Fog computing resource intelligent allocation method and system and medium

Abstract

The invention relates to the technical field of data exchange networks, in particular to a fog computing resource intelligent allocation method and system and a medium. The method comprises the following steps: acquiring various service request data of a vehicle terminal, time efficiency priorities of different service request data in a fog computing platform, historical bandwidth resource data and a network speed at a sampling moment; based on the size of historical bandwidth resource data allocated to the same kind of service request data of the same vehicle terminal in history, determining a computing resource consumption degree; determining a fog resource demand degree in combination with the timeliness priority and the computing resource consumption degree; determining a fog resource tendency degree according to the network speed of the fog computing platform and the allocable bandwidth; and determining allocation weights in combination with the fog resource demand degree and the fog resource tendency degree, and allocating fog node computing resources according to the allocation weights of different vehicle terminals at the current moment. According to the invention, the distribution efficiency of fog node computing resources can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data exchange networks, and in particular to a method, system and medium for intelligent allocation of fog computing resources. Background Art

[0002] With the development of Internet of Vehicles technology, a rich variety of vehicle service businesses have emerged. Each vehicle terminal sends a service request to the fog platform. The distributed architecture of the fog computing platform uses key technologies such as middleware and intelligent sensing access to provide computing and data processing capabilities through the edge network, and allocates the same computing resources to service requests. This not only improves the real-time and responsiveness of the system, but also reduces the burden on cloud servers, thereby supporting the development of diversified applications in the Internet of Vehicles.

[0003] In the traditional IoV fog computing resource allocation process, each vehicle terminal acts as a fog node. The fog node sends multiple service requests to the fog computing platform in real time. The fog computing platform allocates corresponding bandwidth resources to the fog node according to the real-time requirements of the service requests. In this way, due to the different computing resource requirements of each service request, such as the video playback in the in-vehicle entertainment requires more bandwidth resources than the status monitoring, the traditional method only analyzes the real-time nature of the request to obtain low resource allocation efficiency. At the same time, the status of the fog computing platform and the cloud server platform is constantly changing, which further leads to the traditional allocation method. When the fog computing platform has better allocable performance than the cloud server platform, the fog node obtains more bandwidth resources from the cloud server platform, which is a poor allocation result, reducing the overall resource allocation efficiency of fog computing. Summary of the invention

[0004] In order to solve the technical problem of insufficient overall resource allocation efficiency of fog computing, the present invention provides a fog computing resource intelligent allocation method, system and medium. The technical solutions adopted are as follows:

[0005] The present invention proposes a method for intelligently allocating fog computing resources, wherein the fog nodes of the fog computing platform are vehicle terminals, and the method comprises:

[0006] Obtain various types of service request data of vehicle terminals, as well as the timeliness priority and allocated historical bandwidth resource data of different service request data in the fog computing platform, and obtain the network speed of the fog computing platform at all sampling times before the current time;

[0007] Based on the size of historical bandwidth resource data allocated to the same type of service request data in the history of the same vehicle terminal, determine the degree of computing resource consumption of the vehicle terminal in the bandwidth allocation of service request data; combine the timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal to determine the fog resource demand degree of the vehicle terminal;

[0008] Determine the bandwidth that can be allocated to the vehicle terminal under the fog computing platform at the current moment, and determine the degree of fog resource inclination based on the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated to the fog computing platform at the current moment;

[0009] The allocation weight is determined in combination with the fog resource demand degree and the fog resource tendency degree, and the fog node computing resources are allocated according to the allocation weights of different vehicle terminals at the current moment.

[0010] Further, the determining the computing resource consumption degree of the vehicle terminal in the bandwidth allocation of the service request data based on the historical bandwidth resource data size allocated to the same type of service request data of the same vehicle terminal in history includes:

[0011] Taking the average of the historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history as the bandwidth consumption prediction indicator;

[0012] Determine the bandwidth consumption stability index according to the numerical dispersion degree of all historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history;

[0013] The product of the bandwidth consumption prediction index and the bandwidth consumption stability index is calculated, and the maximum and minimum values ​​are normalized to obtain the computing resource consumption degree of the vehicle terminal when allocating bandwidth for data request of this type of service.

[0014] Furthermore, the bandwidth consumption stability index is determined according to the numerical discreteness of all historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history, including:

[0015] Calculate the standard deviation of all historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history;

[0016] The inverse of the standard deviation is normalized and used as a bandwidth consumption stability indicator for the same type of service request data of the same vehicle terminal in history.

[0017] Furthermore, the timeliness priority is represented by a numerical value. The larger the value of the timeliness priority is, the higher the priority of the corresponding service request data is. The timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal are combined to determine the fog resource demand degree of the vehicle terminal, including:

[0018] Calculating the opposite of the computing resource consumption level as a demand instability indicator;

[0019] The product value of the timeliness priority and the demand instability index is normalized to the maximum and minimum values ​​to obtain the fog resource demand degree.

[0020] Furthermore, the determination of the fog resource tendency degree according to the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated under the fog computing platform at the current moment includes:

[0021] Determine the neighboring network fluctuation index at the current moment according to the numerical fluctuation of the network speed at a preset number of sampling moments before the current moment and the numerical fluctuation of the network speed at all sampling moments;

[0022] The product of the neighboring network fluctuation index and the allocatable bandwidth is calculated, and the maximum and minimum values ​​of the inverse of the product are normalized as the fog resource tendency degree.

[0023] Further, the determining of the neighboring network fluctuation index at the current moment according to the numerical fluctuation of the network speed at a preset number of sampling moments before the current moment and the numerical fluctuation of the network speed at all sampling moments includes:

[0024] Calculate the numerical standard deviation of the network speed at a preset number of sampling moments before the current moment as a first standard deviation;

[0025] Calculate the numerical standard deviation of the network speed at all sampling moments before the current moment as the second standard deviation;

[0026] The ratio of the first standard deviation to the second standard deviation is used as a neighboring network fluctuation index.

[0027] Furthermore, the determining of the allocation weight in combination with the fog resource demand degree and the fog resource tendency degree includes:

[0028] The product value of the fog resource demand degree and the fog resource tendency degree is calculated, and normalized to obtain the allocation weight.

[0029] Furthermore, the allocation of fog node computing resources according to the allocation weights of different vehicle terminals at the current moment includes:

[0030] The sum of the allocation weights of the service request data of all vehicle terminals belonging to the same fog computing platform at the current moment is used as the resource demand indicator of the corresponding fog computing platform;

[0031] The resource demand indicators are sorted in descending order, and resource allocation of different fog computing platforms is achieved according to the sorting results.

[0032] On the other hand, the present invention also provides a fog computing resource intelligent allocation system, wherein the fog node of the fog computing platform is a vehicle terminal, and the system comprises:

[0033] An acquisition module is used to acquire various types of service request data of vehicle terminals, as well as the timeliness priority and allocated historical bandwidth resource data of different service request data in the fog computing platform, and to acquire the network speed of the fog computing platform at all sampling moments before the current moment;

[0034] The demand analysis module is used to determine the computing resource consumption degree of the vehicle terminal in the bandwidth allocation of service request data based on the historical bandwidth resource data size allocated to the same type of service request data of the same vehicle terminal in history; and determine the fog resource demand degree of the vehicle terminal in combination with the timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal;

[0035] The tendency analysis module is used to determine the bandwidth that can be allocated to the vehicle terminal under the fog computing platform at the current moment, and determine the degree of fog resource tendency according to the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated to the fog computing platform at the current moment;

[0036] The allocation module is used to determine the allocation weights based on the fog resource demand degree and the fog resource tendency degree, and allocate fog node computing resources according to the allocation weights of different vehicle terminals at the current moment.

[0037] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the steps of any of the aforementioned methods for intelligent allocation of fog computing resources.

[0038] The present invention has the following beneficial effects:

[0039] The embodiment of the present invention obtains various types of service request data, time priority, historical bandwidth resource data allocated and network speed of fog computing platform at all sampling times before the current time of vehicle terminal, performs status analysis of different fog computing platforms, and provides data basis for subsequent bandwidth resource allocation. Compared with directly according to the time sequence of service requests, this solution effectively allocates resources through status analysis to improve the efficiency of overall fog computing resource intelligent allocation. When performing status analysis, firstly, the computing resource consumption degree is determined based on the historical bandwidth resource data allocated by the same type of service request data of the same vehicle terminal in history; the fog resource demand degree is determined in combination with the time priority and computing resource consumption degree. The fog resource demand degree indicates the demand situation when the fog computing platform allocates resources. The larger the value, the more the fog computing platform is needed for processing. At this time, more bandwidth resources need to be allocated; then, the fog resource tendency degree is determined in combination with the network speed and the bandwidth that can be allocated under the fog computing platform, indicating the tendency of bandwidth resource allocation. Therefore, the allocation weight can be determined in combination with the fog resource demand degree and the fog resource tendency degree, and the fog node computing resources are allocated according to the allocation weights of different vehicle terminals at the current time. Compared with directly allocating resources in a sequential manner based on request time, this solution combines the multi-dimensional data of the fog resource platform to perform status analysis on the fog resource platform, thereby allocating bandwidth resources more accurately and reasonably, avoiding the vehicle terminal's overall service request response being untimely due to bandwidth delays, and improving the overall computing resource allocation efficiency of the fog computing platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flow chart of a method for intelligent allocation of fog computing resources provided by an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of a multi-layer architecture of fog computing provided by an embodiment of the present invention;

[0043] Figure 3 A schematic diagram of the architecture of fog computing in a smart Internet of Vehicles application provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a fog computing resource intelligent allocation method, system and medium proposed by the present invention, its specific implementation method, structure, features and effects in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0046] It should be noted that in order to ensure that the calculation results are meaningful, when the embodiments of the present invention perform fractional operations, when the denominator is 0, it is necessary to add a parameter adjustment factor greater than 0 (such as 1) to the denominator to prevent the denominator from being 0. The value of the parameter adjustment factor is set by the implementer according to actual conditions, and this application does not impose any special restrictions.

[0047] In order to enable those skilled in the art to understand the content of the present invention, some basic terms will be explained and elaborated in detail below.

[0048] Fog computing is a computing architecture that is often seen as an extension of edge computing. It distributes computing, storage, network and other resources from the cloud to places closer to users and devices, such as the network edge or local servers. This distributed computing model can provide lower latency, higher bandwidth, stronger processing power and better privacy protection. Fog computing platforms are often used in the fields of Internet of Things (IoT), smart cities, smart transportation, industrial automation, etc.

[0049] The following is a detailed description of a specific solution of a fog computing resource intelligent allocation method provided by the present invention in conjunction with the accompanying drawings.

[0050] See also Figure 1 , which shows a flow chart of a fog computing resource intelligent allocation method provided by an embodiment of the present invention, the method comprising:

[0051] S101: Obtain various types of service request data from vehicle terminals, as well as the timeliness priority and allocated historical bandwidth resource data of different service request data in the fog computing platform, and obtain the network speed of the fog computing platform at all sampling times before the current time.

[0052] It can be understood that the specific scenario of the embodiment of the present invention is the application scenario of fog computing in the Internet of Vehicles, where each vehicle terminal is a fog node, and the service request data is specifically the fog node sending real-time GPS navigation, in-vehicle entertainment, status monitoring and other service requests to the fog computing platform.

[0053] See also Figure 2 and Figure 3 , Figure 2 A schematic diagram of a multi-layer architecture of fog computing provided by an embodiment of the present invention; Figure 3 The schematic diagram of the architecture of fog computing in the smart car networking application provided by one embodiment of the present invention. The fog computing platform allocates computing resources according to the real-time demand characteristics of the service request. If the real-time requirement is high, the fog computing platform allocates more fog computing resources to the fog node, otherwise it allocates fewer fog computing resources, transmits the service request to the cloud, and the cloud allocates the remaining computing resources.

[0054] The computing resources and fog resources in this application can be specifically bandwidth resources. In actual vehicle networking application scenarios, the computing resource requirements of each service request are different. For example, video playback in car entertainment requires more bandwidth resources than state monitoring. If the car entertainment is requested in advance and the state monitoring follows closely, the car entertainment will be allocated more bandwidth resources and the state monitoring will be allocated less bandwidth resources. Therefore, in the related technology, the traditional fog computing resource allocation method only obtains the allocation result based on the real-time characteristics of the service request, which reduces the overall computing resource allocation efficiency of the fog computing platform.

[0055] Therefore, the present application allocates resources at the current moment by combining various types of service request data, the timeliness priorities of different service request data and the allocated historical bandwidth resource data, as well as the network speed at the sampling moment.

[0056] S102: Based on the size of historical bandwidth resource data allocated to the same type of service request data by the same vehicle terminal in history, determine the degree of computing resource consumption of the vehicle terminal in bandwidth allocation for service request data; determine the degree of fog resource demand of the vehicle terminal in combination with the timeliness priority of the service request data and the degree of computing resource consumption of the vehicle terminal.

[0057] The computing resource consumption level indicates the bandwidth information that is predicted to be consumed by the vehicle terminal under the service request data. In combination with all historical data, the computing resources that the current vehicle terminal needs to consume when requesting service data are predicted to determine the computing resource consumption level.

[0058] Furthermore, in some embodiments of the present invention, based on the size of historical bandwidth resource data allocated to the same type of service request data in history by the same vehicle terminal, the degree of computing resource consumption of the vehicle terminal in bandwidth allocation of service request data is determined, including: taking the mean of the historical bandwidth resource data allocated to the same type of service request data in history by the same vehicle terminal as a bandwidth consumption prediction index; determining a bandwidth consumption stability index based on the numerical discreteness of all historical bandwidth resource data allocated to the same type of service request data in history by the same vehicle terminal; calculating the product of the bandwidth consumption prediction index and the bandwidth consumption stability index, and performing maximum and minimum normalization processing to obtain the computing resource consumption degree of the vehicle terminal when allocating bandwidth for this type of service request data.

[0059] In this application, the specific analysis of the computing resource consumption degree is mainly through two parts, one is the bandwidth consumption prediction index, and the other is the bandwidth consumption stability index.

[0060] The bandwidth consumption prediction index is the bandwidth index predicted for the service request data to be allocated. In the present application, the bandwidth consumption prediction index is determined directly by taking the average of all historical bandwidth resource data allocated to the same type of service request data in history.

[0061] The bandwidth consumption stability index represents the stability of the numerical fluctuation of all historical bandwidth resource data allocated to the same type of service request data in history. This application specifically calculates the stability based on the degree of numerical dispersion.

[0062] Because there are differences in the bandwidth resources consumed by different service requests of the same category, for example, the bandwidth consumption for GPS navigation is basically the same, while the bandwidth consumption of different requests in the in-vehicle entertainment category may vary greatly. Therefore, the smaller the difference in bandwidth consumption between service requests of the same category in the historical records, the higher the bandwidth consumption stability index of the bandwidth consumption prediction level.

[0063] Furthermore, in some embodiments of the present invention, a bandwidth consumption stability index is determined based on the numerical discreteness of all historical bandwidth resource data historically allocated to the same type of service request data of the same vehicle terminal, including: calculating the standard deviation of all historical bandwidth resource data historically allocated to the same type of service request data of the same vehicle terminal; normalizing the opposite of the standard deviation as the bandwidth consumption stability index of the same type of service request data of the same vehicle terminal in history.

[0064] Specifically, the numerical dispersion degree in the present application is specifically reflected in the form of standard deviation. Of course, in other embodiments of the present invention, other discrete calculation methods such as variance may also be used, and there is no limitation to this.

[0065] In the embodiment of the present invention, since the larger the value of the standard deviation is, the larger the fluctuation of the values ​​of all historical bandwidth resource data is, that is, the more unstable it is, the standard deviation and the bandwidth consumption stability index are negatively correlated, and the opposite of the standard deviation needs to be normalized to obtain the bandwidth consumption stability index. The normalization can be analyzed according to the actual situation, such as linear normalization of the maximum and minimum values, or nonlinear normalization using an activation function, and there is no limitation on this.

[0066] After determining the bandwidth consumption prediction index and the bandwidth consumption stability index, the product of the bandwidth consumption prediction index and the bandwidth consumption stability index is calculated, and the maximum and minimum values ​​are normalized to obtain the computing resource consumption level of the vehicle terminal when allocating bandwidth for data request of this type of service.

[0067] The computing resource consumption degree can effectively characterize the consumption characteristics of the computing resources to be consumed, that is, the larger the value of the computing resource consumption degree is, the greater the overall consumption of bandwidth resources is, and because the fluctuation is stable, the numerical reliability of the computing resource consumption degree is higher.

[0068] In an embodiment of the present invention, the analysis can also be performed in combination with the timeliness requirement of the service request data itself. The timeliness requirement is specifically based on a pre-set timeliness priority. The timeliness priority is represented by a numerical value. The larger the numerical value of the timeliness priority, the higher the priority of the corresponding service request data.

[0069] For example, during driving, the timeliness priority of the service request data corresponding to GPS navigation is level 5, which is greater than the timeliness priority of level 1 of the service request data corresponding to in-vehicle entertainment.

[0070] The fog resource demand degree indicates the real-time demand for bandwidth resources of the fog computing platform. The larger the value, the greater the real-time demand of the fog computing platform.

[0071] Under the overall fog computing platform architecture, each fog computing platform can freely allocate certain computing resources. The fog computing platform can also directly parse and transmit the service request data to the cloud, which will redistribute it to other fog computing platforms for processing and transmit it to the vehicle terminal of the original fog computing platform. Under this logic, real-time performance cannot be guaranteed. Therefore, it is necessary to analyze the degree of fog resource demand based on the timeliness priority and computing resource consumption of each service request data.

[0072] Furthermore, in some embodiments of the present invention, the fog resource demand degree of the vehicle terminal is determined in combination with the timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal, including: taking the opposite of the computing resource consumption degree as the demand instability index; normalizing the product value of the timeliness priority and the demand instability index to the maximum and minimum values ​​to obtain the fog resource demand degree.

[0073] It is understandable that the lower the computing resource consumption of the service request issued by the fog node, and the higher the real-time requirement, the higher the fog resource demand of the fog computing platform. The lower the computing resource consumption, the smaller the computing resources required, and the higher the timeliness requirement, the more the fog computing platform needs to process directly. Therefore, the product of the inverse of the computing resource consumption and the timeliness priority is directly normalized as the fog resource demand.

[0074] If the timeliness priority of a single fog node sending a service request is higher and the corresponding computing resource consumption of the fog node is lower, the computing resource allocation task of the fog node will tend to be delivered to the fog computing platform rather than the cloud.

[0075] S103: Determine the bandwidth that can be allocated to the vehicle terminal under the fog computing platform at the current moment, and determine the degree of fog resource inclination according to the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated to the fog computing platform at the current moment.

[0076] It is understandable that each fog computing platform has different allocable bandwidth at the current moment. Therefore, it is necessary to specifically determine the bandwidth that can be allocated to the vehicle terminal under the fog computing platform at the current moment, and then, combine the network speed of the fog computing platform at all sampling moments before the current moment to allocate the resources of the fog computing platform. Among them, the fog resource tendency degree represents the tendency of allocating the resources of the fog computing platform.

[0077] Furthermore, in some embodiments of the present invention, the degree of fog resource tendency is determined based on the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated to the fog computing platform at the current moment, including: determining the neighboring network fluctuation index at the current moment based on the numerical fluctuation of the network speed at a preset number of sampling moments before the current moment and the numerical fluctuation of the network speed at all sampling moments; calculating the product of the neighboring network fluctuation index and the allocatable bandwidth, and normalizing the maximum and minimum values ​​of the inverse of the product as the fog resource tendency degree.

[0078] The preset number may be specifically 10, that is, the network speed is collected once every 1 second, and the numerical fluctuation of the network speed 10 seconds before the current moment is determined as the network fluctuation at the adjacent moment.

[0079] Then, the numerical fluctuations of the network data at all sampling moments are regarded as the overall fluctuations, and the network fluctuations at the current moment can be analyzed based on the network fluctuations at adjacent moments and the overall fluctuations.

[0080] Furthermore, in some embodiments of the present invention, the neighboring network fluctuation index at the current moment is determined based on the numerical fluctuation of the network speed at a preset number of sampling moments before the current moment and the numerical fluctuation of the network speed at all sampling moments, including: calculating the numerical standard deviation of the network speed at a preset number of sampling moments before the current moment as the first standard deviation; calculating the numerical standard deviation of the network speed at all sampling moments before the current moment as the second standard deviation; and taking the ratio of the first standard deviation to the second standard deviation as the neighboring network fluctuation index.

[0081] The first standard deviation represents the network fluctuation before the current moment. The larger the value of the first standard deviation, the greater the possibility of network fluctuation at the current moment. The second standard deviation represents the overall network fluctuation. Therefore, by calculating the ratio of the first standard deviation to the second standard deviation, the larger the value, the greater the network fluctuation at the current moment. Therefore, the larger the neighboring network fluctuation index, the more severe the network fluctuation at the current moment.

[0082] In the embodiment of the present invention, since the more severe the network fluctuation is, the worse the overall network environment is, and the tendency of bandwidth resource allocation needs to be reduced, and at the same time, since the more allocatable bandwidth is, the more surplus resources the fog computing platform has, the resource allocation tendency needs to be reduced. Therefore, in the embodiment of the present invention, the product of the neighboring network fluctuation index and the allocatable bandwidth is calculated, and the maximum and minimum values ​​of the inverse of the product are normalized as the degree of fog resource tendency.

[0083] S104: Determine the allocation weight based on the fog resource demand and fog resource tendency, and allocate fog node computing resources according to the allocation weights of different vehicle terminals at the current moment.

[0084] Since the fog resource demand degree indicates the demand index for bandwidth resources, and the fog resource tendency degree indicates the allocation tendency of bandwidth resources, the embodiment of the present invention can determine the allocation weight by combining the fog resource demand degree and the fog resource tendency degree.

[0085] Furthermore, in some embodiments of the present invention, the allocation weight is determined in combination with the fog resource demand degree and the fog resource tendency degree, including: calculating the product value of the fog resource demand degree and the fog resource tendency degree, and normalizing it to obtain the allocation weight.

[0086] That is, the product value of the fog resource demand degree and the fog resource tendency degree is directly normalized to obtain the allocation weight. Of course, in other embodiments of the present invention, the sum of the fog resource demand degree and the fog resource tendency degree can also be normalized to obtain the allocation weight.

[0087] After determining the allocation weight of bandwidth resources, resource allocation is required, that is, the fog node computing resources are allocated according to the allocation weights of different vehicle terminals at the current moment.

[0088] Furthermore, in some embodiments of the present invention, the sum of the allocation weights of the service request data of all vehicle terminals belonging to the same fog computing platform at the current moment is used as the resource demand index of the corresponding fog computing platform; the resource demand indicators are sorted in descending order, and resource allocation of different fog computing platforms is implemented according to the sorting results.

[0089] The fog computing platform can request collaborative processing from the cloud, but when the fog computing platform itself is still in high demand for resources, it still needs to allocate more bandwidth resources to it. Therefore, it is still necessary to use the sum of the allocation weights of the service request data of all vehicle terminals belonging to the same fog computing platform at the current moment as the resource demand indicator of the corresponding fog computing platform. The larger the value of the resource demand indicator, the greater the amount of bandwidth resources allocated to the corresponding fog computing platform.

[0090] The embodiment of the present invention obtains various types of service request data, time priority, historical bandwidth resource data allocated and network speed of fog computing platform at all sampling times before the current time of vehicle terminal, performs status analysis of different fog computing platforms, and provides data basis for subsequent bandwidth resource allocation. Compared with directly according to the time sequence of service requests, this solution effectively allocates resources through status analysis to improve the efficiency of overall fog computing resource intelligent allocation. When performing status analysis, firstly, the computing resource consumption degree is determined based on the historical bandwidth resource data allocated by the same type of service request data of the same vehicle terminal in history; the fog resource demand degree is determined in combination with the time priority and computing resource consumption degree. The fog resource demand degree indicates the demand situation when the fog computing platform allocates resources. The larger the value, the more the fog computing platform is needed for processing. At this time, more bandwidth resources need to be allocated; then, the fog resource tendency degree is determined in combination with the network speed and the bandwidth that can be allocated under the fog computing platform, indicating the tendency of bandwidth resource allocation. Therefore, the allocation weight can be determined in combination with the fog resource demand degree and the fog resource tendency degree, and the fog node computing resources are allocated according to the allocation weights of different vehicle terminals at the current time. Compared with directly allocating resources in a sequential manner based on request time, this solution combines the multi-dimensional data of the fog resource platform to perform status analysis on the fog resource platform, thereby allocating bandwidth resources more accurately and reasonably, avoiding the vehicle terminal's overall service request response being untimely due to bandwidth delays, and improving the overall computing resource allocation efficiency of the fog computing platform.

[0091] On the other hand, an embodiment of the present invention further provides a fog computing resource intelligent allocation system, wherein the fog node of the fog computing platform is a vehicle terminal, and the system includes:

[0092] An acquisition module is used to acquire various types of service request data of vehicle terminals, as well as the timeliness priority and allocated historical bandwidth resource data of different service request data in the fog computing platform, and to acquire the network speed of the fog computing platform at all sampling moments before the current moment;

[0093] The demand analysis module is used to determine the computing resource consumption degree of the vehicle terminal in the bandwidth allocation of service request data based on the historical bandwidth resource data size allocated to the same type of service request data of the same vehicle terminal in history; and determine the fog resource demand degree of the vehicle terminal in combination with the timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal;

[0094] The tendency analysis module is used to determine the bandwidth that can be allocated to the vehicle terminal under the fog computing platform at the current moment, and determine the degree of fog resource tendency according to the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated to the fog computing platform at the current moment;

[0095] The allocation module is used to determine the allocation weights based on the fog resource demand degree and the fog resource tendency degree, and allocate fog node computing resources according to the allocation weights of different vehicle terminals at the current moment.

[0096] On the other hand, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the steps of the aforementioned method for intelligent allocation of fog computing resources are implemented.

[0097] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0098] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for intelligent allocation of fog computing resources, characterized in that: The fog node of the fog computing platform is a vehicle terminal, and the method includes: Obtain various types of service request data of vehicle terminals, as well as the timeliness priority and allocated historical bandwidth resource data of different service request data in the fog computing platform, and obtain the network speed of the fog computing platform at all sampling times before the current time; Based on the size of historical bandwidth resource data allocated to the same type of service request data in the history of the same vehicle terminal, determine the degree of computing resource consumption of the vehicle terminal in the bandwidth allocation of service request data; combine the timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal to determine the fog resource demand degree of the vehicle terminal; Determine the bandwidth that can be allocated to the vehicle terminal under the fog computing platform at the current moment, and determine the degree of fog resource inclination based on the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated to the fog computing platform at the current moment; The allocation weight is determined in combination with the fog resource demand degree and the fog resource tendency degree, and the fog node computing resources are allocated according to the allocation weights of different vehicle terminals at the current moment.

2. A fog computing resource intelligent allocation method as claimed in claim 1, characterized in that: The determining of the computing resource consumption degree of the vehicle terminal in bandwidth allocation of service request data based on the size of historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history includes: Taking the average of the historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history as the bandwidth consumption prediction indicator; Determine the bandwidth consumption stability index according to the numerical dispersion degree of all historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history; The product of the bandwidth consumption prediction index and the bandwidth consumption stability index is calculated, and the maximum and minimum values ​​are normalized to obtain the computing resource consumption degree of the vehicle terminal when allocating bandwidth for data request of this type of service.

3. A fog computing resource intelligent allocation method as claimed in claim 2, characterized in that: Determining the bandwidth consumption stability index according to the numerical discreteness of all historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history includes: Calculate the standard deviation of all historical bandwidth resource data allocated to the same type of service request data of the same vehicle terminal in history; The inverse of the standard deviation is normalized and used as a bandwidth consumption stability indicator for the same type of service request data of the same vehicle terminal in history.

4. The method for intelligent allocation of fog computing resources according to claim 1, characterized in that: The timeliness priority is represented by a numerical value. The larger the value of the timeliness priority is, the higher the priority of the corresponding service request data is. The timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal are combined to determine the fog resource demand degree of the vehicle terminal, including: Calculating the opposite of the computing resource consumption level as a demand instability indicator; The product value of the timeliness priority and the demand instability index is normalized to the maximum and minimum values ​​to obtain the fog resource demand degree.

5. The method for intelligent allocation of fog computing resources according to claim 1, characterized in that: The method of determining the fog resource tendency degree according to the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated under the fog computing platform at the current moment includes: Determine the neighboring network fluctuation index at the current moment according to the numerical fluctuation of the network speed at a preset number of sampling moments before the current moment and the numerical fluctuation of the network speed at all sampling moments; The product of the neighboring network fluctuation index and the allocatable bandwidth is calculated, and the maximum and minimum values ​​of the inverse of the product are normalized as the fog resource tendency degree.

6. A method for intelligent allocation of fog computing resources as claimed in claim 5, characterized in that: The determining of the neighboring network fluctuation index at the current moment according to the numerical fluctuation of the network speed at a preset number of sampling moments before the current moment and the numerical fluctuation of the network speed at all sampling moments includes: Calculate the numerical standard deviation of the network speed at a preset number of sampling moments before the current moment as a first standard deviation; Calculate the numerical standard deviation of the network speed at all sampling moments before the current moment as the second standard deviation; The ratio of the first standard deviation to the second standard deviation is used as a neighboring network fluctuation index.

7. A fog computing resource intelligent allocation method as claimed in claim 1, characterized in that: The step of determining the allocation weight in combination with the fog resource demand degree and the fog resource tendency degree includes: The product value of the fog resource demand degree and the fog resource tendency degree is calculated, and normalized to obtain the allocation weight.

8. The method for intelligent allocation of fog computing resources according to claim 1, characterized in that: The allocation of fog node computing resources according to the allocation weights of different vehicle terminals at the current moment includes: The sum of the allocation weights of the service request data of all vehicle terminals belonging to the same fog computing platform at the current moment is used as the resource demand indicator of the corresponding fog computing platform; The resource demand indicators are sorted in descending order, and resource allocation of different fog computing platforms is achieved according to the sorting results.

9. A fog computing resource intelligent allocation system, characterized in that: The fog node of the fog computing platform is a vehicle terminal, and the system includes: An acquisition module is used to acquire various types of service request data of vehicle terminals, as well as the timeliness priority and allocated historical bandwidth resource data of different service request data in the fog computing platform, and to acquire the network speed of the fog computing platform at all sampling moments before the current moment; The demand analysis module is used to determine the computing resource consumption degree of the vehicle terminal in the bandwidth allocation of service request data based on the historical bandwidth resource data size allocated to the same type of service request data of the same vehicle terminal in history; and determine the fog resource demand degree of the vehicle terminal in combination with the timeliness priority of the service request data and the computing resource consumption degree of the vehicle terminal; The tendency analysis module is used to determine the bandwidth that can be allocated to the vehicle terminal under the fog computing platform at the current moment, and determine the degree of fog resource tendency according to the network speed of the fog computing platform at all sampling moments before the current moment and the bandwidth that can be allocated to the fog computing platform at the current moment; The allocation module is used to determine the allocation weights based on the fog resource demand degree and the fog resource tendency degree, and allocate fog node computing resources according to the allocation weights of different vehicle terminals at the current moment.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the method for intelligent allocation of fog computing resources described in any one of claims 1 to 8 are implemented.

Images