Task Processing Method, Apparatus, Device, and Storage Medium

Abstract

The present application provides a task processing method, apparatus, device, and storage medium, relating to the field of communication technologies. The method includes: monitoring whether the resource processing capacity of the MEC node is greater than or equal to a preset ratio to obtain a judgment result, where the resource processing capacity is used to represent the ratio of the used resources on the MEC node to the available resources of the MEC node; determining a target processing node according to the judgment result, the first delay, and the second delay, where the first delay is the sum of the communication delay and the computing delay for offloading the task to the fog node, and the second delay is the sum of the communication delay and the computing delay for offloading the task to the cloud node, and the target processing node is the fog node or the cloud node or the MEC node; and in response to the target processing node not being the MEC node, offloading the task to the target processing node for processing. The present application can reduce the delay of task processing.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to a task processing method, apparatus, device, and storage medium. Background Art

[0002] With the continuous development of artificial intelligence and mobile Internet technologies, business applications such as augmented reality, face recognition, image rendering, and autonomous driving have emerged in large numbers. These business applications usually require huge amounts of computing resources, storage resources, and energy consumption. Currently, the computing power of terminals is limited and the battery capacity is low, unable to meet the processing requirements of these business applications. Therefore, cloud computing has been proposed and developed rapidly.

[0003] Cloud computing uses virtualization technology to establish a super-large-capacity computing power resource pool, enabling various applications to obtain the required computing resources, storage resources, and software and platform services. The emergence of cloud computing meets the processing requirements of compute-intensive business applications, but business applications such as autonomous driving should also have the characteristic of being latency-sensitive. In many cases, the transmission latency from the terminal to the cloud cannot meet the requirements of these business applications for ultra-low latency. Therefore, the European Telecommunications Standards Institute (ETSI) established the Mobile Edge Computing (MEC) Industry Specification Group (ISG) in December 2014 to initiate the standardization of mobile edge computing to develop mobile edge computing. ETSI defines MEC as a network architecture that can provide Internet Technology (IT) and cloud computing functions near the location of mobile users in the radio access network, aiming to migrate IT and cloud computing from the core network to the edge access network to shorten the end-to-end latency of task processing and ensure data security and privacy. In September 2016, the concept of mobile edge computing was extended to Multi-access Edge Computing (MEC), further extending mobile edge computing from the telecommunications cellular network to other wireless networks to expand the applicability of edge computing in heterogeneous networks including Wireless Fidelity (WiFi) and fixed access technologies.

[0004] The large-scale deployment of edge computing devices and terminals has solved the problems of bandwidth shortage, network congestion, and excessive latency caused by uploading massive amounts of data in the network to the cloud computing center. However, it has also led to the ubiquitous deployment of computing power resources, inevitably resulting in the "computing power island" effect. On the one hand, edge computing nodes do not perform effective collaborative processing tasks, and the computing power resources of a single node cannot meet the computing power resource requirements of ultra-large-scale compute-intensive tasks such as image rendering, and still cannot solve the problem of ultra-low latency requirements for business applications with both compute-intensive and latency-sensitive characteristics. On the other hand, although some edge computing nodes are overloaded and unable to effectively process computing tasks, due to the uneven network load, some computing nodes are bound to remain idle, resulting in the underutilization of the computing power resources of the edge network.

[0005] Therefore, in order to efficiently and collaboratively utilize the heterogeneous computing power resources of the entire network, in 2019, operators, equipment manufacturers, etc. led to propose a technical solution for the integration of computing and network based on a distributed system - Computing-aware Networking (CAN for short), in order to achieve the joint optimization and scheduling of Information and Communications Technology (ICT) systems and provide end-to-end experience guarantee. CAN aims to connect and collaborate the diverse computing powers of cloud computing, edge computing, and terminals through the network, realize the deep integration and collaborative perception of computing and network, as well as the on-demand scheduling and efficient sharing of computing power resources.

[0006] Computing-aware routing and computing power resource allocation are a key issue in the research of computing-aware networks. In traditional network architectures, computing power and network are usually managed separately. In terms of computing power management, as a key technology of edge computing, after the concept of edge computing was proposed, many researchers have proposed task offloading strategies based on single-user multi-node, multi-user single-node, and multi-user multi-node. These strategies essentially perfectly match terminal tasks with edge computing nodes, but there are still problems with relatively high latency in task processing. Summary of the Invention

[0007] This application provides a task processing method, device, equipment, and storage medium to solve the problem of relatively high latency in task processing.

[0008] In a first aspect, the present application provides a task processing method, including: monitoring whether the resource processing capacity of the MEC node is greater than or equal to a preset ratio to obtain a judgment result, where the resource processing capacity is used to represent the ratio of the used resources on the MEC node to the available resources on the MEC node; determining a target processing node according to the judgment result, the first delay, and the second delay, where the first delay is the sum of the communication delay and the computing delay for offloading the task to the fog node, and the second delay is the sum of the communication delay and the computing delay for offloading the task to the cloud node, and the target processing node is the fog node or the cloud node or the MEC node; in response to the target processing node not being the MEC node, offloading the task to the target processing node for processing.

[0009] In a possible implementation manner, determining the target processing node according to the judgment result, the first delay, and the second delay includes: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, determining the magnitude relationship among the service tolerance delay corresponding to the task, the first delay, and the second delay; in response to both the first delay and the second delay being greater than the service tolerance delay, determining the target processing node as the MEC node; in response to only one of the first delay and the second delay being less than the service tolerance delay, determining the target processing node as the node among the cloud node and the fog node with a delay less than the service tolerance delay; in response to both the first delay and the second delay being less than the service tolerance delay, determining the target processing node as the node among the cloud node and the fog node with a smaller delay.

[0010] In a possible implementation manner, the above task processing method may further include: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, obtaining the first energy consumption for offloading the task to the fog node and the second energy consumption for offloading the task to the cloud node; in response to the first energy consumption being less than the second energy consumption, determining the target processing node as the fog node; in response to the first energy consumption being greater than the second energy consumption, determining the target processing node as the cloud node; in response to the first energy consumption being equal to the second energy consumption, determining the target processing node as the fog node or the cloud node.

[0011] In a possible implementation manner, the above task processing method may further include: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, based on the balanced requirement of energy consumption and delay, determining the target processing node as the cloud node or the fog node according to the first energy consumption, the second energy consumption, the first delay, and the second delay.

[0012] In a possible implementation manner, determining a target processing node according to the judgment result, the first time delay, and the second time delay includes: in response to the judgment result that the resource processing capability of the MEC node is less than a preset ratio, determining the magnitude relationship among the third time delay, the first time delay, and the second time delay for the MEC node to process the task; in response to both the first time delay and the second time delay being greater than the third time delay, determining the target processing node as the MEC node; in response to at least one of the first time delay and the second time delay being less than or equal to the third time delay, based on the principle of the minimum time delay, determining the target processing node according to the magnitude relationship among the first time delay, the second time delay, the third time delay, and the service tolerance time delay corresponding to the task.

[0013] In a possible implementation manner, based on the principle of the minimum time delay, determining the target processing node according to the magnitude relationship among the first time delay, the second time delay, the third time delay, and the service tolerance time delay corresponding to the task includes: in response to all of the first time delay, the second time delay, and the third time delay being less than the service tolerance time delay, determining the target processing node as the node with the minimum energy consumption among the MEC node, the cloud node, and the fog node.

[0014] In a possible implementation manner, based on the principle of the minimum time delay, determining the target processing node according to the magnitude relationship among the first time delay, the second time delay, the third time delay, and the service tolerance time delay corresponding to the task further includes: in response to both the first time delay and the second time delay being less than or equal to the third time delay, and the third time delay being greater than or equal to the service tolerance time delay, determining the target processing node according to the following situations: in response to both the first time delay and the second time delay being less than the service tolerance time delay, determining the target processing node as the node with the minimum energy consumption among the cloud node and the fog node; in response to the service tolerance time delay being between the first time delay and the second time delay, determining the target processing node as the node with the smaller time delay among the cloud node and the fog node; in response to both the first time delay and the second time delay being greater than the service tolerance time delay, determining the target processing node as the MEC node.

[0015] In a possible implementation manner, the third time delay is between the first time delay and the second time delay. Correspondingly, based on the principle of the minimum time delay, determining the target processing node according to the magnitude relationship among the first time delay, the second time delay, the third time delay, and the service tolerance time delay corresponding to the task further includes: in response to the third time delay being less than the service tolerance time delay, determining the target processing node as the node with the smaller energy consumption among the MEC node and the node with a time delay less than the third time delay; in response to the third time delay being greater than or equal to the service tolerance time delay, determining the target processing node as the node with a time delay less than the third time delay; in response to being less than the third time delay and greater than the service tolerance time delay, determining the target processing node as the MEC node.

[0016] In a second aspect, the present application provides a task processing device, including:

[0017] A monitoring module, configured to monitor whether the resource processing capacity of the MEC node is greater than or equal to a preset ratio, and obtain a judgment result, where the resource processing capacity is used to represent the ratio of the used resources on the MEC node to the available resources on the MEC node;

[0018] A determination module, configured to determine a target processing node according to the judgment result, a first delay, and a second delay, where the first delay is the sum of the communication delay and the computing delay for offloading a task to a fog node, and the second delay is the sum of the communication delay and the computing delay for offloading a task to a cloud node, and the target processing node is a fog node or a cloud node or an MEC node;

[0019] A processing module, configured to, in response to the target processing node not being the MEC node, offload the task to the target processing node for processing.

[0020] In a possible implementation manner, the determination module is specifically configured to: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, determine the magnitude relationship among the service tolerance delay, the first delay, and the second delay corresponding to the task; in response to both the first delay and the second delay being greater than the service tolerance delay, determine the target processing node as the MEC node; in response to only one of the first delay and the second delay being less than the service tolerance delay, determine the target processing node as the node with a delay less than the service tolerance delay among the cloud node and the fog node; in response to both the first delay and the second delay being less than the service tolerance delay, determine the target processing node as the node with a smaller delay among the cloud node and the fog node.

[0021] In a possible implementation manner, the determination module can also be configured to: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, obtain a first energy consumption for offloading the task to the fog node and a second energy consumption for offloading the task to the cloud node; in response to the first energy consumption being less than the second energy consumption, determine the target processing node as the fog node; in response to the first energy consumption being greater than the second energy consumption, determine the target processing node as the cloud node; in response to the first energy consumption being equal to the second energy consumption, determine the target processing node as the fog node or the cloud node.

[0022] In a possible implementation manner, the determination module can also be configured to: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, based on the balanced requirement of energy consumption and delay, determine the target processing node as the cloud node or the fog node according to the first energy consumption, the second energy consumption, the first delay, and the second delay.

[0023] In a possible implementation, the determining module can also be used to: in response to a judgment result that the resource processing capacity of the MEC node is less than a preset ratio, determine the magnitude relationship among the third time delay, the first time delay, and the second time delay for the MEC node to process a task; in response to both the first time delay and the second time delay being greater than the third time delay, determine that the target processing node is the MEC node; in response to at least one of the first time delay and the second time delay being less than or equal to the third time delay, based on the principle of the minimum time delay, determine the target processing node according to the magnitude relationship among the first time delay, the second time delay, the third time delay, and the service tolerance time delay corresponding to the task.

[0024] In a possible implementation, the determining module can also be used to: in response to all of the first time delay, the second time delay, and the third time delay being less than the service tolerance time delay, determine that the target processing node is the node with the minimum energy consumption among the MEC node, the cloud node, and the fog node.

[0025] In a possible implementation, the determining module can also be used to: in response to both the first time delay and the second time delay being less than or equal to the third time delay, and the third time delay being greater than or equal to the service tolerance time delay, determine the target processing node according to the following situations: in response to both the first time delay and the second time delay being less than the service tolerance time delay, determine that the target processing node is the node with the minimum energy consumption among the cloud node and the fog node; in response to the service tolerance time delay being between the first time delay and the second time delay, determine that the target processing node is the node with the smaller time delay among the cloud node and the fog node; in response to both the first time delay and the second time delay being greater than the service tolerance time delay, determine that the target processing node is the MEC node.

[0026] In a possible implementation, the third time delay is between the first time delay and the second time delay. The determining module is further used to: in response to the third time delay being less than the service tolerance time delay, determine that the target processing node is the node with the smaller energy consumption among the MEC node and the node with a time delay less than the third time delay; in response to the third time delay being greater than or equal to the service tolerance time delay, determine that the target processing node is the node with a time delay less than the third time delay; in response to being less than the third time delay and greater than the service tolerance time delay, determine that the target processing node is the MEC node.

[0027] In a third aspect, the present application provides a task processing device, including: a memory and a processor. The memory is used to store program instructions; the processor is used to call the program instructions in the memory to execute the task processing method of the first aspect.

[0028] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed, the task processing method of the first aspect is implemented.

[0029] Fifth aspect, the present application provides a computer program product, which includes a computer program that, when executed, is used to implement the task processing method of the first aspect.

[0030] For the task processing method, device, equipment, and storage medium provided by the present application, by monitoring whether the resource processing capacity of the MEC node is greater than or equal to a preset ratio, a judgment result is obtained, where the resource processing capacity is used to represent the ratio of the used resources on the MEC node to the available resources on the MEC node; according to the judgment result, the first delay, and the second delay, a target processing node is determined, where the first delay is the sum of the communication delay and the computing delay for offloading the task to the fog node, and the second delay is the sum of the communication delay and the computing delay for offloading the task to the cloud node, and the target processing node is the fog node or the cloud node or the MEC node; in response to the target processing node not being the MEC node, the task is offloaded to the target processing node for processing. The task processing method provided by the present application combines the resource processing capacity of the MEC node, the first delay, and the second delay to determine the target processing node of the task, achieving the effect of dynamically selecting the target processing node of the task to reduce the delay of task processing. Description of the Drawings

[0031] The drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0032] Figure 1 It is a schematic diagram of applying the task processing method provided by an embodiment of the present application to the topology diagram of the computing power network system;

[0033] Figure 2 It is a schematic flowchart of the task processing method provided by an embodiment of the present application;

[0034] Figure 3 It is a schematic structural diagram of the task processing device provided by an embodiment of the present application;

[0035] Figure 4 It is a schematic structural diagram of the task processing equipment provided by an embodiment of the present application.

[0036] Through the above drawings, the clear embodiments of the present application have been shown, and there will be more detailed descriptions later. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application in any way, but to explain the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed Embodiments

[0037] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0038] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented, for example, in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, products, or devices.

[0039] It should be clear that, unless otherwise specified, the "MEC node" mentioned in the embodiments of the present application generally refers to a multi-access edge computing node.

[0040] In view of the problems existing in the related art, the present application proposes a task processing method. When there are many tasks in the MEC node, this method considers offloading the tasks to a fog node or a cloud node for processing. Specifically, based on the relationship between the resource processing capacity of the MEC node and a preset ratio, a target processing node corresponding to the task is determined among the MEC node, the fog node, and the cloud node. When the target processing node is not the MEC node, the task is offloaded to the target processing node. That is, it dynamically senses the resource processing capacity of the MEC node and the delays of the fog node and the cloud node, and selects the target processing node for processing the task, thereby reducing the delay of task processing.

[0041] Furthermore, the energy consumption of the MEC node, the fog node, and the cloud node can be comprehensively considered, and the node with the minimum delay and the least energy consumption is selected as the target processing node.

[0042] Figure 1 Schematic diagram of applying the task processing method provided in an embodiment of the present application to the topology diagram of a computing power network system. As Figure 1As shown in the figure, the computing power network system topology includes terminals, routing nodes (R), MEC nodes, fog nodes (which can also be referred to as "fog node servers" or "Fog Node Servers"), and cloud nodes (which can also be referred to as "central servers" or "CenterServers"). Among them, there can be multiple routing nodes. The fog node server and the central server can store computing resources. The terminal can be a mobile phone, a computer, a laptop, etc.

[0043] Figure 1 It also includes the connections between each node and the weights between each connection. Exemplarily, such as W 12_R represents the weight between routing node 1 and routing node 2, and W 31_N represents the weight between routing node 3 and the MEC node. The representation of the remaining weights is the same. Among them, the weight can be a comprehensive value in dimensions such as path length and bandwidth occupancy rate.

[0044] Exemplarily, based on the known weights, the node server can be selected according to the minimum value of the weight calculation result. Among them, the node server includes the fog node server and the cloud node server. The node server is for the server cluster and does not refer to a single type of server. It is composed of multiple nodes and the management unit of the overall management device. Each node has a module management department for switching the action mode of the node. The module management department switches the individual action of each node or coordinates with other nodes according to the composition information transmitted from the management unit. As Figure 1 shown, in this embodiment, the user can transmit information to the node server through the terminal. The transmitted information will pass through some routes. Different routes will result in different routing nodes reached and different final node servers reached. Exemplarily, based on the known weights, the route can be selected according to the minimum value of the weight calculation result. Different selected routes will have different time delays. The node server can be selected according to the principle of the minimum time delay. If there is a situation where the time delays are the same, the node server can be selected according to the principle of the minimum energy consumption.

[0045] Those skilled in the art can understand that Figure 1 the shown computing power network system topology diagram does not limit the system model applicable to the task processing method, and may include more or fewer routing nodes, as well as cloud node servers and fog node servers than shown in the figure.

[0046] Next, in combination with Figure 1 the example of Figure 2 the task processing method according to the exemplary embodiments of the present application will be described with reference to Figure 1 . It should be noted that the above application scenarios are only shown for the convenience of understanding the spirit and principle of the present application, and the embodiments of the present application are not limited by

[0047] Figure 2 This is a schematic flowchart of a task processing method provided by an embodiment of the present application. As Figure 2 shown, the task processing method in the embodiment of the present application includes the following steps:

[0048] S201: Monitor whether the resource processing capacity of the MEC node is greater than or equal to a preset ratio to obtain a judgment result, where the resource processing capacity is used to represent the ratio of the used resources on the MEC node to the available resources on the MEC node.

[0049] It can be understood that the greater the resource processing capacity, the greater the latency for the MEC node to process tasks. When the resource processing capacity is relatively small, the latency for the MEC node to process tasks is usually within the service tolerance latency. Therefore, by monitoring the resource processing capacity of the MEC node, two judgment results can be obtained: the resource processing capacity of the MEC node is greater than or equal to the preset ratio, or the resource processing capacity of the MEC node is less than the preset ratio.

[0050] In this step, the resource processing capacity of the MEC node can be obtained in various ways. For example, the server uploads the resource capacity of the MEC node to a centralized controller (similar to a resource scheduling and management platform), and the centralized controller can collect information such as the resource processing capacity of the MEC node in real time. Among them, the preset ratio can be artificially set according to the actual situation and / or empirical values. For example, the maximum number of tasks that a certain MEC node can process simultaneously is 500. When this MEC node processes 400 tasks simultaneously and considers using the task processing method proposed in the present application, the preset ratio is 400 to 500, that is, 80%.

[0051] For example, the judgment result includes that the resource processing capacity of the MEC node is greater than or equal to the preset ratio or the resource processing capacity of the MEC node is less than the preset ratio. If the preset ratio is 80%, then it is that the resource processing capacity of the MEC node ≥ 80%, or the resource processing capacity of the MEC node < 80%.

[0052] S202: Determine a target processing node according to the judgment result, the first latency, and the second latency, where the first latency is the sum of the communication latency and the computing latency for offloading the task to the fog node, and the second latency is the sum of the communication latency and the computing latency for offloading the task to the cloud node, and the target processing node is the fog node or the cloud node or the MEC node.

[0053] In this step, on the premise of the judgment result, the target processing node is determined by combining the first latency and the second latency to achieve the purpose of reducing the task processing latency.

[0054] For example, if represents the communication latency for offloading task n to fog node m, represents the computing latency for offloading task n to fog node m, Denotes the communication delay of offloading task n to cloud node c. Denotes the computing delay of offloading task n to cloud node c, T mec_wait_com Denotes the waiting and processing delay of services at the MEC node, E n，mec Denotes the energy consumption of offloading task n to the MEC node. Then the first delay is The second delay is The third delay is T mec_wait_com .

[0055] Among them, when the judgment result is different, the specific implementation of determining the target processing node is also different. For details, please refer to the subsequent embodiments. Among them, the target processing node is a fog node, a cloud node or an MEC node. If the target processing node is an MEC node, the MEC node still processes this task; if the target processing node is a fog node or a cloud node, then step S203 is executed.

[0056] S203: In response to the target processing node not being an MEC node, offload the task to the target processing node for processing.

[0057] The task processing method provided by the embodiments of this application monitors whether the resource processing capacity of the MEC node is greater than or equal to a preset ratio to obtain a judgment result. The resource processing capacity is used to represent the ratio of the used resources on the MEC node to the available resources on the MEC node; according to the judgment result, the first delay and the second delay, determine the target processing node, where the first delay is the sum of the communication delay and the computing delay of offloading the task to the fog node, and the second delay is the sum of the communication delay and the computing delay of offloading the task to the cloud node. The target processing node is a fog node, a cloud node or an MEC node; in response to the target processing node not being an MEC node, offload the task to the target processing node for processing. By combining the resource processing capacity of the MEC node, the first delay and the second delay to determine the target processing node of the task, the effect of dynamically selecting the target processing node of the task is achieved, so as to reduce the delay of task processing.

[0058] Based on the above embodiments, next, it will be described in different cases how to "determine the target processing node according to the judgment result, the first delay and the second delay".

[0059] First of all, the calculation methods of delays, energy consumption, etc. involved in the embodiments of this application will be described. Exemplarily, the parameter descriptions in the calculation formula are as follows:

[0060] T _n_t Denotes the service tolerance delay;

[0061] Z n Denotes the total task volume;

[0062] Denotes a transmission delay introduced when offloading task n to a cloud node;

[0063] R n，c Denotes the transmission rate of task n to a cloud node;

[0064] R n，m Denotes the transmission rate of task n to a fog node;

[0065] R n，m (a) Denotes the transmission rate of task n to fog node m when task n exclusively occupies the channel of fog node m;

[0066] n m (a) Denotes the number of tasks selecting fog node m;

[0067] E n，m Denotes the energy consumption of offloading task n to fog node m;

[0068] E n，c Denotes the energy consumption of offloading task n to cloud node c;

[0069] Denotes the transmission power of task n communicating with a cloud node;

[0070] Denotes the transmission power of task n communicating with a fog node;

[0071] f c Denotes the computing power of a cloud node;

[0072] f m Denotes the computing power of a fog node;

[0073] E n，mec Denotes the energy consumption of offloading task n to an MEC node;

[0074] Denotes the transmission power of task n communicating with an MEC node;

[0075] R n，mec Denotes the transmission rate of task n to an MEC node;

[0076] n mec (a) Denotes the number of tasks selecting an MEC node.

[0077] Among them, T _n_t and T mec_wait_com 、Z n 、R n，c 、R n，m 、R n，mec 、n m (a)、n mec (a)、 f c and f m are known quantities that can be obtained in advance.

[0078] According to the above parameter descriptions, through calculation, the delay and energy consumption parameters required for the embodiments of this application are obtained. The calculation process is as follows:

[0079] (1) Transmission rate of task n to fog node m:

[0080] (2) The communication delay, computing delay, and energy consumption for offloading task n to fog node m can be expressed as follows:

[0081] Communication delay:

[0082] Computing delay: where γ n represents the computing efficiency. In an ideal case, γ n = 1.

[0083] Energy consumption:

[0084] (3) The communication delay, computing delay, and energy consumption for offloading task n to cloud node c can be expressed as follows:

[0085] Communication delay:

[0086] Computing delay:

[0087] Energy consumption:

[0088] (4) The energy consumption for offloading task n to the MEC node can be expressed as:

[0089] Energy consumption:

[0090] Exemplarily, a preset ratio is set to 80%.

[0091] Based on the above, in the first case, determining the target processing node according to the judgment result, the first delay, and the second delay may further include: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, determining the magnitude relationship among the service tolerance delay, the first delay, and the second delay corresponding to the task; in response to both the first delay and the second delay being greater than the service tolerance delay, determining the target processing node as the MEC node; in response to only one of the first delay and the second delay being less than the service tolerance delay, determining the target processing node as the node among the cloud node and the fog node with a delay less than the service tolerance delay; in response to both the first delay and the second delay being less than the service tolerance delay, determining the target processing node as the node among the cloud node and the fog node with a smaller delay.

[0092] That is to say, when the resource processing capacity of the MEC node ≥ 80%:

[0093] If and Then the task can be directly processed on the MEC node, that is, the target processing node is the MEC node. In this case, since the delays of the fog node and the cloud node in processing the task are both greater than the service tolerance delay corresponding to the task, offloading the task to the fog node or the cloud node for processing will not significantly reduce the task processing delay. Therefore, the task is still processed by the MEC node at this time.

[0094] If and Then offload the task to the cloud node for processing;

[0095] If and Then offload the task to the fog node for processing;

[0096] If and Then offload the task to the node with a smaller delay among the cloud node and the fog node for processing. For example, when both the first delay and the second delay are less than the service tolerance delay, compare the magnitudes of the first delay and the second delay. If the first delay is greater than the second delay, select the cloud node as the target processing node for the service to be processed; if the first delay is less than the second delay, select the fog node as the target processing node for the service to be processed; if the first delay is equal to the second delay, select the cloud node or the fog node as the target processing node for the service to be processed.

[0097] Further, the target processing node can also be determined by combining the energy consumption of the cloud node and the fog node for processing tasks. In one implementation, the above task processing method may further include: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, obtaining the first energy consumption of offloading the task to the fog node and the second energy consumption of offloading the task to the cloud node; in response to the first energy consumption being less than the second energy consumption, determining the target processing node as the fog node; in response to the first energy consumption being greater than the second energy consumption, determining the target processing node as the cloud node; in response to the first energy consumption being equal to the second energy consumption, determining the target processing node as the fog node or the cloud node.

[0098] Exemplarily, when the resource processing capacity of the MEC ≥ 80%, energy consumption is introduced, and on the basis of ensuring a small delay, energy consumption is considered:

[0099] If and the energy consumption of the two nodes can be compared and considered in different cases:

[0100] Case 1: If E n，m <E n，c , the task is offloaded to the fog node for processing;

[0101] Case 2: If E n，c <E n，m , the task is offloaded to the cloud node for processing.

[0102] The embodiments of the present application comprehensively consider server resources and energy consumption usage, dynamically sense server delay, energy consumption, etc., and perform traffic node offloading, thereby achieving a significant reduction in network energy consumption.

[0103] In addition, based on the balance requirement of energy consumption and delay, the target processing node can also be determined by combining the energy consumption weight and the delay weight. Therefore, the above task processing method may further include: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, based on the balance requirement of energy consumption and delay, according to the first energy consumption, the second energy consumption, the first delay, and the second delay, determining the target processing node as the cloud node or the fog node.

[0104] Exemplarily, if delay and energy consumption are equally important for the method proposed in the present application, the delay weight and the energy consumption weight can each be 0.5. If delay is more important than energy consumption for the method proposed in the present application, the delay weight can be 0.7, and the energy weight can be 0.3. The specific weight values can be set according to actual requirements or historical experience, and the embodiments of the present application do not limit them.

[0105] In the second case, the above task processing method may further include: in response to the judgment result that the resource processing capacity of the MEC node is less than the preset ratio, determining the magnitude relationship among the third delay, the first delay, and the second delay for the MEC node to process the task; in response to both the first delay and the second delay being greater than the third delay, determining the target processing node as the MEC node; in response to at least one of the first delay and the second delay being less than or equal to the third delay, based on the principle of the minimum delay, determining the target processing node according to the magnitude relationship among the first delay, the second delay, the third delay, and the service tolerance delay corresponding to the task.

[0106] Exemplarily, when the resource processing capacity of the MEC node < 80%, if and then the task can be directly processed on the MEC node; for other cases, reference can be made to the subsequent embodiments.

[0107] Optionally, based on the principle of the minimum delay, determining the target processing node according to the magnitude relationship among the first delay, the second delay, the third delay, and the service tolerance delay corresponding to the task includes: in response to all of the first delay, the second delay, and the third delay being less than the service tolerance delay, determining the target processing node as the node with the minimum energy consumption among the MEC node, the cloud node, and the fog node.

[0108] Exemplarily, when the resource processing capacity of the MEC node < 80%, compare the relationship between the service tolerance delay T _n_t and the waiting and processing delay T mec_wait_com of the service at the MEC node. Specifically, when T mec_wait_com < T _n_t :

[0109] If and then select the node with the minimum energy consumption as the target processing node, that is, select Min{E n，m , E n，c , E n，mec};

[0110] If The target processing node can be selected according to the relationship between T _n_t and the sum of the communication delay and the computing delay for offloading the task to the cloud node. That is, if then select the node with the minimum energy consumption as the target processing node, that is, select Min{E n，m , E n，c , E n，mec}.

[0111] If The target processing node can be selected according to the relationship between T _n_tRelating to the sum of the communication delay and the computing delay for offloading a task to a fog node, select a target processing node, that is, if then select the node with the minimum energy consumption as the target processing node, that is, select Min{E n，m , E n，c , E n，mec}.

[0112] Furthermore, based on the principle of minimum delay, determine the target processing node according to the magnitude relationship between the first delay, the second delay, the third delay and the service tolerance delay corresponding to the task. It may further include: in response to both the first delay and the second delay being less than or equal to the third delay, and the third delay being greater than or equal to the service tolerance delay, determine the target processing node according to the following situations: in response to both the first delay and the second delay being less than the service tolerance delay, determine the target processing node as the node with the minimum energy consumption among the cloud node and the fog node; in response to the service tolerance delay being between the first delay and the second delay, determine the target processing node as the node with the smaller delay among the cloud node and the fog node; in response to both the first delay and the second delay being greater than the service tolerance delay, determine the target processing node as the MEC node.

[0113] Exemplarily, when the resource processing capacity of the MEC node < 80%, if and the service tolerance delay T _n_t and the waiting and processing delay T mec_wait_com of the service at the MEC node can be compared. If T _n_t ≤T mec_wait_com , select the target processing node according to the relationship between T _n_t and the sum of the communication delay and the computing delay for offloading the task to the fog node, and the sum of the communication delay and the computing delay for offloading the task to the cloud node. Specifically:

[0114] If then select the node with the minimum energy consumption as the target processing node according to the principle of minimum energy consumption, that is, select Min{E n，m , E n，c}.

[0115] If then select the fog node as the target processing node.

[0116] If then the task can be processed on the MEC node or the processing of the task can be abandoned.

[0117] Further, the third time delay is between the first time delay and the second time delay. Based on the principle of minimum time delay, according to the magnitude relationship between the first time delay, the second time delay, the third time delay and the service tolerance time delay corresponding to the task, determining the target processing node further includes: in response to the third time delay being less than the service tolerance time delay, determining that the target processing node is the node with the smaller energy consumption among the MEC node and the nodes with a time delay less than the third time delay; in response to the third time delay being greater than or equal to the service tolerance time delay, determining that the target processing node is the node with a time delay less than the third time delay; in response to the time delay being less than the third time delay and greater than the service tolerance time delay, determining that the target processing node is the MEC node.

[0118] Exemplarily, when the resource processing capacity of the MEC node < 80%, if the service tolerance time delay T _n_t and the waiting and processing time delay T mec_wait_com of the service at the MEC node can be compared, when T mec_wait_com < T _n_t , if then according to the principle of minimum energy consumption, select the node with the minimum energy consumption as the target processing node, that is, select Min{E n，m , E n，mec}; if then according to the principle of minimum energy consumption, select the node with the minimum energy consumption as the target processing node, that is, select Min{E n，c , E n，mec}.

[0119] Exemplarily, when the resource processing capacity of the MEC node < 80%, if the service tolerance time delay T _n_t and the waiting and processing time delay T mec_wait_com of the service at the MEC node can be compared, when T _n_t ≤ T mec_wait_com , if then select the fog node as the target processing node; if then select the cloud node as the target processing node.

[0120] Exemplarily, when the resource processing capacity of the MEC node < 80%, if the service tolerance time delay T _n_t and the waiting and processing time delay T mec_wait_com of the service at the MEC node can be compared, when T _n_t ≤ T mec_wait_com , if or Then, it can be selected to process the task on the MEC node or abandon processing the task.

[0121] In summary, the task processing method provided by this application realizes the purpose of dynamically selecting the target processing node by judging the resource processing ability of the MEC node and combining the time delay, which can reduce the time delay of task processing; further, by introducing energy consumption and comprehensively considering the time delay and energy consumption, the goal of processing tasks with lower time delay and less energy consumption is achieved.

[0122] The following is the device embodiment of this application, which can be used to execute the method embodiment of this application. For the details not disclosed in the device embodiment of this application, please refer to the method embodiment of this application.

[0123] Figure 3 It is a schematic structural diagram of a task processing device provided by an embodiment of this application. For the sake of simplicity, only the parts related to the embodiment of this application are shown. As Figure 3 shown, the task processing device 30 includes: a monitoring module 31, a determination module 32, and a processing module 33. Among them,

[0124] The monitoring module 31 is configured to monitor whether the resource processing ability of the MEC node is greater than or equal to a preset ratio to obtain a judgment result, and the resource processing ability is used to represent the ratio of the used resources on the MEC node to the available resources of the MEC node.

[0125] The determination module 32 is configured to determine the target processing node according to the judgment result, the first time delay, and the second time delay. The first time delay is the sum of the communication time delay and the calculation time delay for offloading the task to the fog node, and the second time delay is the sum of the communication time delay and the calculation time delay for offloading the task to the cloud node. The target processing node is the fog node or the cloud node or the MEC node.

[0126] The processing module 33 is configured to, in response to the target processing node not being the MEC node, offload the task to the target processing node for processing.

[0127] In a possible implementation manner, the determination module 32 is specifically configured to: in response to the judgment result that the resource processing ability of the MEC node is greater than or equal to the preset ratio, determine the magnitude relationship among the service tolerance time delay, the first time delay, and the second time delay corresponding to the task; in response to both the first time delay and the second time delay being greater than the service tolerance time delay, determine the target processing node as the MEC node; in response to only one of the first time delay and the second time delay being less than the service tolerance time delay, determine the target processing node as the node with a time delay less than the service tolerance time delay among the cloud node and the fog node; in response to both the first time delay and the second time delay being less than the service tolerance time delay, determine the target processing node as the node with a smaller time delay among the cloud node and the fog node.

[0128] In a possible implementation, the determining module 32 can also be used to: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to a preset ratio, and both the first delay and the second delay are less than the service tolerance delay, obtain the first energy consumption for offloading the task to the fog node and the second energy consumption for offloading the task to the cloud node; in response to the first energy consumption being less than the second energy consumption, determine that the target processing node is the fog node; in response to the first energy consumption being greater than the second energy consumption, determine that the target processing node is the cloud node; in response to the first energy consumption being equal to the second energy consumption, determine that the target processing node is the fog node or the cloud node.

[0129] In a possible implementation, the determining module 32 can also be used to: in response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to a preset ratio, and both the first delay and the second delay are less than the service tolerance delay, based on the balanced requirements of energy consumption and delay, determine the target processing node as the cloud node or the fog node according to the first energy consumption, the second energy consumption, the first delay, and the second delay.

[0130] In a possible implementation, the determining module 32 can also be used to: in response to the judgment result that the resource processing capacity of the MEC node is less than the preset ratio, determine the magnitude relationship among the third delay, the first delay, and the second delay for the MEC node to process the task; in response to both the first delay and the second delay being greater than the third delay, determine that the target processing node is the MEC node; in response to at least one of the first delay and the second delay being less than or equal to the third delay, based on the principle of minimum delay, determine the target processing node according to the magnitude relationship among the first delay, the second delay, the third delay, and the service tolerance delay corresponding to the task.

[0131] In a possible implementation, the determining module 32 can also be used to: in response to all of the first delay, the second delay, and the third delay being less than the service tolerance delay, determine that the target processing node is the node with the minimum energy consumption among the MEC node, the cloud node, and the fog node.

[0132] In a possible implementation, the determining module 32 can also be used to: in response to both the first delay and the second delay being less than or equal to the third delay, and the third delay being greater than or equal to the service tolerance delay, determine the target processing node according to the following situations: in response to both the first delay and the second delay being less than the service tolerance delay, determine that the target processing node is the node with the minimum energy consumption among the cloud node and the fog node; in response to the service tolerance delay being between the first delay and the second delay, determine that the target processing node is the node with the smaller delay among the cloud node and the fog node; in response to both the first delay and the second delay being greater than the service tolerance delay, determine that the target processing node is the MEC node.

[0133] In a possible implementation, the third time delay is between the first time delay and the second time delay. The determining module 32 is further configured to: in response to the third time delay being less than the service tolerance time delay, determine that the target processing node is the node with lower energy consumption among the MEC node and the nodes with time delays less than the third time delay; in response to the third time delay being greater than or equal to the service tolerance time delay, determine that the target processing node is the node with a time delay less than the third time delay; and in response to the time delay being less than the third time delay and greater than the service tolerance time delay, determine that the target processing node is the MEC node.

[0134] The task processing device provided in the embodiments of the present application has the same implementation principle and technical effects as the above embodiments. For specific references, please refer to the above embodiments and will not be elaborated here.

[0135] Figure 4 It is a schematic structural diagram of a task processing device provided in an embodiment of the present application. As Figure 4 shown, the task processing device 40 includes: at least one processor 410, a memory 420, a communication interface 430, and a system bus 440. Among them, the memory 420 and the communication interface 430 are connected to the processor 410 through the system bus 440 to complete mutual communication. The memory 420 is used to store instructions, the communication interface 430 is used to communicate with other devices, and the processor 410 is used to call the instructions in the memory to execute the solutions of the above task processing method embodiments. The specific implementation manners and technical effects are similar and will not be elaborated here.

[0136] Figure 4 The processor 410 mentioned in

[0137] The memory 420 may include a Random Access Memory (RAM), and may also include a Static Random Access Memory (SRAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), an Erasable Programmable Read-Only Memory (EPROM), a Programmable Read-Only Memory (PROM), a Read Only Memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disc, such as at least one magnetic disk memory.

[0138] The communication interface 430 is used to implement communication between the task processing device and other devices (such as clients).

[0139] The system bus 440 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus 440 can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity, only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0140] Those skilled in the art can understand that Figure 4 The illustrated task processing device does not constitute a limitation on the task processing device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0141] The embodiments of the present application also provide a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed, the above task processing method is implemented.

[0142] The embodiments of the present application also provide a computer program product, including a computer program, which implements the above task processing method when executed.

[0143] The embodiments of the present application also provide a chip for running instructions. The chip is used to execute the task processing method in any of the above method embodiments.

[0144] Other embodiments of the present application will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include well-known knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and examples are only illustrative, and the true scope and spirit of the present application are pointed out by the following claims.

[0145] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A task processing method, characterized in that, Including: Monitoring whether the resource processing capacity of a multi-access edge computing (MEC) node is greater than or equal to a preset ratio to obtain a judgment result, where the resource processing capacity is used to represent the ratio of the used resources on the MEC node to the available resources on the MEC node; Determining a target processing node according to the judgment result, a first delay, and a second delay, where the first delay is the sum of the communication delay and the computing delay for offloading a task to a fog node, and the second delay is the sum of the communication delay and the computing delay for offloading a task to a cloud node, and the target processing node is the fog node or the cloud node or the MEC node; In response to the target processing node not being the MEC node, offloading the task to the target processing node for processing.

2. The task processing method according to claim 1, wherein The determining the target processing node according to the judgment result, the first delay, and the second delay includes: In response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, determining the magnitude relationship among the service tolerance delay corresponding to the task, the first delay, and the second delay; In response to both the first delay and the second delay being greater than the service tolerance delay, determining the target processing node as the MEC node; In response to only one of the first delay and the second delay being less than the service tolerance delay, determining the target processing node as the node among the cloud node and the fog node with a delay less than the service tolerance delay; In response to both the first delay and the second delay being less than the service tolerance delay, determining the target processing node as the node with a smaller delay among the cloud node and the fog node.

3. The task processing method according to claim 2, wherein It further includes: In response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, obtaining a first energy consumption for offloading the task to the fog node and a second energy consumption for offloading the task to the cloud node; In response to the first energy consumption being less than the second energy consumption, determining the target processing node as the fog node; In response to the first energy consumption being greater than the second energy consumption, determining the target processing node as the cloud node; In response to the first energy consumption being equal to the second energy consumption, determining the target processing node as the fog node or the cloud node.

4. The task processing method according to claim 3, wherein It further includes: In response to the judgment result that the resource processing capacity of the MEC node is greater than or equal to the preset ratio, and both the first delay and the second delay are less than the service tolerance delay, based on the balanced requirement of energy consumption and delay, determining the target processing node as the cloud node or the fog node according to the first energy consumption, the second energy consumption, the first delay, and the second delay.

5. The task processing method according to any one of claims 1 to 4, characterized in that, The determining the target processing node according to the judgment result, the first delay, and the second delay includes: In response to the judgment result that the resource processing capacity of the MEC node is less than the preset ratio, determining the magnitude relationship among a third delay for the MEC node to process the task, the first delay, and the second delay; In response to both the first delay and the second delay being greater than the third delay, determine that the target processing node is the MEC node; In response to at least one of the first delay and the second delay being less than or equal to the third delay, based on the principle of minimum delay, determine the target processing node according to the magnitude relationship between the first delay, the second delay, the third delay and the service tolerance delay corresponding to the task.

6. The task processing method according to claim 5, characterized in that, The determining the target processing node according to the magnitude relationship between the first delay, the second delay, the third delay and the service tolerance delay corresponding to the task based on the principle of minimum delay includes: In response to the first delay, the second delay and the third delay all being less than the service tolerance delay, determine that the target processing node is the node with the minimum energy consumption among the MEC node, the cloud node and the fog node.

7. The task processing method according to claim 5, wherein The determining the target processing node according to the magnitude relationship between the first delay, the second delay, the third delay and the service tolerance delay corresponding to the task based on the principle of minimum delay further includes: In response to the first delay and the second delay both being less than or equal to the third delay, and the third delay being greater than or equal to the service tolerance delay, determine the target processing node according to the following situations: In response to the first delay and the second delay both being less than the service tolerance delay, determine that the target processing node is the node with the minimum energy consumption among the cloud node and the fog node; In response to the service tolerance delay being between the first delay and the second delay, determine that the target processing node is the node with the smaller delay among the cloud node and the fog node; In response to the first delay and the second delay both being greater than the service tolerance delay, determine that the target processing node is the MEC node.

8. The task processing method according to claim 5, wherein When the third delay is between the first delay and the second delay, the determining the target processing node according to the magnitude relationship between the first delay, the second delay, the third delay and the service tolerance delay corresponding to the task based on the principle of minimum delay further includes: In response to the third delay being less than the service tolerance delay, determine that the target processing node is the node with the smaller energy consumption among the MEC node and the node with a delay less than the third delay; In response to the third delay being greater than or equal to the service tolerance delay, determine that the target processing node is the node with a delay less than the third delay; In response to being less than the third delay and greater than the service tolerance delay, determine that the target processing node is the MEC node.

9. A task processing device, characterized in that, including: A monitoring module, configured to monitor whether the resource processing capacity of a multi-access edge computing (MEC) node is greater than or equal to a preset ratio to obtain a judgment result, where the resource processing capacity is used to characterize the ratio of the used resources on the MEC node to the available resources on the MEC node; A determination module, configured to determine a target processing node according to the determination result, a first time delay, and a second time delay, where the first time delay is the sum of a communication time delay and a computing time delay for offloading a task to a fog node, the second time delay is the sum of a communication time delay and a computing time delay for offloading a task to a cloud node, and the target processing node is the fog node or the cloud node or the MEC node; A processing module, configured to, in response to the target processing node not being the MEC node, offload the task to the target processing node for processing.

10. A task processing device, characterized in that, Comprising: A memory and a processor; The memory is configured to store program instructions; The processor is configured to call the program instructions in the memory to execute the task processing method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, Computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed, the task processing method according to any one of claims 1 to 8 is implemented.

Images