Service scheduling method and device, electronic equipment and computer readable storage medium

Abstract

The present disclosure provides a service scheduling method and device, and electronic equipment and computer readable storage medium, the service scheduling method is applied to a management and control node in a cloud edge computing network, the cloud edge computing network also includes a plurality of edge computing nodes, comprising: obtaining network state information, service type information and service scheduling request times uploaded by each edge computing node in the cloud edge computing network; determining the activation utility value of each edge computing node according to the network state information, service type information and service scheduling request times uploaded by each edge computing node in the cloud edge computing network; determining a candidate service scheduling node among the edge computing nodes according to the activation utility value; activating the service scheduling module of the candidate service scheduling node, and taking the candidate service scheduling node as the current service scheduling node in the cloud edge computing network to process the service scheduling request in the cloud edge computing network. Through the embodiment of the present disclosure, the service request delay on the cloud edge computing network can be reduced.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to a service scheduling method and apparatus, electronic device and computer-readable storage medium. Background Technology

[0002] In a cloud-edge computing network, when the computing tasks of an edge computing node become too heavy, it will request business scheduling to offload the computing tasks to other nodes for computation.

[0003] In a cloud-edge computing network, the specific decision of who allocates and processes the computational offloading tasks for edge computing nodes to determine how each edge computing node offloads its computational tasks has a crucial impact on the speed and efficiency of the computational offloading tasks in the cloud-edge computing network.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure. Summary of the Invention

[0005] The purpose of this disclosure is to provide a service scheduling method, apparatus, electronic device, and computer-readable storage medium that can determine a candidate service scheduling node based on network status information, service type information, and service scheduling request information uploaded by each edge computing node, so as to activate the service scheduling module in the candidate service scheduling node to process service scheduling requests in the cloud-edge computing network and minimize the response time of the cloud-edge computing network to service scheduling requests.

[0006] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0007] This disclosure provides a service scheduling method applied to a control node in a cloud-edge computing network, which also includes multiple edge computing nodes. The method includes: acquiring network status information, service type information, and the number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; determining an activation utility value for each edge computing node based on the uploaded network status information, service type information, and the number of service scheduling requests; determining candidate service scheduling nodes among the edge computing nodes based on the activation utility value; activating the service scheduling module of the candidate service scheduling node, and using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network.

[0008] In some embodiments, activating the service scheduling module of the candidate service scheduling node and using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network includes: activating the service scheduling module of the candidate service scheduling node; notifying multiple edge computing nodes in the cloud-edge computing network that the candidate service scheduling node is the current service scheduling node in the cloud-edge computing network, so that each edge computing node sends a service scheduling request to the current service scheduling node; determining the previous service scheduling node in the cloud-edge computing network; and controlling the previous service scheduling node to close its service scheduling module.

[0009] In some embodiments, the previous service scheduling node includes target scheduling service data. After activating the service scheduling module of the candidate service scheduling node, and before controlling the previous service scheduling node to close its service scheduling module, activating the service scheduling module of the candidate service scheduling node and using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network further includes: controlling the previous service scheduling node to transfer the target scheduling service data to the current service scheduling node, so that the current service scheduling node continues to perform service scheduling processing on the target scheduling service data.

[0010] In some embodiments, determining candidate service scheduling nodes among various edge computing nodes based on the activation utility value includes: determining the edge computing node with the smallest activation utility value among various edge computing nodes as the candidate service scheduling node; and determining the candidate service scheduling node based on the candidate service scheduling node if it is determined that the service scheduling module in the candidate service scheduling node is not activated.

[0011] In some embodiments, if it is determined that the service scheduling module in the candidate service scheduling node is not activated, then the candidate service scheduling node is determined based on the candidate service scheduling node, including: if it is determined that the service scheduling module in the candidate service scheduling node is not activated, then it is further determined whether the target counter in the control node is in an on state; if the target counter is in an on state, then the candidate service scheduling node is determined based on the target counter and the candidate service scheduling node; if the target counter is in a off state, then the candidate service scheduling node is used as the candidate service scheduling node.

[0012] In some embodiments, if the target counter is enabled, determining the candidate service scheduling node based on the target counter and the candidate service scheduling node includes: obtaining the target counter from the management node, the target counter including a target node index and a target count value, the target count value being used to determine the number of times the edge computing node corresponding to the target node index is consecutively selected as a candidate service scheduling node; if the target node index is the index of the candidate service scheduling node and the target count value is a first value, then the candidate service scheduling node is selected as the candidate service scheduling node; if the target node index is the index of the candidate service scheduling node and the target count value is less than the first value, then the candidate service scheduling node is not selected as the candidate service scheduling node, and the target count value is incremented by one.

[0013] In some embodiments, if the target counter is in a closed state, then the candidate service scheduling node is selected as the candidate service scheduling node, including: if it is determined that the target counter is in a closed state, then continue to determine the number of times the cloud-edge computing network switches the current service scheduling node within a target time period; determine that the number of times the cloud-edge computing network switches the current service scheduling node within the target time period is greater than a second value; turn on the target counter, set the target node index of the target counter to the index of the candidate service scheduling node, and set the target count value of the target counter to a third value, so that the cloud-edge computing network determines the candidate service scheduling node based on the target counter.

[0014] In some embodiments, the plurality of edge computing nodes includes a first edge computing node, and the service scheduling module deployed in the first edge computing node is a target service scheduling module. The determination of the activation utility value of each edge computing node based on network status information, service type information, and the number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network includes: determining the target latency for each edge computing node to initiate a service scheduling request to the first edge computing node based on the network status information uploaded by each edge computing node in the cloud-edge computing network; determining the target service module enabled in the first edge computing node based on the server type of the first edge computing node; determining the module compatibility value between the target service module and the target service scheduling module; determining the scheduling service probability of the first edge computing node in making service scheduling requests in the cloud-edge computing network based on the number of service scheduling requests made by the first edge computing node; determining the target priority function value of the first edge computing node based on the module compatibility value and the scheduling service probability; and determining the activation utility value of the first edge computing node based on the target latency and the target priority function value.

[0015] In some embodiments, determining the module compatibility value between the target service module and the target service scheduling module includes: obtaining the target third-party module called by the target service module and the target service scheduling module; identifying a first calling module jointly called by the target service module and the target service scheduling module in the target third-party module; identifying a second calling module with a different calling version by the target service module and the target service scheduling module in the first calling module; and determining the module compatibility value between the target service module and the target service scheduling module based on the second calling module, the first calling module, and the third-party module.

[0016] In some embodiments, determining the module compatibility value between the target service module and the target service scheduling module includes: determining the affinity module and the rejection module of the target service scheduling module in the target service module; determining the affinity score value corresponding to the affinity module and the rejection score value corresponding to the rejection module; and determining the module compatibility value between the target service module and the target service scheduling module in the first edge computing node based on the affinity score value, the number of affinity modules, the rejection score value, and the number of rejection modules.

[0017] In some embodiments, the plurality of edge computing nodes includes a second edge computing node, and the target latency includes a second latency for the second edge computing node to initiate a service scheduling request to the first edge computing node. Determining the target latency for each edge computing node to initiate a service scheduling request to the first edge computing node based on network status information uploaded by each edge computing node in the cloud-edge computing network includes: determining the connectivity and connection speed between each edge computing node in the cloud-edge computing network based on the network status information uploaded by each edge computing node in the cloud-edge computing network; determining the optimal path with the shortest connection time from the second edge computing node to the first edge computing node based on the connectivity and connection speed between each edge computing node; acquiring the service scheduling characteristics of the second edge computing node when sending the service scheduling request, and the data characteristics of the first edge computing node issuing a service scheduling decision to the first edge computing node; and determining the second latency for the second edge computing node to initiate a service scheduling request to the first edge computing node based on the service scheduling characteristics of the second edge computing node when sending the service scheduling request, the data characteristics of the first edge computing node issuing a service scheduling decision to the first edge computing node, and the connection speed between each edge computing node on the optimal path.

[0018] This disclosure provides a service scheduling device applied to a control node in a cloud-edge computing network. The cloud-edge computing network also includes multiple edge computing nodes, including: a data acquisition module, an activation utility value determination module, a candidate service scheduling node determination module, and an activation module.

[0019] The data acquisition module is used to acquire network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network. The activation utility value determination module can be used to determine the activation utility value of each edge computing node based on the network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network. The candidate service scheduling node determination module can be used to determine candidate service scheduling nodes among the edge computing nodes based on the activation utility value. The activation module can be used to activate the service scheduling module of the candidate service scheduling node and use the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network.

[0020] This disclosure provides an electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the service scheduling method described above.

[0021] This disclosure provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the service scheduling method as described in any of the preceding embodiments.

[0022] This disclosure provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned service scheduling method.

[0023] The service scheduling method, apparatus, electronic device, and computer-readable storage medium provided in this disclosure determine the activation utility value of each edge computing node based on network status information, service type information, and service scheduling request information uploaded by each edge computing node in the edge computing network. Based on this activation utility value, a candidate service scheduling node is determined from among the edge computing nodes. The service scheduling module on the candidate service scheduling node is then activated, and this candidate service scheduling module is used as the current service scheduling module to schedule service scheduling requests in the cloud-edge computing network. Through this technical solution, on the one hand, the candidate service scheduling node is determined from multiple edge computing nodes based on the activation utility value. This ensures that the determination of the candidate service scheduling node comprehensively considers the network status information, service type, and number of service scheduling requests of each edge computing node in the cloud-edge computing network, enabling the candidate service scheduling node to complete service scheduling in the cloud-edge computing network as quickly as possible. On the other hand, this technical solution schedules service scheduling requests in the cloud-edge computing network using only one service scheduling node, avoiding multiple tasks being scheduled to the same node and reducing the possibility of node congestion.

[0024] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this disclosure. Attached Figure Description

[0025] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0026] Figure 1 A schematic diagram of an exemplary system architecture that can be applied to the service scheduling method or service scheduling apparatus of the present disclosure embodiments is shown.

[0027] Figure 2 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0028] Figure 3 This is a schematic diagram of a cloud-edge computing network topology according to an exemplary embodiment.

[0029] Figure 4 This is a schematic diagram illustrating a multi-functional service scheduling method according to an exemplary embodiment.

[0030] Figure 5 This is a schematic diagram illustrating a single effective service scheduling according to an exemplary embodiment.

[0031] Figure 6 This is a schematic diagram of inter-node bandwidth according to an exemplary embodiment.

[0032] Figure 7 This is a schematic diagram illustrating an inter-node request transmission rate according to an exemplary embodiment.

[0033] Figure 8 This is an example of a method for determining activation utility value.

[0034] Figure 9 This is a service scheduling module activation method illustrated according to an exemplary embodiment.

[0035] Figure 10 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0036] Figure 11 This is a flowchart illustrating a method for determining candidate service scheduling nodes based on a target counter, according to an exemplary embodiment.

[0037] Figure 12 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0038] Figure 13 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0039] Figure 14 This is a block diagram illustrating a service scheduling apparatus according to an exemplary embodiment.

[0040] Figure 15 A schematic diagram of the structure of an electronic device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0041] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.

[0042] The features, structures, or characteristics described in this disclosure can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more specific details omitted, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0043] The accompanying drawings are merely illustrative of this disclosure, and the same reference numerals in the drawings denote the same or similar parts, thus omitting repeated descriptions of them. Some block diagrams shown in the drawings do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0044] The flowchart shown in the accompanying drawings is merely illustrative and does not necessarily include all content and steps, nor does it require execution in the described order. For example, some steps may be broken down, while others may be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0045] In this specification, the terms “a,” “an,” “the,” “the,” and “at least one” are used to indicate the presence of one or more elements / components / etc.; the terms “comprising,” “including,” and “having” are used to indicate an open-ended inclusion and to mean that there may be other elements / components / etc. in addition to the listed elements / components / etc.; the terms “first,” “second,” and “third,” etc., are used only as markings and are not a limitation on the number of objects.

[0046] The exemplary embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0047] Figure 1 A schematic diagram of an exemplary system architecture that can be applied to the service scheduling method or service scheduling apparatus of the present disclosure embodiments is shown.

[0048] like Figure 1As shown, the system architecture 100 may include: a control node 101, edge computing nodes 102 and 103 (for ease of understanding, this embodiment only uses two edge computing nodes in the cloud-edge computing network as an example, but this disclosure is not limited to this), and a terminal device 104. The control node 101 and the edge computing nodes 102 and 103, or the edge computing nodes 102 and 103 and the terminal device 104, can communicate through a network, which may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0049] In this embodiment of the disclosure, the control node 101 and the edge computing nodes 102 and 103 can constitute a cloud-edge computing network to provide various services to the terminal device 104.

[0050] In this disclosure, the control node 101 may refer to a control node, which may be any cloud such as public cloud, private cloud, central cloud, edge cloud, etc., and this disclosure does not limit it; the control node may also refer to a central data center (for example), and this disclosure does not limit it.

[0051] It is understandable that the devices corresponding to the control node 101 can include any device with computing capabilities, such as servers and terminal devices. The terminal devices can be various electronic devices with a display screen and support web browsing, including but not limited to smartphones, tablets, laptops, desktop computers, wearable devices, virtual reality devices, smart home devices, etc.

[0052] In this disclosure, edge computing nodes 102 and 103 may refer to service nodes controlled by managed node 101 that provide services to terminal device 104 nearby.

[0053] In this disclosure, edge computing nodes 102 and 103 can be an edge cloud, which can include any device with computing capabilities, such as servers and terminal devices. The terminal devices can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, desktop computers, wearable devices, virtual reality devices, smart home devices, etc.

[0054] In this disclosure, the terminal device 104 can be various electronic devices with a display screen and support web browsing, including but not limited to smartphones, tablets, laptops, desktop computers, wearable devices, virtual reality devices, smart home devices, etc., and this disclosure does not limit them.

[0055] In some embodiments, the terminal device (or server) 104 can request service from the edge computing node in the cloud-edge computing network through the network medium. When the device resources of a certain edge computing node in the cloud-edge computing network are limited, resulting in insufficient processing capacity, the edge computing node will offload the computing task to other edge computing nodes in the cloud-edge computing network.

[0056] In some embodiments, the control node 101 may periodically execute the following steps to schedule computing offloading tasks in the cloud-edge computing network. Specifically, this method may include: acquiring network status information, service type information, and the number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; determining the activation utility value of each edge computing node based on the network status information, service type information, and the number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; determining candidate service scheduling nodes among the edge computing nodes based on the activation utility value; activating the service scheduling module of the candidate service scheduling node, using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network, acquiring the previous service scheduling node in the cloud-edge computing network, and stopping the service scheduling request module of the previous service scheduling node.

[0057] It should be noted that the network medium between the terminal device (or server) 101 and each edge computing node, or between the edge computing node and the management node, can include various connection types, such as wired or wireless communication links or fiber optic cables, etc., and this disclosure does not limit them.

[0058] The aforementioned servers can be servers that provide various services, independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. This disclosure does not impose any restrictions on these.

[0059] Edge computing nodes can consist of multiple terminal devices and / or servers, and this disclosure does not limit this. Control nodes can also consist of multiple terminals and / or servers, and this disclosure does not limit this.

[0060] It should be understood that Figure 1 The number of terminal devices, servers, and edge computing nodes shown is merely illustrative; any number of terminal devices, servers, and edge computing nodes can be included depending on actual needs.

[0061] Figure 2 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0062] In some embodiments, a cloud-edge computing network can be a service network built using cloud-edge computing technology, which can provide backend services to users. The cloud-edge computing network may include control nodes (e.g., a central cloud) and edge computing nodes (e.g., edge clouds) located near user terminals. Multiple edge computing nodes in the cloud-edge computing network can further constitute an edge computing network to provide services to user terminals in close proximity.

[0063] The technical solutions provided in the embodiments of this disclosure can be implemented by a control node or by an edge computing node, and this disclosure does not limit them.

[0064] Figure 3 This diagram illustrates a common multi-access edge computing network topology. The multi-access edge computing network includes a control node (not shown in the diagram) and multiple edge computing nodes (e.g., edge computing nodes A, B, C, or D). The control node may contain a network monitoring module that periodically collects network status information. It also contains a dynamic activation module that can activate the service scheduling modules within the edge computing nodes. Edge computing nodes possess certain resources (e.g., computing or storage resources). Some edge computing nodes communicate with each other, while others do not.

[0065] In an edge network composed of several edge computing nodes, due to the limited computing resources and unstable network link quality, a service scheduling module is needed to schedule services within the edge computing network to ensure user service quality and achieve network load balancing. To ensure consistency in the network state read when different computing tasks make computation offloading decisions, this embodiment only allows one edge computing node to activate its service scheduling module to provide service scheduling services for all computing tasks in the entire network. Deploying the service scheduling service to the network node with the optimal location in the network topology minimizes the average latency of other edge computing nodes requesting the service scheduling service, thus reducing the impact of latency caused by requesting the service scheduling service on the deadlines of computing tasks.

[0066] In this embodiment, a 5G-oriented multi-access edge computing network can be considered, consisting of a control node and edge settlement nodes. The control node is responsible for providing deployment strategies for corresponding modules in the edge computing nodes, and several edge computing nodes can complete the deployment of modules according to the deployment strategies of the control node. The network topology of the edge side of this cloud-edge computing network can be as follows: Figure 3 As shown, the connecting lines between edge computing nodes represent communication links. Due to the instability of network link quality, some computing nodes cannot communicate with each other. Considering the strong consistency of network state during service scheduling, we need to deploy the service scheduling module to the optimal computing node in the entire network. Therefore, finding a suitable service scheduling node among the edge computing nodes becomes the core issue.

[0067] The following will be through Figure 4 and Figure 5 Explain the advantages of using a single node to uniformly schedule computation offloading tasks in an edge computing network.

[0068] Figure 4 This is an example of a multi-efficiency mode. A multi-efficiency mode refers to a scenario where, at any given time, service scheduling is provided through service scheduling modules on multiple edge computing nodes within the entire cloud-edge computing network. For example... Figure 4 As shown, each node is an abstraction of an edge computing node. When multiple nodes receive tasks simultaneously, the multi-efficiency mode schedules the tasks separately through different edge computing nodes. However, when scheduling each task separately, the network conditions considered by different edge computing nodes are not consistent, which may lead to multiple tasks being scheduled to the same node, thus causing that node to overload. And as... Figure 5 The single-effective mode shown effectively avoids this problem and can achieve network load balancing.

[0069] The single-effective mode refers to the fact that in the entire cloud-edge computing network, only the service scheduling module on one edge computing node provides service scheduling at any given time. Although other nodes also have service scheduling modules deployed, they do not provide services to the outside world.

[0070] The advantages of the single-effective mode are illustrated as follows: Figure 4 and Figure 5As shown, two scenarios can be considered: a) multiple service scheduling modules provide service scheduling services simultaneously in the edge computing network (multiple effective mode); b) only one service scheduling module provides service scheduling services simultaneously in the network (single effective mode). When multiple nodes in the network have tasks arriving simultaneously, in a), the node will select the nearest service scheduling module to make the service scheduling request. Note that although task 1 and task 2 arrive at the same time, the scheduling decision is made in different service scheduling modules. On the one hand, the different latency of the two task requests being transmitted to different service scheduling modules will lead to differences in the network conditions when making the decision; on the other hand, scheduling the two tasks independently may result in both tasks being scheduled to the same resource-sufficient and idle computing node, which will lead to resource waste and node overload.

[0071] Therefore, the embodiments of this disclosure adopt a single effective mode, that is, at any given time, the cloud-edge computing network provides service scheduling services through the service scheduling module of only one edge computing node.

[0072] Before implementing the embodiments of this disclosure, network pre-configuration and initial module deployment need to be performed in the cloud-edge computing network:

[0073] Network pre-configuration: In a cloud-edge computing network, one management node and multiple edge computing nodes can be deployed. The management node is equipped with a dynamic activation module, which is responsible for the dynamic activation of the service scheduling module and the collection of logs of service scheduling module request data. The edge computing nodes are responsible for deploying the modules and processing related service requests based on the deployment decisions of the management node.

[0074] Initial Module Deployment: In the initial network phase, a service scheduling module is deployed on all edge computing nodes. The management node can determine the optimal node among multiple edge computing nodes using a certain static scheduling strategy (e.g., determining the node with the best computing or storage capacity among multiple edge computing nodes as the optimal node), and then activate the service scheduling module on the optimal node.

[0075] The service scheduling module is a module that provides service scheduling and resource allocation for user services. When a user service arrives at an edge computing node, the edge computing node can send a request to the service scheduling module. The service scheduling module combines the network status and service characteristics in the edge computing network to schedule the user service to the optimal node and allocate the corresponding network resources to provide the user with the best quality of service.

[0076] Strong consistency of network status means that service scheduling must ensure that the network status information obtained by the service scheduling module is consistent, so as to ensure that its service scheduling decision is optimal.

[0077] The technical solution provided in this disclosure can be applied to the management node in a cloud-edge computing network, which can execute the service scheduling method according to a certain period.

[0078] In some embodiments, the execution cycle of the service scheduling method can be determined by the following methods.

[0079] The execution period τ of the above business scheduling method is determined by the actual scenario (for example, it can be 50ms), and its specific determination rule is given by the following formula:

[0080]

[0081] The average arrival rate of tasks can be defined as the average arrival rate of requests sent from each edge computing node to the control node.

[0082] Reference Figure 2 The service scheduling method provided in this disclosure may include the following steps.

[0083] Step S202: Obtain network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network.

[0084] In some embodiments, edge computing nodes in a cloud-edge computing network upload their own network status information and the network status information between themselves and other edge computing nodes to the management and control node in real time.

[0085] In addition, cloud edge computing nodes will also upload the types of services they are enabling to the management and control nodes in real time.

[0086] In some embodiments, the management node will record in real time the number of times the service type node on each edge computing node requests the business scheduling service and write it into the cache.

[0087] In addition, control nodes can also deploy network monitoring modules (such as Prometheus (an open-source system monitoring and alarm module) or Grafana (an automated monitoring tool)) to obtain information on the service activation status, service request count, and network performance data of each edge computing node in the network, thereby integrating network status information.

[0088] Step S204: Determine the activation utility value of each edge computing node based on the network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network.

[0089] In some embodiments, the control node can periodically update the network status between various nodes in the cloud-edge computing network based on network performance data, and send the latest network status to the dynamic activation module.

[0090] Network status is a data structure that describes the resource load of each node and link in the network. It includes the computing and storage resource load of nodes, the quality and connectivity of links, etc.

[0091] In some embodiments, the dynamic activation module can calculate the value of the activation utility function of each edge computing node based on the latest network state of the cloud-edge computing network.

[0092] An activation utility function is a function used to measure the performance of a node's deployment of a service scheduling module. The design of the activation utility function directly determines the performance of dynamic activation. This technical solution assumes that the optimal deployment node should simultaneously have a good priority and minimize the average latency for other nodes to request the service scheduling module.

[0093] Let U(n) be the utility function of the service scheduling module deployed on node n, and its calculation method is as shown in formula (1):

[0094]

[0095] In formula (1), Let N be the set of edge computing nodes in the network. The number of nodes in the array, where i is an integer greater than or equal to 1 and n is an integer greater than or equal to 1.

[0096] G(n) is the priority function deployed by the service scheduling module on node n, and its calculation method is as shown in formula (2).

[0097] G(n)=σ a G A (n)+σ f G F (n), 0≤σ a ,σ f ≤1,σ a +σ f =1 (2)

[0098] In formula (2), G A (n) is the affinity function for deploying the service scheduling module on node n, used to characterize the compatibility between the service scheduling module in the current node n and the service modules corresponding to the service types already enabled on that node. Intuitively, the edge computing node corresponding to the service scheduling module with better compatibility with other service modules should be activated; G Fσ represents the probability that node n requests service scheduling in the cloud-edge computing network. Intuitively, the service scheduling module in the edge computing node with a higher probability of requesting service scheduling should be activated. For example, if an edge computing node frequently requests service scheduling, then using that node as the service scheduling node can reduce the latency of service scheduling request transmission in the cloud-edge computing network, thereby reducing the completion time of the service scheduling request. a With σ f The weights of the two functions, σ by default a =σ f =0.5.

[0099] Optionally, the above affinity priority function G A (n) can also be obtained by: determining the affinity module and the rejection module of the target business scheduling module in the target service module; determining the affinity score value corresponding to the affinity module and the rejection score value corresponding to the rejection module; and determining the module compatibility value between the target service module and the target business scheduling module in the first edge computing node based on the affinity score value, the number of affinity modules, the rejection score value and the number of rejection modules.

[0100] Specifically, it can be obtained through the following formula:

[0101]

[0102] In formula (3), v a A value greater than 0 represents the score (an empirical value) of the affinity module on the node (the module that performs more efficiently when deployed on the same node). i >0 represents the score (an empirical value) of the least efficient module (the one that performs worse when deployed on the same node) on the node, and N(·) is the number of modules. By default, v a =v i =1.

[0103] In this context, the affinity module in node n can refer to a service module that has good compatibility and minimal dependency conflicts with the business scheduling module in node n. For example, service modules that use the same third-party software as the business scheduling module can be considered modules with good compatibility with the business scheduling module, or service modules with sufficient computing resources or high stability can also be considered modules with good compatibility with the business scheduling module.

[0104] Service modules that call different third-party software than the business scheduling module, or even if they call the same third-party software but different versions, may have significant dependency conflicts with the business scheduling module.

[0105] In some embodiments, various factors that may affect compatibility with the service scheduling module can be listed, and different weights can be set for different factors to determine the affinity value of each service module in the node relative to the service scheduling module. Then, based on the compatibility value of each service module relative to the service scheduling module, the affinity module (service module with affinity greater than a certain threshold) and the rejection module (service module with affinity lower than a certain threshold) of the service scheduling module can be determined.

[0106] It should be noted that the affinity module and the rejection module mentioned in this application must be service modules corresponding to services that have been enabled on node n.

[0107] In some embodiments, the control center may pre-store affinity values ​​between the service scheduling module and each service module.

[0108] Optionally, the probability priority function G in formula (2) above... F (n) can be approximated by a frequency function, which is the ratio of the number of times node n requests service scheduling to the total number of service scheduling requests in the cloud-edge computing network.

[0109] τ in formula (1) i The delay for node i to initiate a request to the service scheduling module of node n is calculated as shown in formula (4).

[0110]

[0111] In formula (3), The size (in bits) of the service feature data uploaded by node i to the service scheduling module of node n. The size (in bits) of the service scheduling decision data fed back from the service scheduling module of node n to node i. Let node i and node k be s The transmission rate (in Mbps), The set of intermediate nodes marked is the optimal path from node i to node n, i.e., the path from node i to node n is i→k. s →k next →…→k t →n,k s It is an integer greater than or equal to 1.

[0112] In some embodiments, the plurality of edge computing nodes includes a second edge computing node, and the target latency includes a second latency for the second edge computing node to initiate a service scheduling request to the first edge computing node. The second latency can then be determined by the following methods: determining the connectivity and connection speed between edge computing nodes in the cloud-edge computing network based on network status information uploaded by each edge computing node; determining the optimal path with the shortest connection time from the second edge computing node to the first edge computing node based on the connectivity and connection speed between each edge computing node; obtaining the service scheduling characteristics of the second edge computing node when sending the service scheduling request, and the data characteristics of the first edge computing node issuing a service scheduling decision to the first edge computing node; and determining the second latency for the second edge computing node to initiate a service scheduling request to the first edge computing node based on the service scheduling characteristics of the second edge computing node when sending the service scheduling request, the data characteristics of the first edge computing node issuing a service scheduling decision to the first edge computing node, and the connection speed between each edge computing node on the optimal path.

[0113] Specifically, the optimal path can be found using graph theory algorithms, where the weight of each edge represents the time delay. For specific solution strategies, please refer to [link / reference needed]. Figure 6 and Figure 7 And related explanations.

[0114] Figure 6 and Figure 7 This is an example of finding the optimal path. Figure 6 This is a schematic diagram illustrating the network bandwidth between various edge computing nodes according to an exemplary embodiment. Based on the network bandwidth diagram of the various edge computing nodes, the following can be obtained: Figure 7 The diagram shows the network transmission rate between the edge computing nodes. After obtaining the latency of transmitting request data on each edge (assuming the size of the request data is a unit value of 1), the network can be regarded as a weighted graph. Then, the Dijkstra algorithm (an optimal path solving algorithm) can be used to find the shortest latency and path from the request node to the node where the service scheduling module is deployed.

[0115] Step S206: Determine candidate service scheduling nodes among each edge computing node based on the activation utility value.

[0116] In some embodiments, edge computing nodes can be sorted according to their activation utility function values, and then candidate service nodes can be determined based on the sorting results of the activation utility function values. For example, the edge computing node with the largest activation utility function value can be selected as a candidate service node.

[0117] In addition, candidate service nodes can be identified in each edge computing node through the following dynamic activation strategy.

[0118] The dynamic activation strategy aims to minimize the utility function, and its calculation method is shown in formula (5):

[0119]

[0120] The constraint conditions for the dynamic activation strategy are calculated as shown in formula (6):

[0121]

[0122] In the formula, t0 is the current time, f0,r0 are the computing and storage resources required to activate the dynamic activation module, and F n (t0),R n (t0) represents the remaining computing and storage resources on node n at the current time. * (t0) The node currently having the dynamic activation module enabled. Constraints C1 and C2 require that the node having the dynamic activation module enabled should be able to allocate the computing and storage resources required to enable the module; constraint C3 requires that the new node and the old node cannot be the same node.

[0123] Step S208: Activate the service scheduling module of the candidate service scheduling node, and use the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network.

[0124] In some embodiments, when a candidate service scheduling node is used as the current service scheduling node to schedule services for service scheduling requests in the cloud-edge computing network, it is also necessary to shut down the service scheduling module of the old service scheduling node to avoid service scheduling under multiple effective modes.

[0125] The existing strategy related to module deployment is a static scheduling strategy. This means that when a module is started, the control node deploys it to the optimal node based on the instantaneous network state and continues to execute it. Clearly, this deployment strategy does not consider subsequent changes in network state or the correlation between module deployment and network status. It can only schedule based on the instantaneous network state during deployment and cannot proactively update when the network topology changes. When significant changes occur in the network state later, the link quality of the module deployment node deteriorates, and other nodes requesting the module incur significant latency, drastically reducing network resource utilization and impacting user experience.

[0126] The technical solution provided in this embodiment periodically monitors the network status in the cloud-edge computing network and periodically determines whether the current service scheduling node needs to be updated, so that the service scheduling node in the cloud-edge computing network can dynamically change with the changes in network status and service scheduling request status, thereby reducing the latency of service scheduling requests in the cloud-edge computing network.

[0127] The technical solution provided in this embodiment addresses the need for multi-access edge computing networks to minimize the impact of latency caused by the request service scheduling module on user service deadlines. Traditional static scheduling strategies cannot simultaneously detect user services arriving at the same time, and cannot achieve adaptive adjustment of service scheduling module nodes when network topology changes, thus affecting user service quality.

[0128] This application proposes a single-effective mode for service scheduling modules that guarantees strong consistency of network state. Based on the static scheduling strategy, it designs a dynamic activation strategy for service scheduling modules based on graph theory algorithms by comprehensively considering the latency of service requests, the affinity of network nodes, and the probability of requests. It performs secondary encapsulation without changing the logic of the existing scheduling design paradigm, thereby combining static and dynamic scheduling.

[0129] In addition, the technical solution provided in this embodiment, when determining a new service scheduling node, takes into account the compatibility of the service scheduling module in the service scheduling node with other service modules and the probability of the service scheduling node requesting service scheduling. It comprehensively considers multiple aspects to reduce the latency of each edge computing node in the edge computing network sending service scheduling requests to the service scheduling node.

[0130] Figure 8 This is an example of a method for determining activation utility value.

[0131] In some embodiments, the plurality of edge computing nodes includes a first edge computing node, wherein the service scheduling module deployed in the first edge computing node is a target service scheduling module.

[0132] refer to Figure 8 The above-mentioned method for determining activation utility value may include the following steps.

[0133] Step S802: Determine the target latency for each edge computing node to initiate a service scheduling request to the first edge computing node based on the network status information uploaded by each edge computing node in the cloud-edge computing network.

[0134] In some embodiments, the target latency for each edge computing node to initiate a service scheduling request to the first edge computing node can be determined according to formula (4).

[0135] Step S804: Determine the target service module enabled in the first edge computing node based on the server type of the first edge computing node.

[0136] Step S806: Determine the module compatibility value between the target service module and the target business scheduling module.

[0137] In some embodiments, the compatibility value with the target service module can be determined based on the affinity and dependency conflict between the target service module and the target business scheduling module.

[0138] For example, service modules that are the same as the third-party software called by the business scheduling module can be modules with good compatibility with the business scheduling module, or service modules with sufficient computing resources or high stability can be modules with good compatibility with the business scheduling module.

[0139] Service modules that call different third-party software than the business scheduling module, or even if they call the same third-party software but different versions, may have significant dependency conflicts with the business scheduling module.

[0140] For example, the compatibility between the target service module and the target service scheduling module can be determined by the following method: obtaining the target third-party module called by the target service module and the target service scheduling module; identifying a first calling module jointly called by the target service module and the target service scheduling module in the target third-party module; identifying a second calling module with a different calling version by the target service module and the target service scheduling module in the first calling module; and determining the module compatibility value between the target service module and the target service scheduling module based on the second calling module, the first calling module, and the third-party module.

[0141] In some embodiments, various factors that may affect compatibility with the service scheduling module can be listed, and different weights can be set for different factors to determine the affinity value of each service module in the node relative to the service scheduling module. Then, based on the compatibility value of each service module relative to the service scheduling module, the affinity module (service module with affinity greater than a certain threshold) and the rejection module (service module with affinity lower than a certain threshold) of the service scheduling module can be determined.

[0142] Step S608: Determine the scheduling service probability of the first edge computing node in the cloud-edge computing network based on the number of service scheduling requests made by the first edge computing node.

[0143] In some embodiments, the scheduling service probability G of the first edge computing node in the cloud-edge computing network for making service scheduling requests can be approximated using a frequency function.F (n).

[0144] The ratio of the number of times the first edge computing node n requests service scheduling to the total number of service scheduling requests in the cloud-edge computing network is taken as the scheduling service probability corresponding to the first edge computing node.

[0145] Step S610: Determine the target priority function value of the first edge computing node based on the module compatibility value and the scheduling service probability.

[0146] In some embodiments, the target priority function value G(n) of the first edge computing node n can be determined according to formula (2).

[0147] Step S612: Determine the activation utility value of the first edge computing node based on the target latency and the target priority function value.

[0148] In some embodiments, the activation utility value of the first edge computing node can be determined according to formula (1).

[0149] The technical solution provided in the above embodiments, when determining the activation utility value of the first edge computing node, comprehensively considers the compatibility between the service modules and the business scheduling module of each enabled service in the first edge computing node, as well as the probability of the first edge computing node requesting the business scheduling service. This makes it possible to select edge computing nodes with high compatibility between the business scheduling module and other server modules and high probability of business scheduling requests when determining the business scheduling node based on the activation utility value, so as to reduce the latency of requests issued by each edge computing node reaching the business scheduling node.

[0150] Figure 9 This is a service scheduling module activation method illustrated according to an exemplary embodiment.

[0151] refer to Figure 9 The above-mentioned service scheduling module activation method may include the following steps.

[0152] Step S902: Activate the service scheduling module of the candidate service scheduling node.

[0153] Step S904: Notify the multiple edge computing nodes in the cloud-edge computing network that the candidate service scheduling node is the current service scheduling node in the cloud-edge computing network, so that each edge computing node can send a service scheduling request to the current service scheduling node.

[0154] Step S906: Determine the previous service scheduling node in the cloud-edge computing network.

[0155] Step S908: Control the previous service scheduling node to shut down the service scheduling module.

[0156] In some embodiments, the previous service scheduling node includes target scheduling service data. After activating the service scheduling module of the candidate service scheduling node, and before controlling the previous service scheduling node to close its service scheduling module, activating the service scheduling module of the candidate service scheduling node and using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network further includes: controlling the previous service scheduling node to transfer the target scheduling service data to the current service scheduling node, so that the current service scheduling node continues to perform service scheduling processing on the target scheduling service data.

[0157] The technical solution provided in this embodiment, on the one hand, activates a new service scheduling node while not closing the existing service scheduling node, ensuring that at least one service scheduling node exists in the cloud-edge computing network to schedule the service scheduling service for the service scheduling request; on the other hand, after the new service scheduling node (i.e., the current service scheduling node) is activated, the old service scheduling node (i.e., the previous service scheduling node) will transfer the target scheduling service data in the previous service scheduling request to the new service scheduling node, avoiding the existence of multiple service scheduling nodes in the cloud-edge computing network at the same time, and reducing the possibility of node congestion problems in edge computing nodes.

[0158] Figure 10 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0159] refer to Figure 10 The above-mentioned business scheduling method may include the following steps.

[0160] Step S1002: Determine the edge computing node with the smallest activation utility value among all edge computing nodes as the candidate service scheduling node.

[0161] In some embodiments, if it is determined that the service scheduling module in the candidate service scheduling node is not activated, the candidate service scheduling node is determined based on the candidate service scheduling node, which may specifically include steps S1004 to S1008.

[0162] Step S1004: If it is determined that the service scheduling module in the candidate service scheduling node is not activated, then continue to determine whether the target counter in the control node is in the on state.

[0163] Step S1006: If the target counter is in the enabled state, the candidate service scheduling node is determined based on the target counter and the candidate service scheduling node.

[0164] Figure 11This is a flowchart illustrating a method for determining candidate service scheduling nodes based on a target counter, according to an exemplary embodiment.

[0165] refer to Figure 11 The method for determining candidate service scheduling nodes based on the target counter may include the following steps.

[0166] Step S1102: Obtain the target counter from the control node. The target counter includes a target node index and a target count value. The target count value is used to determine the number of times the edge computing node corresponding to the target node index has been continuously selected as a service scheduling node.

[0167] The counter is a tuple that counts the number of times a single node has been the optimal node consecutively. It consists of two elements: (node ​​index) and (number of consecutive times the node has been the optimal node). If the latest optimal node (i.e., the candidate service scheduling node) is not a node in the counter, the data in the counter will be cleared.

[0168] Step S1104: If the target node index is the index of the candidate service scheduling node and the target count value is the first value, then the candidate service scheduling node is selected as the candidate service scheduling node.

[0169] Step S1106: If the target node index is the index of the candidate service scheduling node, and the target count value is less than the first value, then the candidate service scheduling node is not selected as the candidate service scheduling node, and the target count value is incremented by one.

[0170] Step S1008: If the target counter is in a closed state, then the candidate service scheduling node is selected as the candidate service scheduling node.

[0171] Figure 12 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0172] refer to Figure 12 If the target counter is in a closed state, then selecting the candidate service scheduling node as the candidate service scheduling node may include the following steps:

[0173] Step S1202: If it is determined that the target counter is in a closed state, then continue to determine the number of times the cloud-edge computing network switches the current service scheduling node within the target time period.

[0174] Step S1204: Determine that the number of times the cloud-edge computing network switches the current service scheduling node within the target time period is greater than the second value.

[0175] Step S1206: Activate the target counter, set the target node index of the target counter to the index of the candidate service scheduling node, and set the target count value of the target counter to the third value, so that the cloud-edge computing network can determine the candidate service scheduling node based on the target counter.

[0176] The technical solution provided in the above embodiments can reduce the resource waste caused by switching nodes with the service scheduling module enabled when the cloud-edge computing network is unstable and the current service scheduling node is frequently switched, by using a target counter.

[0177] To better understand the above-mentioned objectives, features and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0178] Figure 13 This is a flowchart illustrating a service scheduling method according to an exemplary embodiment.

[0179] refer to Figure 13 The above-mentioned business scheduling method may include the following steps.

[0180] Step S1301, Network Pre-configuration. In the edge network, there is one control node and multiple edge computing nodes. The control node deploys a dynamic activation module, which is responsible for the dynamic activation of the service scheduling module and the collection of logs of service scheduling module request data. The edge computing nodes are responsible for deploying the modules according to the deployment decisions of the control node and processing related service requests.

[0181] Step S1302, Initial Module Deployment. In the initial network phase, all edge computing nodes deploy the service scheduling module. The management node uses the existing static scheduling strategy to activate the service scheduling module on the optimal node. Note that, considering the strong consistency of network status, the activation of the service scheduling module should follow a single-validity mode.

[0182] Step S1303, Reference Data Acquisition. The management node records in real time the types of services started on the node and the number of times the node requests the business scheduling service, and writes them to the cache.

[0183] Step S130, Network Status Update. The control node updates the network status periodically based on network performance data and sends the latest network status to the dynamic activation module.

[0184] Step S1305: Update the activation utility function. The dynamic activation module calculates the activation utility function based on the latest network status according to the time period and sorts them according to the function values.

[0185] Step S1306: Determine whether the current optimal node (i.e., the candidate service scheduling node) is the node with the service scheduling module currently enabled. If the node with the highest value is the node with the service scheduling module currently enabled, no action is taken, and network status continues to be monitored. If the node with the highest value is not the node with the service scheduling module currently enabled, proceed to step S1307 to determine whether the counter is enabled.

[0186] If the counter is not enabled, proceed to step S1308 and assign the counter value to (current optimal node index, 1).

[0187] If the counter is already enabled, proceed to step S1309 to determine whether the node in the counter is the current optimal node. If the node with the highest value is not the node with the currently enabled service scheduling module, and the node index in the counter is not the index of the current optimal node, then reset the counter and assign it the value (optimal node index, 1).

[0188] If the node with the highest value is the node that currently has the service scheduling module enabled, then determine whether the value in the counter is greater than N (e.g., 9), where N is a preset integer greater than 0.

[0189] If the value in the counter is less than N, then execute step S1311 to increment the value in the counter by 1; if the value in the counter is greater than or equal to N, then execute step S1312 to dynamically activate the service scheduling module in the optimal node.

[0190] Step S1312 may include steps S1313 to S1316.

[0191] If the node with the highest value is not the node currently having its service scheduling module enabled, and the node index in the counter is the index of the current optimal node, and the counter value is 9, then network status updates and utility function updates are paused, and preparation is made to activate the service scheduling module of the current optimal node.

[0192] Step S1313, Dynamic Activation Preparation. The dynamic activation module of the control node sends a confirmation message to the new node (the candidate service scheduling node preparing to activate the service scheduling module), preparing to activate the dynamic activation module on the new node. The new node allocates resources for the activation process of the service scheduling module. The control node sends a confirmation message to the old node (the node currently activating the service scheduling module), preparing to shut down the dynamic activation module on the old node. The old node collects the traffic it is currently serving. After receiving the confirmation message, the new node and the old node send a successful activation message to the dynamic activation module and begin the dynamic activation process.

[0193] Step S1314: The new node starts the service scheduling module, but does not provide services at first. After the module starts successfully, it sends a module start success message to the management node.

[0194] In step S1315, after receiving the module activation information, the control node sends traffic transfer information to the old node. Upon receiving the traffic transfer information, the old node transfers the traffic to the new node.

[0195] In step S1316, after the transmission is complete, both the old node and the new node send transmission completion information to the control node, and the new node's service scheduling module begins providing services. Upon receiving both transmission completion messages, the control node notifies the old node to shut down its service scheduling module. Upon receiving the shutdown command, the old node shuts down its service scheduling module.

[0196] In some embodiments, after the dynamic activation process ends, the counter is cleared, the dynamic activation module restarts network state updates and utility function updates, and the dynamic activation module continues to perform the next round of listening.

[0197] This disclosure is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims.

[0198] Therefore, the technical solution provided by this disclosure has the following beneficial effects: designing a single effective mode for the service scheduling module to ensure strong consistency of network state during service scheduling; using graph theory algorithms to find the optimal path for computing nodes to request the service scheduling module and to find the optimal node to activate the service scheduling module; and designing a dynamic activation strategy for the service scheduling module to reduce the impact on network stability during dynamic activation.

[0199] This embodiment proposes a dynamic activation method for service scheduling modules in 5G multi-access edge computing networks. This method designs a single effective mode for service scheduling modules that ensures strong consistency of network state. Based on the static scheduling strategy, it designs a dynamic activation strategy for service scheduling modules based on graph theory algorithms by comprehensively considering the latency of service requests, the affinity of network nodes, and the request frequency. It performs secondary encapsulation without changing the logic of the existing scheduling design paradigm, thereby combining static and dynamic scheduling.

[0200] Furthermore, this embodiment designs a single-active mode for the module. This effectively avoids the overload problem caused by tasks arriving from different nodes being offloaded to the same idle node when multiple tasks arrive at multiple nodes simultaneously in the entire network. Through this updated design pattern, only one active service scheduling module is allowed in the entire network, responsible for service scheduling across all nodes, ensuring strong consistency of network status. Since the amount of data transmitted for calculating offloading decisions is not large, it does not generate excessive latency and has no significant impact on service execution.

[0201] Furthermore, this embodiment proposes an activation utility function for dynamic module activation. It aims to minimize the average latency of network request service scheduling and maximize the node priority function, ensuring that the impact of latency caused by request computation offloading services on the deadline of computation tasks is minimized. Simultaneously, it considers the scenario where nodes in the network are not interconnected, and uses graph theory algorithms to find the optimal node request service path for solving the utility function, demonstrating excellent practical value.

[0202] Finally, the dynamic activation method proposed in this implementation always deploys the compute offloading service to the optimal network node in the entire network topology, minimizing the average latency for other network nodes requesting the compute offloading service. Without altering the existing scheduling logic, it adaptively adjusts the service scheduling module to the optimal service node based on network conditions, encapsulating the existing scheduling design to combine static and dynamic scheduling. By setting a counter function, it also avoids frequent module switching, ensuring that network resources are not misused or wasted.

[0203] Figure 14 This is a block diagram illustrating a service scheduling device according to an exemplary embodiment. It is applied to a control node in a cloud-edge computing network, which also includes multiple edge computing nodes.

[0204] Reference Figure 14 The service scheduling device 1400 provided in this embodiment may include: a data acquisition module 1401, an activation utility value determination module 1402, a candidate service scheduling node determination module 1403, and an activation module 1404.

[0205] The data acquisition module 1401 can be used to acquire network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; the activation utility value determination module 1402 can be used to determine the activation utility value of each edge computing node based on the network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; the candidate service scheduling node determination module 1403 can be used to determine candidate service scheduling nodes among the edge computing nodes based on the activation utility value; and the activation module 1404 can be used to activate the service scheduling module of the candidate service scheduling node, and use the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network.

[0206] In some embodiments, the activation module 1404 may include: an activation submodule, a notification submodule, a previous service scheduling node determination submodule, and a control shutdown module.

[0207] The activation submodule can be used to activate the service scheduling module of the candidate service scheduling node; the notification submodule can be used to notify multiple edge computing nodes in the cloud-edge computing network that the candidate service scheduling node is the current service scheduling node in the cloud-edge computing network, so that each edge computing node can send a service scheduling request to the current service scheduling node; the previous service scheduling node determination submodule can be used to determine the previous service scheduling node in the cloud-edge computing network; and the control shutdown module can be used to control the previous service scheduling node to shut down its service scheduling module.

[0208] In some embodiments, the previous service scheduling node includes target scheduling service data, and the activation module 1404 may further include a data transfer submodule.

[0209] The data transfer submodule can, after activating the service scheduling module of the candidate service scheduling node and before controlling the previous service scheduling node to close its service scheduling module, control the previous service scheduling node to transfer the target scheduling service data to the current service scheduling node, so that the current service scheduling node can continue to perform service scheduling processing on the target scheduling service data.

[0210] In some embodiments, the candidate service scheduling node determination module 1403 may include: a candidate service scheduling node determination submodule and an inactive determination submodule.

[0211] The candidate service scheduling node determination submodule can be used to determine the edge computing node with the smallest activation utility value among all edge computing nodes as the candidate service scheduling node; the non-activated determination submodule can be used to determine the candidate service scheduling node based on the candidate service scheduling node if the service scheduling module among the candidate service scheduling nodes is not activated.

[0212] In some embodiments, the inactive determination submodule may include: an on-state determination unit, a first candidate service scheduling node determination unit, and a second candidate service scheduling node determination unit.

[0213] The activation state determination unit can be used to determine whether the target counter in the control node is in the activated state if the service scheduling module in the candidate service scheduling node is not activated. The first candidate service scheduling node determination unit can be used to determine the candidate service scheduling node based on the target counter and the candidate service scheduling node if the target counter is in the activated state. The second candidate service scheduling node determination unit can be used to select the candidate service scheduling node as the candidate service scheduling node if the target counter is in the deactivated state.

[0214] In some embodiments, the candidate service scheduling node determination first unit may include: a target counter acquisition subunit, an index determination subunit, and a counter value increment unit.

[0215] The target counter acquisition subunit can be used to acquire the target counter from the control node. The target counter includes a target node index and a target count value. The target count value is used to determine the number of times the edge computing node corresponding to the target node index has been consecutively selected as a candidate service scheduling node. The index determination subunit can be used to select the candidate service scheduling node as a candidate service scheduling node if the target node index is the index of the candidate service scheduling node and the target count value is a first value. The count value increment unit can be used to not select the candidate service scheduling node as a candidate service scheduling node and increment the target count value if the target node index is the index of the candidate service scheduling node and the target count value is less than the first value.

[0216] In some embodiments, the candidate service scheduling node determining second unit may include: a shutdown status determining subunit, a switching count determining subunit, and a target counter enabling subunit.

[0217] The "Closed State Determination Subunit" can be used to determine the number of times the cloud-edge computing network switches the current service scheduling node within the target time period if the target counter is determined to be in a closed state. The "Switching Count Determination Subunit" can be used to determine that the number of times the cloud-edge computing network switches the current service scheduling node within the target time period is greater than a second value. The "Target Counter Enabling Subunit" can be used to enable the target counter, set the target node index of the target counter to the index of the candidate service scheduling node, and set the target count value of the target counter to a third value, so that the cloud-edge computing network can determine the candidate service scheduling node based on the target counter.

[0218] In some embodiments, the plurality of edge computing nodes include a first edge computing node, and the service scheduling module deployed in the first edge computing node is a target service scheduling module; wherein, the activation utility value determination module 1402 may include: a target latency determination submodule, a target service module determination submodule, a module compatibility value determination submodule, a scheduling service probability determination submodule, a target priority function value determination submodule, and an activation utility value determination submodule.

[0219] The target latency determination submodule can be used to determine the target latency for each edge computing node to initiate a service scheduling request to the first edge computing node based on the network status information uploaded by each edge computing node in the cloud-edge computing network; the target service module determination submodule can be used to determine the target service module enabled in the first edge computing node based on the server type of the first edge computing node; the module compatibility value determination submodule can be used to determine the module compatibility value between the target service module and the target service scheduling module; the scheduling service probability determination submodule can be used to determine the scheduling service probability of the first edge computing node in making service scheduling requests in the cloud-edge computing network based on the number of service scheduling requests made by the first edge computing node; the target priority function value determination submodule can be used to determine the target priority function value of the first edge computing node based on the module compatibility value and the scheduling service probability; and the activation utility value determination submodule can be used to determine the activation utility value of the first edge computing node based on the target latency and the target priority function value.

[0220] In some embodiments, the module compatibility value determination submodule may include: a target third-party module determination unit, a first calling module determination submodule, a second calling module determination submodule, and a module compatibility value determination submodule.

[0221] The target third-party module determination unit can be used to obtain the target service module and the target business scheduling module calling the target third-party module; the first calling module determination submodule can be used to determine the first calling module jointly called by the target service module and the target business scheduling module in the target third-party module; the second calling module determination submodule can be used to determine the second calling module with a different version called by the target service module and the target business scheduling module in the first calling module; the module compatibility value determination submodule can be used to determine the module compatibility value between the target service module and the target business scheduling module based on the second calling module, the first calling module, and the third-party module.

[0222] In some embodiments, the first calling module and the third-party module include: an affinity module determination submodule, an affinity score determination submodule, and a module compatibility value determination submodule.

[0223] The affinity module determination submodule can be used to determine the affinity module and rejection module of the target service scheduling module in the target service module; the affinity score determination submodule can be used to determine the affinity score value corresponding to the affinity module and the rejection score value corresponding to the rejection module; the module compatibility value determination submodule can be used to determine the module compatibility value between the target service module and the target service scheduling module in the first edge computing node based on the affinity score value, the number of affinity modules, the rejection score value, and the number of rejection modules.

[0224] In some embodiments, the plurality of edge computing nodes includes a second edge computing node, and the target latency includes a second latency for the second edge computing node to initiate a service scheduling request to the first edge computing node; wherein, the target latency determination submodule may include: a connectivity determination unit, an optimal path determination unit, a data feature acquisition unit, and a second latency determination unit.

[0225] The connectivity determination unit can be used to determine the connectivity and connection speed between edge computing nodes in the cloud-edge computing network based on the network status information uploaded by each edge computing node in the cloud-edge computing network; the optimal path determination unit can be used to determine the optimal path with the shortest connection time from the second edge computing node to the first edge computing node based on the connectivity and connection speed between each edge computing node; the data feature acquisition unit can be used to acquire the service scheduling features of the second edge computing node when sending a service scheduling request, and the data features of the first edge computing node issuing a service scheduling decision to the first edge computing node; the second latency determination unit can be used to determine the second latency of the second edge computing node initiating a service scheduling request to the first edge computing node based on the service scheduling features of the second edge computing node when sending a service scheduling request, the data features of the first edge computing node issuing a service scheduling decision to the first edge computing node, and the connection speed between each edge computing node on the optimal path.

[0226] Since the functions of the device 1400 have been described in detail in their respective method embodiments, they will not be repeated here.

[0227] The modules and / or sub-modules and / or units and / or sub-units described in the embodiments of this application can be implemented in software or hardware. The described modules and / or sub-modules and / or units and / or sub-units can also be located in a processor. The names of these modules and / or sub-modules and / or units and / or sub-units do not, in some cases, constitute a limitation on the module and / or sub-module and / or unit and / or sub-unit itself.

[0228] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0229] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0230] Figure 15 A schematic diagram of the structure of an electronic device suitable for implementing embodiments of the present disclosure is shown. It should be noted that... Figure 15 The illustrated electronic device 1500 is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0231] like Figure 15As shown, the electronic device 1500 includes a central processing unit (CPU) 1501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1502 or a program loaded from a storage section 1508 into a random access memory (RAM) 1503. The RAM 1503 also stores various programs and data required for the operation of the electronic device 1500. The CPU 1501, ROM 1502, and RAM 1503 are interconnected via a bus 1504. An input / output (I / O) interface 1505 is also connected to the bus 1504.

[0232] The following components are connected to I / O interface 1505: an input section 1506 including a keyboard, mouse, etc.; an output section 1507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1508 including a hard disk, etc.; and a communication section 1509 including a network interface card such as a LAN card, modem, etc. The communication section 1509 performs communication processing via a network such as the Internet. A drive 1510 is also connected to I / O interface 1505 as needed. Removable media 1511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 1510 as needed so that computer programs read from them can be installed into storage section 1508 as needed.

[0233] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 1509, and / or installed from removable medium 1511. When the computer program is executed by central processing unit (CPU) 1501, it performs the functions defined above in the system of this application.

[0234] It should be noted that the computer-readable storage medium disclosed herein may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0235] In another aspect, this application also provides a computer-readable storage medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable storage medium carries one or more programs, which, when executed by the device, enable the device to perform the following functions: acquiring network status information, service type information, and the number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; determining the activation utility value of each edge computing node based on the network status information, service type information, and the number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; determining candidate service scheduling nodes among the edge computing nodes based on the activation utility value; activating the service scheduling module of the candidate service scheduling node, and using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network.

[0236] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in various optional implementations of the above embodiments.

[0237] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions of the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computing device (such as a personal computer, server, mobile terminal, or smart device, etc.) to execute the method according to the embodiments of this disclosure, for example... Figure 2 , Figure 6 , Figure 9 , Figure 10 , Figure 11 , Figure 12 ,or Figure 13 One or more of the steps shown.

[0238] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not claimed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0239] It should be understood that this disclosure is not limited to the detailed structures, drawing arrangements or implementations shown herein; rather, this disclosure is intended to cover various modifications and equivalent arrangements contained within the spirit and scope of the appended claims.

Claims

1. A service scheduling method, characterized in that, A management and control node applied in a cloud-edge computing network, wherein the cloud-edge computing network also includes multiple edge computing nodes, the method comprising: Obtain network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network; The activation utility value of each edge computing node is determined based on the network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network. Candidate service scheduling nodes are determined among each edge computing node based on the activation utility value; The service scheduling module of the candidate service scheduling node is activated, and the candidate service scheduling node is used as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network; The plurality of edge computing nodes include a first edge computing node, wherein the service scheduling module deployed in the first edge computing node is a target service scheduling module; wherein the activation utility value of each edge computing node is determined based on the network status information, service type information, and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network, including: The target latency for each edge computing node to initiate a service scheduling request to the first edge computing node is determined based on the network status information uploaded by each edge computing node in the cloud-edge computing network. The target service module enabled in the first edge computing node is determined based on the server type of the first edge computing node; Determine the module compatibility value between the target service module and the target business scheduling module; The probability of the first edge computing node performing a service scheduling request in the cloud-edge computing network is determined based on the number of service scheduling requests made by the first edge computing node. The target priority function value of the first edge computing node is determined based on the module compatibility value and the scheduling service probability. The activation utility value of the first edge computing node is determined based on the target latency and the target priority function value.

2. The method according to claim 1, characterized in that, Activating the service scheduling module of the candidate service scheduling node, and using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network, including: Activate the service scheduling module of the candidate service scheduling node; The candidate service scheduling node is notified to multiple edge computing nodes in the cloud-edge computing network that it is the current service scheduling node in the cloud-edge computing network, so that each edge computing node can send a service scheduling request to the current service scheduling node. Determine the previous service scheduling node in the cloud-edge computing network; Control the previous service scheduling node to shut down the service scheduling module.

3. The method according to claim 2, characterized in that, The previous service scheduling node includes target scheduling service data. After activating the service scheduling module of the candidate service scheduling node and before controlling the previous service scheduling node to close its service scheduling module, activating the service scheduling module of the candidate service scheduling node and using the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network further includes: The previous service scheduling node is controlled to transfer the target scheduling service data to the current service scheduling node, so that the current service scheduling node can continue to perform service scheduling processing on the target scheduling service data.

4. The method according to claim 1, characterized in that, Based on the activation utility value, candidate service scheduling nodes are determined among each edge computing node, including: Among all edge computing nodes, the edge computing node with the smallest activation utility value is selected as the candidate service scheduling node; If it is determined that the service scheduling module in the candidate service scheduling node is not activated, then the candidate service scheduling node is determined based on the candidate service scheduling node.

5. The method according to claim 4, characterized in that, If it is determined that the service scheduling module among the candidate service scheduling nodes is not activated, then the candidate service scheduling nodes are determined based on the candidate service scheduling nodes, including: If it is determined that the service scheduling module in the candidate service scheduling node is not activated, then it is further determined whether the target counter in the control node is in the on state. If the target counter is enabled, the candidate service scheduling node is determined based on the target counter and the candidate service scheduling node. If the target counter is in a closed state, then the candidate service scheduling node will be used as the candidate service scheduling node.

6. The method according to claim 5, characterized in that, If the target counter is enabled, the candidate service scheduling node is determined based on the target counter and the candidate service scheduling node, including: The target counter is obtained from the control node. The target counter includes a target node index and a target count value. The target count value is used to determine the number of times the edge computing node corresponding to the target node index has been continuously selected as a service scheduling node. If the target node index is the index of the candidate service scheduling node, and the target count value is the first value, then the candidate service scheduling node is selected as the candidate service scheduling node. If the target node index is the index of the candidate service scheduling node, and the target count value is less than the first value, then the candidate service scheduling node will not be used as the candidate service scheduling node, and the target count value will be incremented by one.

7. The method according to claim 5, characterized in that, If the target counter is in a closed state, then the candidate service scheduling node is selected as the candidate service scheduling node, including: If it is determined that the target counter is in a closed state, then the number of times the cloud-edge computing network switches the current service scheduling node within the target time period is determined. It is determined that the number of times the cloud-edge computing network switches the current service scheduling node within the target time period is greater than the second value; The target counter is activated, and the target node index of the target counter is set to the index of the candidate service scheduling node. The target count value of the target counter is set to the third value, so that the cloud-edge computing network can determine the candidate service scheduling node based on the target counter.

8. The method according to claim 1, characterized in that, Determining the module compatibility value between the target service module and the target service scheduling module includes: The target service module and the target business scheduling module call the target third-party module; In the target third-party module, determine the first calling module jointly invoked by the target service module and the target business scheduling module; In the first calling module, a second calling module with a different calling version from the target service module and the target business scheduling module is determined; The module compatibility value between the target service module and the target business scheduling module is determined based on the second calling module, the first calling module, and the third-party module.

9. The method according to claim 1, characterized in that, Determining the module compatibility value between the target service module and the target service scheduling module includes: In the target service module, the affinity module and the rejection module of the target service scheduling module are determined; Determine the affinity score value corresponding to the affinity module and the rejection score value corresponding to the rejection module; The module compatibility value between the target service module and the target business scheduling module in the first edge computing node is determined based on the affinity score, the number of affinity modules, the rejection score, and the number of rejection modules.

10. The method according to claim 1, characterized in that, The plurality of edge computing nodes includes a second edge computing node, and the target latency includes a second latency for the second edge computing node to initiate a service scheduling request to the first edge computing node; wherein, determining the target latency for each edge computing node to initiate a service scheduling request to the first edge computing node based on the network status information uploaded by each edge computing node in the cloud-edge computing network includes: The connectivity and connection speed between edge computing nodes in the cloud-edge computing network are determined based on the network status information uploaded by each edge computing node in the cloud-edge computing network. The optimal path with the shortest connection time from the second edge computing node to the first edge computing node is determined based on the connectivity and connection speed between each edge computing node. The service scheduling characteristics of the second edge computing node when sending a service scheduling request, and the data characteristics of the first edge computing node issuing service scheduling decisions to the first edge computing node are obtained. Based on the service scheduling characteristics of the second edge computing node when sending a service scheduling request, the data characteristics of the first edge computing node issuing a service scheduling decision to the first edge computing node, and the connection speed between each edge computing node on the optimal path, the second delay for the second edge computing node to initiate a service scheduling request to the first edge computing node is determined.

11. A service scheduling device, characterized in that, A management and control node applied in a cloud-edge computing network, wherein the cloud-edge computing network also includes multiple edge computing nodes, and the service scheduling device includes: The data acquisition module is used to acquire network status information, service type information, and number of business scheduling requests uploaded by each edge computing node in the cloud-edge computing network. The activation utility value determination module is used to determine the activation utility value of each edge computing node based on the network status information, service type information and number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network. The candidate service scheduling node determination module is used to determine candidate service scheduling nodes among various edge computing nodes based on the activation utility value; An activation module is used to activate the service scheduling module of the candidate service scheduling node, and to use the candidate service scheduling node as the current service scheduling node in the cloud-edge computing network to process service scheduling requests in the cloud-edge computing network. The plurality of edge computing nodes include a first edge computing node, in which a service scheduling module is deployed as a target service scheduling module; wherein, the activation utility value of each edge computing node is determined based on network status information, service type information, and the number of service scheduling requests uploaded by each edge computing node in the cloud-edge computing network, including: The target latency for each edge computing node to initiate a service scheduling request to the first edge computing node is determined based on the network status information uploaded by each edge computing node in the cloud-edge computing network. The target service module enabled in the first edge computing node is determined based on the server type of the first edge computing node; Determine the module compatibility value between the target service module and the target business scheduling module; The probability of the first edge computing node performing a service scheduling request in the cloud-edge computing network is determined based on the number of service scheduling requests made by the first edge computing node. The target priority function value of the first edge computing node is determined based on the module compatibility value and the scheduling service probability. The activation utility value of the first edge computing node is determined based on the target latency and the target priority function value.

12. An electronic device, characterized in that, include: Memory; as well as A processor coupled to the memory, the processor being used to execute the service scheduling method as described in any one of claims 1-10 based on instructions stored in the memory.

13. A computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the service scheduling method as described in any one of claims 1-10.

Images