guarantee good service quality for some delay-sensitive tasks.

[0007] From the perspective of optimization objectives, existing work almost always aims to maximize the benefits of mobile terminal devices. However, in many scenarios, mobile terminal devices are selfish. From the perspective of deployment, the deployment of edge servers and their utility also need to be considered. Summary of the invention

[0008] The embodiments of the present invention provide a cloud-end secondary market resource collaborative procurement method for edge computing, so as to at least solve the technical problem that the edge server receives a large number of computing demands during the peak period of resource demand, and the computing resources on the edge server are deployed according to the peak value of computing demand, which will cause a large number of resources to be idle during the non-peak period of computing demand. This waste will greatly increase the cost of the edge server.

[0009] According to one aspect of an embodiment of the present invention, a cloud-end secondary market resource collaborative procurement method for edge computing is provided. The method may include: when a user with resource demand passes a user computing task to an edge server, and the computing resources of the edge server cannot process the user computing task, and it is necessary to purchase computing resources from a mobile terminal device or a cloud server, obtain the number of computing resources ordered from the mobile terminal device in each time slot, the number of computing resources requested to be reserved from the cloud server, the penalty value generated by the unit computing resource lacking from the edge server, and the number of computing resources that the mobile terminal device can provide, wherein the price of each computing resource of the mobile terminal device is a first price, the price of each computing resource requested to be reserved by the cloud server is a second price, and the price of each computing resource purchased by the cloud server is a third price; based on the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources that the mobile terminal device can provide, determine the cost of the mobile terminal. transaction volume; after receiving the computing resources delivered by the mobile terminal device, obtain the income generated by the unit computing resources of the cloud server and the mobile terminal device, and determine the ideal total number of computing resources for the mobile terminal device and the cloud server based on the income generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server and the third price; based on the ideal total number of computing resources of the mobile terminal device and the cloud server, the transaction volume of the mobile terminal, the first price, the second price, the third price, the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources requested to be reserved from the cloud server, establish the edge server maximization profit function; based on the edge server maximization profit function, determine the optimal number of computing resources ordered from the mobile terminal device in each time slot and the optimal number of computing resources requested to be reserved from the cloud server.

[0010] Optionally, based on the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources that the mobile terminal device can provide, the expression for determining the transaction volume of the mobile terminal is: Q = min{R, Rd} Among them, Q represents the transaction volume of the mobile terminal, Rd is the number of computing resources ordered from the mobile terminal device for each time slot, R is the number of computing resources that the mobile terminal device can provide, and min is used to represent the minimum value of R and Rd.

[0011] Optionally, based on the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server, and the third price, the expression for determining the ideal total number of computing resources of the mobile terminal device and the cloud server is:

[0012] in, represents the inverse of the cumulative distribution function of the user's computing task, represents the total number of ideal computing resources for mobile terminal devices and cloud servers, It represents the revenue generated by the computing resources of cloud servers and mobile terminal devices. Indicates the penalty value generated by the lack of computing resources per unit on the edge server. Indicates the third price.

[0013] Optionally, based on the total number of ideal computing resources of the mobile terminal device and the cloud server, the transaction volume of the mobile terminal, the first price, the second price, the third price, the number of computing resources ordered from the mobile terminal device in each time slot, and the number of computing resources requested to be reserved from the cloud server, the expression for maximizing the profit function of the edge server is established as follows:

[0014] in, represents the edge server maximizing profit, Indicates the amount of computing resources ordered from the mobile terminal device in each time slot, Indicates the number of computing resources requested from the cloud server for each time slot. represents the second price, Indicates the first price Express expectations, It represents the expected revenue generated by the edge server when the number of computing resources delivered by the cloud server and mobile terminal devices is n.

[0015] Optionally, the expected revenue generated by the edge server is expressed as:

[0016] in, represents the expected revenue generated by the edge server when the number of computing resources delivered by the cloud server and mobile terminal devices is n, It represents the benefits that the computing resources of the cloud and mobile terminals bring to the edge server. Indicates the computing needs of users with resource requirements, represents the probability density function of the computational requirements, Represents the total number of computing resources delivered by mobile terminal devices and cloud servers.

[0017] Optionally, based on the edge server profit maximization function, the number of optimal computing resources ordered from the mobile terminal device in each time slot and the number of optimal computing resources requested to be reserved from the cloud server are determined, including: based on the edge server profit maximization function, determining the constraint function and decision variables of the edge server profit maximization function; solving the edge server profit maximization function based on the constraint function and decision variables of the edge server profit maximization function to obtain the number of optimal computing resources ordered from the mobile terminal device in each time slot and the number of optimal computing resources requested to be reserved from the cloud server.

[0018] Beneficial effects of the present invention: The present invention proposes a cloud-end secondary market resource collaborative procurement method for edge computing, by considering a resource procurement framework built by an edge server, a group of user computing tasks, a group of heterogeneous mobile terminal devices that can provide computing resources, and a cloud server; prioritizes recruiting computing resources from mobile terminals, and then supplements the computing resources procurement from cloud servers. Using the resource reservation-based method idea, the obtained information of the edge server, mobile terminal device, and cloud server is expressed as a problem of maximizing the utility of the edge server; the problem is converted into a single variable problem with a concave profit function, so as to derive the optimal order quantity and the optimal reservation quantity based on the expectation and classic resource procurement model, solves the technical problem that the edge server receives a large number of computing demands during the peak period of resource demand, and the computing resources on the edge server are deployed according to the peak of computing demand, which will cause a large number of resources to be idle during the non-peak period of computing demand. This waste will greatly increase the cost of the edge server, and achieves the technical effect that the edge server receives a large number of computing demands during the peak period of resource demand, and purchases computing resources from mobile terminal devices and cloud servers according to certain purchase rules to ensure that the computing needs of users are completed at the lowest cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 is a flow chart of a game control method for a flexible capture means of a space target according to an embodiment of the present invention; Figure 2 It is a scenario diagram of a cloud-end secondary market resource collaborative procurement method for edge computing according to an embodiment of the present invention; Figure 3 It is a schematic diagram of a solution of a cloud-end secondary market resource collaborative procurement method for edge computing according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] Example 1 According to an embodiment of the present invention, a cloud-end secondary market resource collaborative procurement method for edge computing is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system comprising at least one set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0023] Figure 1 is a flow chart of a cloud-end secondary market resource collaborative procurement method for edge computing according to an embodiment of the present invention. Figure 1 As shown, the method may include the following steps: Step S101, when a user with resource demand passes the user computing task to the edge server, and the computing resources of the edge server cannot process the user computing task, it is necessary to purchase computing resources from the mobile terminal device or the cloud server. In this scenario, the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources requested to be reserved from the cloud server, the penalty value generated by the unit computing resource lacking from the edge server, and the number of computing resources that can be provided by the mobile terminal device are obtained, wherein the price of each computing resource of the mobile terminal device is a first price, the price of each computing resource requested to be reserved by the cloud server is a second price, and the price of each computing resource purchased by the cloud server is a third price.

[0024] In the technical solution provided in the above step S101 of the present invention, Figure 2 is a scenario diagram of the cloud-end secondary market resource collaborative procurement method for edge computing according to an embodiment of the present invention. The task model of the computing resource demander is modeled as follows: the user with resource demand has limited resources and can only perform tasks with small computing requirements locally. Therefore, when the demand for computing resources exceeds the capacity of the user with resource demand, the user with resource demand will offload the task to the edge server to minimize the computing time and its own energy consumption. Indicates the task set currently received by the edge server. Indicates the total number of tasks. By triple express. Representation Task The required CPU cycle resources are the number of computing resources; Indicates that the task The utility brought to the mobile user upon completion. This utility is determined according to the different requirements of the user's tasks: for example, delay-sensitive tasks expect the fastest service, while energy-sensitive tasks expect the service with the lowest energy consumption. Indicates uninstall strategy. = 0, users with resource requirements will offload user computing tasks to the edge server. At the same time, it is assumed that a single computing task request only lasts for one time slot, that is, a single computing task can be completed within one time slot, so that the task will not be terminated due to the departure of the mobile terminal device. The user computing task demand within the coverage area of ​​the edge server is set to C, and according to historical data, it is assumed to have a probability density function fc and a cumulative distribution function Fc.

[0025] Mobile terminal equipment ( You can use a triple < , , > to indicate, Indicates the start time when computing resources of the mobile terminal device become available. Indicates the end time when the computing resources of the mobile terminal device are available, that is . Mobile terminal equipment Resource availability time A single time slot times. Indicates the unit idle computing resources that the mobile terminal device can provide. The computing resources of each mobile terminal device can be equally divided into unit idle computing resources times, which can be understood as the unit idle computing resources that the mobile terminal device can provide is indivual The number of available resources, that is, its ability to supply computing resources, is set to R. At the same time, based on historical data and forecasts, it is considered to have a probability density function fr and a cumulative distribution function Fr.

[0026] The unit price of computing resources purchased by edge servers for mobile terminal devices is ; The reserved unit price of the edge server for reserving resources to the cloud server is ; and the purchase price of computing resources from the cloud server is At the same time, due to the completion of the user's computing needs, the price per unit resource on the edge server is According to the actual conditions, it can be assumed that: + > : Indicates that the resource price of mobile terminal equipment is relatively cheap due to its own nature.

[0027] Figure 3 Schematic diagram of a solution for a cloud-end secondary market resource collaborative procurement method for edge computing according to an embodiment of the present invention; the demand C of the user's computing task is random, and there are two suppliers: a mobile terminal device and a cloud server. The number of computing resources that the mobile terminal device can provide is random R, while the cloud server is completely reliable. In each time slot, the edge server orders Rd units of computing resources from the mobile terminal device and sends Rd units of computing resources to the cloud server. In addition, after observing the amount of computing resources delivered by the mobile terminal device, the edge server can choose to reserve Rc units of computing resources at a price of Purchase no more than Rc units of computing resources from the cloud server at a price of

[0028] Step S102, determining the transaction volume of the mobile terminal based on the quantity of computing resources ordered from the mobile terminal device in each time slot and the quantity of computing resources that can be provided by the mobile terminal device.

[0029] In the technical solution provided in step S102 of the present invention, the transaction volume of the mobile terminal is obtained according to the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources that the mobile terminal device can provide. The edge server only pays for the transaction volume Q = min{R, Rd} delivered by the mobile terminal device, and the price per unit is .

[0030] Step S103, after receiving the computing resources delivered by the mobile terminal device, obtain the income generated by the unit computing resources of the cloud server and the mobile terminal device, and determine the ideal total number of computing resources for the mobile terminal device and the cloud server based on the income generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server, and the third price.

[0031] In the technical solution provided in the above step S103 of the present invention, the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server and the third price are calculated to obtain the total ideal computing resources of the mobile terminal device and the cloud server.

[0032] Step S104, establish an edge server profit maximization function based on the total number of ideal computing resources of the mobile terminal device and the cloud server, the transaction volume of the mobile terminal, the first price, the second price, the third price, the number of computing resources ordered from the mobile terminal device in each time slot, and the number of computing resources requested to be reserved from the cloud server.

[0033] In the technical solution provided in the above step S104 of the present invention, the total number of ideal computing resources of mobile terminal devices and cloud servers, the transaction volume of mobile terminals, the first price, the second price, the third price, the number of computing resources ordered from mobile terminal devices in each time slot, and the number of computing resources requested to be reserved from the cloud server are modeled to obtain the edge server's maximum profit function.

[0034] Step S105, based on the edge server's maximum profit function, determine the number of optimal computing resources ordered from the mobile terminal device in each time slot and the number of optimal computing resources requested to be reserved from the cloud server.

[0035] In the technical solution provided in the above step S105 of the present invention, the edge server's maximum profit function is solved to obtain the number of optimal computing resources ordered from the mobile terminal device in each time slot and the number of optimal computing resources requested to be reserved from the cloud server.

[0036] The above method of this embodiment is further introduced below.

[0037] As an optional embodiment, in step S102, based on the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources that the mobile terminal device can provide, the expression for determining the transaction volume of the mobile terminal is: Q = min{R, Rd} Among them, Q represents the transaction volume of the mobile terminal, Rd is the number of computing resources ordered from the mobile terminal device for each time slot, R is the number of computing resources that the mobile terminal device can provide, and min is used to represent the minimum value of R and Rd.

[0038] In this embodiment, the minimum value between the quantity of computing resources ordered from the mobile terminal device in each time slot and the quantity of computing resources that can be provided by the mobile terminal device is determined as the transaction volume of the mobile terminal.

[0039] As an optional embodiment, in step S103, based on the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking by the edge server, and the third price, the expression for determining the ideal total number of computing resources of the mobile terminal device and the cloud server is:

[0040] in, represents the inverse of the cumulative distribution function of the user's computing task, represents the total number of ideal computing resources for mobile terminal devices and cloud servers, It represents the revenue generated by the computing resources of cloud servers and mobile terminal devices. Indicates the penalty value generated by the lack of computing resources per unit on the edge server. Indicates the third price.

[0041] In this embodiment, the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server, and the third price are calculated to obtain the total ideal computing resources of the mobile terminal device and the cloud server.

[0042] As an optional implementation method, step S104, based on the total number of ideal computing resources of the mobile terminal device and the cloud server, the transaction volume of the mobile terminal, the first price, the second price, the third price, the number of computing resources ordered from the mobile terminal device in each time slot, and the number of computing resources requested to be reserved from the cloud server, establish an expression for maximizing the profit function of the edge server as follows:

[0043] in, represents the edge server maximizing profit, Indicates the amount of computing resources ordered from the mobile terminal device in each time slot, Indicates the number of computing resources requested from the cloud server for each time slot. represents the second price, Indicates the first price Express expectations, It represents the expected revenue generated by the edge server when the number of computing resources delivered by the cloud server and mobile terminal devices is n.

[0044] In this embodiment, the total number of ideal computing resources of mobile terminal devices and cloud servers, the transaction volume of mobile terminals, the first price, the second price, the third price, the number of computing resources ordered from mobile terminal devices in each time slot, and the number of computing resources requested to be reserved from the cloud server are modeled to obtain the edge server's maximum profit function.

[0045] As an optional embodiment, the expected income generated by the edge server is expressed as:

[0046] in, represents the expected revenue generated by the edge server when the number of computing resources delivered by the cloud server and mobile terminal devices is n, It represents the benefits that the computing resources of the cloud and mobile terminals bring to the edge server. Indicates the computing needs of users with resource requirements, represents the probability density function of the computational requirements, Represents the total number of computing resources delivered by mobile terminal devices and cloud servers.

[0047] In this embodiment, , and By performing integral calculations from 0 to n and from n to positive infinity, we can obtain the expected revenue generated by the edge server when the number of computing resources delivered by the cloud server and mobile terminal devices is n.

[0048] As an optional implementation method, step S105, based on the edge server's profit maximization function, determines the number of optimal computing resources ordered from the mobile terminal device in each time slot and the number of optimal computing resources requested to be reserved from the cloud server, including: based on the edge server's profit maximization function, determining the constraint function and decision variables of the edge server's profit maximization function; solving the edge server's profit maximization function based on the constraint function and decision variables of the edge server's profit maximization function to obtain the number of optimal computing resources ordered from the mobile terminal device in each time slot and the number of optimal computing resources requested to be reserved from the cloud server.

[0049] In this embodiment, the constraint function is expressed as:

[0050] The expression of decision variables is: ,

[0051]

[0052] in, Indicates mobile terminal equipment is selected.

[0053] The specific process of solving the objective function is: transform the problem into a single variable problem with a concave profit function, and then derive the optimal order quantity and the optimal reservation quantity based on the expectation and classic resource procurement model. The specific method is: first further expand the profit expression of the edge server, such as the expression related to expectation, etc. The further analysis is as follows: When there is only a cloud provider (i.e. Rd=0), we can get: , which is the purchase price Expected profit of resource procurement problem. Indicates the number of cloud resources purchased for this resource procurement problem, namely:

[0054] In the export right After the first-order condition of , it can be proved that it can be further restricted to Rc< Search for the optimal reservation quantity in . That is, when only the cloud is available, we only need to consider values ​​less than the optimal order quantity. Assume Rc> , we can get , and there are: . Then discuss each case separately (because which side is more expensive often determines the final derived expression of Rd and Rc) (1) When : Obviously, the purchase price of the cloud is higher, so the amount of resources ordered by the edge server from the mobile terminal device group will not be less than the amount without the cloud. In other words, the availability of the cloud should not affect the optimal resource purchase quantity of all mobile terminals. So far, the expressions of the optimal terminal device purchase quantity Rd* and the optimal cloud reservation quantity Rc* can be derived.

[0055] ① Optimal purchase volume on mobile devices: Rd* =

[0056] The above formula shows that the optimal booking quantity in the cloud is maximized Given. Among them, , = represents the optimal order quantity for the terminal equipment of the first-tier supplier. It can be seen that is concave. After determining the resource purchase amount of the mobile terminal device group, the uncertainty of the profit function of the edge server comes from how to optimally utilize the cloud provider. Just like the standard resource reservation problem with a concave profit function.

[0057] ② Therefore, the optimal booking quantity on the cloud is determined by the first-order condition, and there are two thresholds here:

[0058]

[0059] The optimal reservation quantity on the specific cloud depends on the reservation price on the cloud and the size of the threshold: > , then choosing not to reserve any capacity from the cloud is the best option. Here is a predetermined cost The critical value of , determines whether cloud resource provisioning should be enabled. , then Rc* satisfies ;if , Rc* satisfies It should be noted that when the maximum amount of resources that the device can provide > The second case is not applicable when = .

[0060] (2) When : When it is cheaper to buy resources from the cloud, the sum of the optimal order quantity of the terminal and the optimal order quantity of the cloud is the solution to the classic resource reservation problem, unless it is limited by the maximum capacity of the mobile terminal. In other words, unless the mobile terminal device group is the only available supplier, part of the order should be reserved for the cloud resource supplier. is the optimal purchase quantity of the first-tier supplier's mobile terminal device group, assuming that the cloud supply has reserved resources Rc. Then, for Rc< ,have .

[0061] It is considered as a univariate optimization problem, because the optimal cloud resource reservation is given by =argmax{ } is given. Among them, .because < , which is equivalent to the optimal ordering problem with two suppliers, one with random resource supply and one with completely reliable resource supply. This is a special case of the two unreliable supplier model studied by the researchers, where the profit function is unimodal. This means that is also unimodal, the optimal reservation quantity of the cloud can be determined from the first-order conditions. and , we can get:

[0062] Because the volume , in any other case produce:

[0063] After some algebra:

[0064] Finally, the solution is obtained, and the optimal terminal device purchase quantity Rd* and the optimal cloud booking quantity Rc* can be expressed as follows: There are also two thresholds here, which are and , , in Represents the optimal order quantity on the mobile terminal device group.

[0065] .

[0066] if , ;if : satisfy , Note that this only applies when the maximum amount of resources available from the mobile device group is less than Applicable if :but satisfy ,and .

[0067] In an embodiment of the present invention, when a user with resource demand transmits a user computing task to an edge server, and the computing resources of the edge server cannot process the user computing task, and computing resources need to be purchased from a mobile terminal device or a cloud server, the number of computing resources ordered from the mobile terminal device in each time slot, the number of computing resources requested to be reserved from the cloud server, the penalty value generated by the unit computing resource lacking from the edge server, and the number of computing resources that can be provided by the mobile terminal device are obtained, wherein the price of each computing resource of the mobile terminal device is a first price, the price of each computing resource requested to be reserved by the cloud server is a second price, and the price of each computing resource purchased by the cloud server is a third price; based on the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources that can be provided by the mobile terminal device, the mobile terminal device is determined. transaction volume; after receiving the computing resources delivered by the mobile terminal device, obtain the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, and determine the ideal total number of computing resources for the mobile terminal device and the cloud server based on the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server, and the third price; establish an edge server maximization profit function based on the ideal total number of computing resources of the mobile terminal device and the cloud server, the transaction volume of the mobile terminal, the first price, the second price, the third price, the number of computing resources ordered from the mobile terminal device in each time slot, and the number of computing resources requested to be reserved from the cloud server; based on the edge server maximization profit function, determine the optimal number of computing resources ordered from the mobile terminal device in each time slot and the optimal number of computing resources requested to be reserved from the cloud server. The technical problem that when the edge server receives a large number of computing demands during the peak period of resource demand, the computing resources on the edge server are deployed according to the peak value of computing demand, which will cause a large number of resources to be idle during the non-peak period of computing demand, and this waste will greatly increase the cost of the edge server is solved. The technical effect is achieved that when the edge server receives a large number of computing demands during the peak period of resource demand, the computing resources are purchased from mobile terminal devices and cloud servers according to certain purchasing rules to ensure that the computing needs of users are met at the lowest cost.

[0068] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0069] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0070] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0071] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0072] In addition, each functional unit in each embodiment of the present invention may be integrated into a first processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0073] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A cloud-end secondary market resource collaborative procurement method for edge computing, characterized in that: include: In a scenario where a user with resource demand transfers a user computing task to an edge server, and the computing resources of the edge server cannot process the user computing task, and computing resources need to be purchased from a mobile terminal device or a cloud server, the number of computing resources ordered from the mobile terminal device and the number of computing resources requested to be reserved from the cloud server in each time slot, the penalty value generated by the unit computing resource lacking from the edge server, and the number of computing resources that can be provided by the mobile terminal device are obtained, wherein the price of each computing resource of the mobile terminal device is a first price, the price of each computing resource requested to be reserved by the cloud server is a second price, and the price of each computing resource purchased by the cloud server is a third price; Determining the transaction volume of the mobile terminal based on the amount of computing resources ordered from the mobile terminal device in each time slot and the amount of computing resources available from the mobile terminal device; After receiving the computing resources delivered by the mobile terminal device, obtain the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, and determine the ideal total number of computing resources of the mobile terminal device and the cloud server based on the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server, and the third price; Based on the total number of ideal computing resources of the mobile terminal device and the cloud server, the transaction volume of the mobile terminal, the first price, the second price, the third price, the number of computing resources ordered from the mobile terminal device in each time slot, and the number of computing resources reserved from the cloud server, a profit maximization function of the edge server is established; Based on the edge server's maximized profit function, the optimal number of computing resources ordered from the mobile terminal device and the optimal number of computing resources reserved from the cloud server are determined in each time slot.

2. The method according to claim 1, characterized in that The expression for determining the transaction volume of the mobile terminal based on the number of computing resources ordered from the mobile terminal device in each time slot and the number of computing resources that the mobile terminal device can provide is: Q = min{R, Rd} Among them, Q represents the transaction volume of the mobile terminal, Rd is the number of computing resources ordered from the mobile terminal device for each time slot, R is the number of computing resources that the mobile terminal device can provide, and min is used to represent the minimum value of R and Rd.

3. The method according to claim 1, characterized in that The expression for determining the ideal total number of computing resources of the mobile terminal device and the cloud server based on the revenue generated by the unit computing resources of the cloud server and the mobile terminal device, the penalty value generated by the unit computing resources lacking from the edge server, and the third price is: in, represents the inverse of the cumulative distribution function of the user's computing task, represents the total number of ideal computing resources for mobile terminal devices and cloud servers, It represents the revenue generated by the computing resources of cloud servers and mobile terminal devices. Indicates the penalty value generated by the lack of computing resources per unit on the edge server. Indicates the third price.

4. The method according to claim 1, characterized in that The expression for establishing the edge server's profit maximization function based on the total number of ideal computing resources of the mobile terminal device and the cloud server, the transaction volume of the mobile terminal, the first price, the second price, the third price, the number of computing resources ordered from the mobile terminal device in each time slot, and the number of computing resources reserved from the cloud server is: in, represents the edge server maximizing profit, Indicates the amount of computing resources ordered from the mobile terminal device in each time slot, Indicates the number of computing resources requested from the cloud server for each time slot. represents the second price, represents the first price, Express expectations, It represents the expected revenue generated by the edge server when the number of computing resources delivered by the cloud server and mobile terminal devices is n.

5. The method according to claim 4, characterized in that The expected income generated by the edge server is expressed as: in, represents the expected revenue generated by the edge server when the number of computing resources delivered by the cloud server and mobile terminal devices is n, It represents the benefits that the computing resources of the cloud and mobile terminals bring to the edge server. Indicates the computing needs of users with resource requirements, represents the probability density function of the computational requirements, Represents the total number of computing resources delivered by mobile terminal devices and cloud servers.

6. The method according to claim 1, characterized in that The method of determining the number of optimal computing resources ordered from the mobile terminal device and the number of optimal computing resources reserved from the cloud server in each time slot based on the edge server maximization profit function includes: Based on the edge server's profit maximization function, determining the constraint function and decision variables of the edge server's profit maximization function; The edge server's profit maximization function is solved based on its constraint functions and decision variables to obtain the optimal number of computing resources ordered from the mobile terminal device and the optimal number of computing resources reserved from the cloud server in each time slot.

7. A computer system, characterized in that include: One or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method of claim 1.

8. A computer-readable storage medium, characterized in that Computer executable instructions are stored, and when the instructions are executed, they are used to implement the method of claim 1.

9. A computer program product, characterized in that The invention comprises computer executable instructions, which are used to implement the method of claim 1 when being executed.

Images