Dynamic Service Deployment Method, System, Device and Storage Medium for Internet of Vehicles

Abstract

The present invention discloses a method, system, device and storage medium for dynamic service deployment in a vehicle networking. This method obtains a service request and environmental data, and generates a service deployment decision through a policy function according to the service request and the environmental data. Then, a decision action and a decision response are obtained according to the service deployment decision. Next, according to the decision action and the decision response, a decision quality value of the service deployment decision is generated through a value function. Then, a target service deployment decision is determined from the service deployment decisions according to the decision quality value. Subsequently, services are deployed according to the target service deployment decision. This method is more in line with the actual dynamic scenario, and can effectively reduce the response time of services and maintain the resource balance of edge servers, so as to effectively cope with emergencies. The present invention can be widely applied in the technical field of vehicle networking edge computing technology.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Vehicles (IoV) edge computing, and in particular to a dynamic service deployment method, system, device and storage medium for the Internet of Vehicles (IoV). Background Art

[0002] The emergence of 5G technology has promoted the development of the Internet of Vehicles (IoV) technology. 5G technology helps to build an intelligent and sustainable in-vehicle network with security, reliability, efficient transportation, low latency and wider network coverage.

[0003] The Internet of Vehicles (IoV) can meet various service requirements during vehicle driving. However, due to problems such as limited edge resources, high mobility of vehicles, growing service requirements, and dynamics of service request types, the service deployment of the Internet of Vehicles (IoV) has become a challenging task. In related technologies, the static deployment solution cannot effectively solve practical problems because it does not consider traffic mobility and service dynamics. Summary of the Invention

[0004] An object of the present invention is to solve at least to some extent one of the technical problems existing in the prior art.

[0005] To this end, an object of an embodiment of the present invention is to provide a dynamic service deployment method, system, device and medium for the Internet of Vehicles (IoV), which can reduce the response time of services and maintain the resource balance of edge servers.

[0006] To achieve the above technical object, the technical solutions adopted by the embodiments of the present invention include:

[0007] On the one hand, an embodiment of the present invention provides a dynamic service deployment method for the Internet of Vehicles (IoV), including the following steps:

[0008] Obtain service requests and environmental data;

[0009] According to the service request and the environmental data, generate a service deployment decision through a policy function;

[0010] Obtain a decision action and a decision response according to the service deployment decision;

[0011] According to the decision action and the decision response, generate a decision quality value of the service deployment decision through a value function;

[0012] Determine a target service deployment decision from the service deployment decisions according to the decision quality value;

[0013] Deploy the service according to the target service deployment decision.

[0014] Further, the environmental data includes a vehicle ID and a vehicle location. The obtaining of the service request and the environmental data specifically includes:

[0015] Receiving the vehicle ID, the vehicle location, and the service request sent through a communication module on the vehicle.

[0016] Further, the generating of a service deployment decision on the edge server according to the service request and the environmental data specifically includes:

[0017] Extracting the features of the environmental data;

[0018] Generating a service deployment decision through a policy function according to the service request and the features of the environmental data.

[0019] Further, the extracting of the features of the environmental data specifically includes:

[0020] Obtaining the resource consumption amount and the total service time of the service request;

[0021] Obtaining the available resources on the edge and the available resources in the cloud;

[0022] Obtaining the edge node load according to the available resources on the edge, the available resources in the cloud, and the resource consumption amount;

[0023] Taking the total service time and the edge node load as the features of the environmental data.

[0024] Further, the obtaining of the total service time of the service request according to the environmental data specifically includes:

[0025] Obtaining the signal propagation speed of the service request;

[0026] Obtaining the Euclidean distance between the edge node and the vehicle according to the environmental data;

[0027] Obtaining the total service time of the service request through a preset formula according to the propagation speed and the Euclidean distance.

[0028] Further, after the step of deploying the service according to the target service deployment decision, the method further includes:

[0029] Determining that the service has completed the response, deleting the current service on the edge server and uploading it to the cloud server.

[0030] Further, the generating of a service deployment decision on the edge server according to the service request and the environmental data includes at least one of the following:

[0031] Generate a service deployment decision through a policy function in an edge node according to the service request and the environmental data;

[0032] Or

[0033] Generate a service deployment decision through a policy function in a cloud server according to the service request and the environmental data.

[0034] On the other hand, an embodiment of the present invention provides a dynamic service deployment system for an Internet of Vehicles, including:

[0035] On the other hand, an embodiment of the present invention provides a dynamic service deployment device for an Internet of Vehicles, including:

[0036] At least one processor;

[0037] At least one memory for storing at least one program;

[0038] When the at least one program is executed by the at least one processor, the at least one processor implements the dynamic service deployment method for the Internet of Vehicles described above.

[0039] On the other hand, an embodiment of the present invention provides a storage medium, in which instructions executable by a processor are stored, and the instructions executable by the processor are used to implement the dynamic service deployment method for the Internet of Vehicles when executed by the processor.

[0040] The present invention discloses a dynamic service deployment method for an Internet of Vehicles, having the following beneficial effects:

[0041] In this embodiment, by obtaining a service request and environmental data, and then according to the service request and the environmental data, generating a service deployment decision through a policy function, so as to generate a preliminary service deployment strategy for subsequent steps to correct it. Then, obtain a decision action and a decision response according to the service deployment decision, and according to the decision action and the decision response, generate a decision quality value of the service deployment decision through a value function to objectively evaluate the generated service deployment strategy. Then, determine a target service deployment decision from the service deployment decisions according to the decision quality value, correct the service deployment strategy through the decision quality value to obtain a strategy with better deployment effect, and then perform service deployment on the service according to the target service deployment decision, which can make the response time of the service shorter and maintain the resource balance of the edge server, so as to effectively cope with emergencies. Description of the Drawings

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following provides an introduction to the relevant technical solution drawings in the embodiments of the present invention or the prior art. It should be understood that the drawings in the following introduction are only for conveniently and clearly presenting some embodiments of the technical solutions in the present invention. For those skilled in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0043] Figure 1 A schematic structural diagram of an Internet of Vehicles provided by an embodiment of the present invention;

[0044] Figure 2 A schematic structural diagram of an Internet of Vehicles service deployment system provided by an embodiment of the present invention;

[0045] Figure 3 A schematic module diagram of an Internet of Vehicles service deployment system provided by an embodiment of the present invention.

[0046] Figure 4 A schematic flowchart of a dynamic service deployment method for an Internet of Vehicles provided by an embodiment of the present invention;

[0047] Figure 5 A schematic structural diagram of a dynamic service deployment system for an Internet of Vehicles provided by an embodiment of the present invention;

[0048] Figure 6 A schematic structural diagram of a dynamic service deployment device for an Internet of Vehicles provided by an embodiment of the present invention. Detailed Embodiments

[0049] This part will describe in detail the specific embodiments of the present invention. The preferred embodiments of the present invention are shown in the drawings. The role of the drawings is to supplement the description in the text part of the specification, enabling people to intuitively and vividly understand each technical feature and the overall technical solution of the present invention. However, it should not be construed as a limitation on the protection scope of the present invention.

[0050] In the description of the embodiments of the present invention, the meaning of "several" is one or more, the meaning of "multiple" is two or more, "greater than", "less than", "exceeding", etc. are understood as not including the present number, "above", "below", "within", etc. are understood as including the present number, "at least one" means one or more, and "at least one of the following" and its similar expressions refer to any combination of these items, including any combination of single items or plural items. If there is a description of "first", "second", etc., it is only for the purpose of distinguishing technical features and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence relationship of the indicated technical features.

[0051] It should be noted that in the embodiments of the present invention, words such as "set", "installed", "connected", etc. should be understood in a broad sense. Those skilled in the art can reasonably determine the specific meanings of the above words in the embodiments of the present invention in combination with the specific content of the technical solution. For example, the term "connected" can be a mechanical connection, an electrical connection, or a connection that allows mutual communication; it can be directly connected or indirectly connected through an intermediate medium.

[0052] In the description of the embodiments of the present invention, the descriptions referring to terms such as "one embodiment / embodiment mode", "another embodiment / embodiment mode", or "certain embodiments / embodiment modes", "in the above embodiments / embodiment modes", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least two embodiments or embodiment modes of the present disclosure. In the present disclosure, the schematic expressions of the above terms do not necessarily refer to the same embodiment or embodiment mode. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or embodiment modes.

[0053] It should be noted that the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0054] The connotation of the Internet of Vehicles mainly refers to: on-vehicle devices on vehicles effectively utilize all vehicle dynamic information in the information network platform through wireless communication technology to provide different functional services during vehicle operation, such as navigation, remote driving, and automatic parking, etc.

[0055] In the current related technologies, most use static solution plans (SSPs) for service deployment, which are ineffective for the mobile and dynamic scenarios of the IoV network. However, the number of vehicles requesting services and their distances from different edge servers are dynamic, and the real-time environment must be considered when mapping services to edge servers. And some related technologies only consider one-to-one deployment, with a fixed number of services and the availability of a single instance for each service. However, deploying one instance for one service can only serve a limited number of vehicles. If the demand for any specific service increases, multiple instances are required for each service.

[0056] To this end, the present application proposes a dynamic service deployment method, system, device and storage medium for the vehicle networking. By obtaining service requests and environmental data, and based on the service requests and the environmental data, generating a service deployment decision through a policy function. Then, obtaining a decision action and a decision response according to the service deployment decision. Next, generating a decision quality value of the service deployment decision through a value function according to the decision action and the decision response. Then, determining a target service deployment decision from the service deployment decisions according to the decision quality value. Subsequently, performing service deployment on the service according to the target service deployment decision. This method is more in line with the actual dynamic scenario, and can effectively reduce the response time of the service and maintain the resource balance of the edge server, so as to effectively cope with emergencies.

[0057] Referring to Figure 1 , in an embodiment of the present invention, a vehicle networking is provided, including a cloud server 101, a MEC node 102 and a terminal 103. Among them, the cloud server 101 can provide services such as highly aggregated centralized computing and storage. The MEC node 102 is located at the edge of the Internet and is close to the data source. The MEC node is connected to the Internet of Things, and nodes can communicate with each other to provide computing and storage functions. The terminal 103 can be sensors on the vehicle, various intelligent in-vehicle devices, which can sense, measure and collect relevant raw data according to the service requirements and upload it to the MEC node 102. Among them, MEC (Multi-Access Edge Computing) refers to migrating traffic and service computing from the centralized cloud to the network edge, closer to the user. All data is analyzed, processed and stored at the network edge instead of being sent to the cloud for processing. Collecting and processing data closer to the customer can reduce latency and provide real-time performance for high-bandwidth applications.

[0058] Referring to Figure 2 , in an embodiment of the present invention, a vehicle networking service deployment system is provided. The system includes a data collection module 201, a policy function module 202 and a value function module 203. Among them, the data collection module 201 is used to collect environmental data required for the service (such as: service requests, vehicle IDs, vehicle positions, signal propagation speeds, etc.); the policy function module 202 is used to receive the data collected by the data collection module 201 and the decision quality value generated by the value function module 203, and generate a service deployment policy according to the collected data; the value function module 203 is used to receive the service deployment policy of the policy function module 202, generate a decision quality value according to the service deployment policy, and determine the target service deployment policy through the decision quality value.

[0059] Referring to Figure 3 , Figure 3It is a schematic diagram of the modules of the vehicle networking service deployment system involved in various embodiments of the present application. In the embodiments of the present application, the vehicle networking service deployment system may include: a communication bus, a network communication module 301, a processor 302 (such as a Central Processing Unit, CPU), and a memory 303. Among them, the communication bus is used to realize the connection communication between these components; the memory 303 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. Optionally, the memory 303 may also be a storage device independent of the aforementioned processor 302. Those skilled in the art can understand that Figure 3 the hardware structure shown in it does not constitute a limitation to the present application, and may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.

[0060] The memory 303, as a readable storage medium, may include an operating system, application program modules, and a control program for the distributed platform. In Figure 3 this case, the network communication module 301 is mainly used to obtain an instruction issuing request and transmit the request to the processor 302; and the processor 302 may call the control program of the vehicle networking service deployment system stored in the memory 303 and execute the dynamic service deployment method of the vehicle networking provided in the embodiments of the present application.

[0061] It should be noted that only a part of the modules of the vehicle networking service deployment system are exemplarily given in the embodiments of the present application. The vehicle networking service deployment system may also include other components included in other existing vehicle networking service deployment system devices to implement corresponding functions, which are not specifically limited.

[0062] Based on Figure 2 and Figure 3 the hardware structure shown, as Figure 4 shown, this embodiment provides a dynamic service deployment method for vehicle networking, including but not limited to steps S401, S402, S403, S404, S405, and S406:

[0063] S401. Obtain a service request and environmental data.

[0064] It should be noted that the service request in step S401 refers to a request sent by in-vehicle devices (such as GPS global positioning systems, smart terminals, etc.) on the vehicle to the MEC node after receiving the service instructions required by the user. When the in-vehicle device sends a service request, the environmental data corresponding to the service will be sent to the MEC node at the same time. Among them, the environmental data includes vehicle ID, vehicle location, and the distance to the nearest MEC node, etc. The specific environmental data sent is determined according to the type of service request, and the environmental data can be collected through sensors on the vehicle, GPS global positioning systems, etc., and the data is cleaned and integrated through the processor of the in-vehicle smart terminal. Specifically, the user obtains the service of "automatic parking" by performing a certain interaction operation (such as clicking, touching, gestures, input in the input box, selecting from the drop-down menu, etc.) on the display interface of the in-vehicle smart terminal, thereby triggering the "request automatic parking service" instruction to the processor of the terminal. After receiving the operation instruction, the processor responds and processes it, and then sends the environmental data and service request of the vehicle requesting the service to the MEC node through the communication module on the vehicle.

[0065] S402. Generate a service deployment decision through a policy function according to the service request and the environmental data.

[0066] Optionally, step S402 can be further divided into the following steps:

[0067] Extract the features of the environmental data;

[0068] Generate a service deployment decision through a policy function according to the service request and the features of the environmental data.

[0069] According to the previous description, when obtaining the service request, the corresponding environmental data will also be obtained. Before generating the service deployment policy, extracting the features of the environmental data can reduce the computing cost of the policy function module 202, and at the same time can also help the policy function module 202 generate a service deployment policy with a smaller service response.

[0070] Among them, extracting the features of the environmental data can be further divided into the following steps:

[0071] Obtain the resource consumption amount and total service time of the service request;

[0072] Obtain the available resources at the edge and the available resources in the cloud;

[0073] Obtain the edge node load according to the available resources at the edge, the available resources in the cloud, and the resource consumption amount;

[0074] Take the total service time and the edge node load as the features of the environmental data.

[0075] In the step of extracting the features of the environmental data, first, the resource consumption amount and the total service time of the service request need to be obtained. It can be understood that for different service requests, the resource consumption amount and the total service response time are different. Therefore, it is necessary to obtain the corresponding features of the service first for subsequent steps to call. In this embodiment, the total service time of the service (i.e., the total response time of the service) and the edge node load (i.e., the ratio of the resources occupied by the service to the available resources at the edge) are selected as the features corresponding to the service.

[0076] Specifically, the MEC computing system is modeled as an M / D / 1 queue according to the Markov random model. For service type s, it has an arrival rate λs with a uniform distribution, and the service processing rate is deterministic (providing services at a rate C). The total service delay observed when a vehicle requests service s from MEC node e refers to the total time from when the vehicle sends the service request s to when it receives the corresponding response from the MEC node.

[0077] For the edge node load, the edge node load refers to the ratio between the resources consumed by the service instance and the available resources at the edge. To obtain the edge node load, it is necessary to obtain the available resources at the edge and the available resources in the cloud. Among them, multiple MEC nodes form an edge network with a server of limited capacity, and the available resources at the edge refer to the capacity of the resources that this edge network can use. The available resources at the edge C e can be calculated by the following formula:

[0078]

[0079] where Re represents the virtual resource unit, N represents the total number of edge resource units, and n represents the nth edge resource unit.

[0080] And the available resources in the cloud refer to the capacity of the resources that can be used in the cloud server. The available resources in the cloud C c can be calculated by the following formula:

[0081]

[0082] where Rc represents the virtual resource unit, M represents the total number of resource units in the cloud, and m represents the mth edge resource unit.

[0083] Then, according to the available resources at the edge, the available resources in the cloud, and the resource consumption amount, the edge node load (i.e., the load of each MEC node) can be calculated.

[0084] For the total service time, the total service time refers to the total time from when the service instance is initiated to when it gets a response. The total service time consists of the propagation delay and the queuing delay, that is: Since the average propagation delay is the ratio of the distance to the propagation speed on the medium, the total service time can be rewritten as the following formula:

[0085]

[0086] where dist(v, s) is the Euclidean distance between vehicle v and the MEC node deploying service s, and c is the propagation speed of the service request signal through the communication medium. Therefore, in this embodiment, it is necessary to obtain the signal propagation speed of the service request and also obtain the Euclidean distance between the edge node and the vehicle according to the environmental data.

[0087] After obtaining the above-mentioned parameters, the load on the MEC node can be obtained through the following formula, denoted by :

[0088]

[0089]

[0090] where C is the service rate, λs is the arrival rate of service requests, I S is the resources that service s is about to consume, and Rs represents the amount of resources consumed by deploying each service s at edge node E.

[0091] After obtaining the total service time of the service request and the edge node load as features, the policy function can generate the corresponding service deployment policy. Among them, the purpose of the service deployment policy is to minimize the service delay (i.e., reduce the total service time) and make the allocation of MEC resources at each moment as small and average as possible (i.e., the edge node load distribution of each node is the smallest and average). The policy function module can choose to use integer linear programming to implement the service deployment policy, and those skilled in the art can choose a suitable algorithm to implement the service deployment policy according to the actual situation, which is not specifically limited here.

[0092] Optionally, in some embodiments, generating a service deployment decision on the edge server according to the service request and the environmental data through the policy function includes at least one of the following:

[0093] Generating a service deployment decision according to the service request and the environmental data through the policy function in the edge node;

[0094] Or, generating a service deployment decision according to the service request and the environmental data through the policy function in the cloud server.

[0095] In this embodiment, the policy function module 202 can be set on the cloud server, or on the edge server (MEC node), or the policy function module 202 can be separately set on the cloud server and the edge server (MEC node) according to the functional requirements. If it is set on the cloud server, the powerful computing power of the cloud server can be used to save the computing resources of the edge server, so as to relieve the computing pressure of the edge server; if it is set on the edge server, the service can be responded to faster, thereby further reducing the total response time required for the service. Those skilled in the art can adjust it according to the actual situation to make the setting of the policy function module 202 more in line with the actual requirements.

[0096] S403. Obtain a decision action and a decision response according to the service deployment decision.

[0097] In step S403, it is necessary to obtain the decision action and the decision response corresponding to the service deployment decision according to the service deployment decision described above. The policy made by each service deployment decision is the decision action corresponding to the service deployment decision, and the response or effect obtained by the decision action is called the decision response corresponding to the service deployment decision.

[0098] S404. Generate a decision quality value of the service deployment decision through a value function according to the decision action and the decision response.

[0099] The task of the value function module is to evaluate the performance of the policy function module according to the actions taken and the responses received. The value function is responsible for calculating the decision quality value Q(ω,α) of the decision made by the policy function module. A high decision quality value Q(ω,α) means a high-quality decision. Therefore, an action with the maximum quality value must be selected, α = arg maxQ(ω,α). Where ω is a set of data sets obtained from the environment, such as: ω = {[v1,loc1,s],[v2,loc2,s]…[v n ,loc n ,S]}, where each parameter represents the service request s at time t, the set of vehicle IDs {v1,v2…v n}, the set of vehicle positions {loc1,loc2…loc n}, and α represents the action taken by the service deployment decision at time t.

[0100] S405. Determine a target service deployment decision from the service deployment decisions according to the decision quality value.

[0101] According to the foregoing description, the value function module continuously corrects the service deployment decisions generated by the policy function module, so that the result of the service deployment decision minimizes the service latency (i.e., reduces the total service time) and makes the allocation of MEC resources at each moment as small and average as possible (i.e., the load distribution of edge nodes at each node is minimized and averaged). In this embodiment, by setting the loss function of the value function module, taking the global minimum value of the loss function as the training end condition, the loss function is as follows:

[0102]

[0103] where σ(Ds,R(ω,a)) is the standard deviation between the delay threshold and the reward. The greater the deviation, the better the model is in terms of delay.

[0104] where is a target value, and the calculation formula is

[0105]

[0106] The value function module updates its parameter θ to minimize the mean square loss function

[0107] When the service deployment decision generated by the policy function module makes the loss function reach the global minimum value, the correction of the policy function module by the value function module is ended, and the service deployment decision that makes the loss function reach the global minimum value is used as the target service deployment decision.

[0108] S406. Perform service deployment on the service according to the target service deployment decision.

[0109] According to the foregoing steps, the required target service deployment decision has been obtained. In step S406, only by performing service deployment according to the target service deployment decision can the dynamic service deployment method of the vehicle networking in the embodiment of the present application be implemented.

[0110] Optionally, in an embodiment of the present application, the service deployment method further includes:

[0111] Determine that the service has completed the response, delete the current service on the edge server and upload it to the cloud server.

[0112] In this embodiment, the current service refers to the service that has completed the response. When the edge server (i.e., the MEC node) determines that the service has completed the response, the service will be deleted and transmitted back to the cloud server. Due to the limitation of the MEC node capacity, the service will only be deployed at the edge when there is a demand for the service. In the case of no demand, the edge node will delete the service instance s from its resources and transmit it back to the cloud to reduce the load of the MEC node and provide better performance for new services.

[0113] Based on the foregoing description, taking the case where the policy function module is set on the edge server as an example, as Figure 4 shown, a dynamic service deployment method for an Internet of Vehicles according to an embodiment of the present application specifically includes the following steps:

[0114] Obtain a service request and environmental data.

[0115] According to the service request and the environmental data, generate a service deployment decision through a policy function set on the edge server.

[0116] Obtain a decision action and a decision response according to the service deployment decision.

[0117] According to the decision action and the decision response, generate a decision quality value of the service deployment decision through a value function.

[0118] Determine a target service deployment decision from the service deployment decisions according to the decision quality value.

[0119] Perform service deployment on the service according to the target service deployment decision.

[0120] Exemplarily, a user can obtain the service of "automatic parking" by performing a certain interaction operation (such as clicking, touching, gesturing, inputting in an input box, selecting from a drop-down menu, etc.) on the display interface of an in-vehicle intelligent terminal, thereby triggering an instruction of "requesting automatic parking service" to the processor of the terminal. After receiving the operation instruction, the processor responds and processes it, and then sends the environmental data and service request of the vehicle requesting the service to the MEC node through the communication module on the vehicle. The MEC node generates a service deployment decision through a policy function set on the MEC node according to the data sent by the terminal. Subsequently, a decision action and a decision response are obtained according to the service deployment decision, and the value function module will conduct an objective evaluation on the generated service deployment decision to obtain a decision quality value. The decision quality value will be returned to the policy function module to further correct the generated service deployment decision until the loss function on the value function module reaches the global minimum. The service deployment decision that makes the loss function reach the global minimum is used as the target service deployment decision, and service deployment is performed according to the target service deployment decision.

[0121] As can be seen from the above, this application obtains a service request and environmental data, and then generates a service deployment decision through a policy function according to the service request and the environmental data, so as to generate a preliminary service deployment policy for subsequent steps to correct it. Then, a decision action and a decision response are obtained according to the service deployment decision, and a decision quality value of the service deployment decision is generated through a value function according to the decision action and the decision response, so as to objectively evaluate the generated service deployment policy. Then, a target service deployment decision is determined from the service deployment decisions according to the decision quality value, and the service deployment policy is corrected through the decision quality value to obtain a policy with better deployment effect. Then, the service is deployed according to the target service deployment decision, which can make the response time of the service shorter and maintain the resource balance of the edge server, so as to effectively cope with emergencies.

[0122] Referring to Figure 5 , a dynamic service deployment system for an Internet of Vehicles proposed in an embodiment of the present invention includes:

[0123] A first module 501, configured to obtain a service request and environmental data;

[0124] A second module 502, configured to generate a service deployment decision through a policy function according to the service request and the environmental data;

[0125] A third module 503, configured to obtain a decision action and a decision response according to the service deployment decision;

[0126] A fourth module 504, configured to generate a decision quality value of the service deployment decision through a value function according to the decision action and the decision response;

[0127] A fifth module 505, configured to determine a target service deployment decision from the service deployment decisions according to the decision quality value;

[0128] A sixth module 506, configured to deploy the service according to the target service deployment decision.

[0129] The content in the above method embodiments is applicable to the system embodiments of the present invention. The functions specifically implemented by the system embodiments of the present invention are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0130] Referring to Figure 6 , an embodiment of the present invention provides a dynamic service deployment device for an Internet of Vehicles, including:

[0131] At least one processor 601;

[0132] At least one memory 602, configured to store at least one program;

[0133] When the at least one program is executed by the at least one processor 601, it causes the at least one processor 601 to implement Figure 4 the dynamic service deployment method of the vehicle networking shown.

[0134] The content in the above method embodiments is applicable to the device embodiments of the present invention. The functions specifically implemented by the device embodiments of the present invention are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0135] The embodiments of the present invention also provide a storage medium, in which instructions executable by a processor are stored. The instructions executable by the processor are used to implement Figure 4 the dynamic service deployment method of the vehicle networking shown.

[0136] The above has specifically described the preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the present invention, and these equivalent deformations or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A dynamic service deployment method for an Internet of Vehicles, characterized in that It includes the following steps: Obtain a service request and environmental data; Generate a service deployment decision through a policy function according to the service request and the environmental data; Obtain a decision action and a decision response according to the service deployment decision; Generate a decision quality value of the service deployment decision through a value function according to the decision action and the decision response; Determine a target service deployment decision from the service deployment decisions according to the decision quality value; Perform service deployment on the service according to the target service deployment decision; The service request and the environmental data generate a service deployment decision on an edge server through a policy function, specifically including: Obtain the resource consumption amount and the total service time of the service request; Obtain the available resources on the edge and the available resources in the cloud; Obtain the edge node load according to the available resources on the edge, the available resources in the cloud, and the resource consumption amount; Take the total service time and the edge node load as the features of the environmental data; Generate a service deployment decision through a policy function according to the service request and the features of the environmental data; Among them, the goal of the policy function is to minimize the total service time as much as possible and make the edge node load distribution of each node the smallest and average.

2. The dynamic service deployment method for the vehicle networking according to claim 1, wherein The environmental data includes a vehicle ID and a vehicle location. The obtaining of the service request and the environmental data specifically includes: Receive the vehicle ID, the vehicle location, and the service request sent through a communication module on the vehicle.

3. The dynamic service deployment method for the vehicle networking according to claim 1, wherein The obtaining of the total service time of the service request according to the environmental data specifically includes: Obtain the signal propagation speed of the service request; Obtain the Euclidean distance between the edge node and the vehicle according to the environmental data; Determine the total service time of the service request according to the propagation speed and the Euclidean distance.

4. The dynamic service deployment method for the vehicle networking according to claim 1, wherein After the step of performing service deployment on the service according to the target service deployment decision, the method further includes: Determine that the service has completed the response, delete the current service on the edge server and upload it to the cloud server.

5. The dynamic service deployment method for an Internet of Vehicles according to claim 1, wherein The generating of a service deployment decision on an edge server through a policy function according to the service request and the environmental data includes at least one of the following: Generate a service deployment decision through a policy function in an edge node according to the service request and the environmental data; Or Generate a service deployment decision through a policy function in a cloud server according to the service request and the environmental data.

6. A dynamic service deployment system for an Internet of Vehicles, characterized in that It includes: A first module for obtaining a service request and environmental data; A second module for generating a service deployment decision through a policy function according to the service request and the environmental data; A third module for obtaining a decision action and a decision response according to the service deployment decision; A fourth module for generating a decision quality value of the service deployment decision through a value function according to the decision action and the decision response; A fifth module for determining a target service deployment decision from the service deployment decisions according to the decision quality value; A sixth module for performing service deployment on the service according to the target service deployment decision; The service request and the environment data generate a service deployment decision on the edge server through a policy function, specifically including: Obtain the resource consumption amount and the total service time of the service request; Obtain the available resources at the edge and the available resources in the cloud; Obtain the edge node load according to the available resources at the edge, the available resources in the cloud, and the resource consumption amount; Use the total service time and the edge node load as the features of the environment data; Generate a service deployment decision through the policy function according to the service request and the features of the environment data; Among them, the goal of the policy function is to minimize the total service time as much as possible, and make the edge node load distribution of each node the smallest and average.

7. A dynamic service deployment device for an Internet of Vehicles, characterized in that Including: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the dynamic service deployment method for the vehicle networking as described in any one of claims 1-5.

8. A computer-readable storage medium storing instructions executable by a processor, characterized in that, The instructions executable by the processor are used to implement the dynamic service deployment method for the vehicle networking as described in any one of claims 1-5 when executed by the processor.

Images