A data processing method, device and readable storage medium based on edge computing

Abstract

The present application discloses a data processing method, device and readable storage medium based on edge computing nodes, the method comprising: obtaining the total idle computing power resource information of the edge computing node cluster; predicting the total change computing power resource information of the target application for the edge computing node cluster within the target time period; if the edge computing node cluster meets the node sleep condition, the edge computing node to be dormant is determined in the edge computing node cluster according to the total idle computing power resource information and the total change computing power resource information, and the dormant edge computing node to be dormant is processed within the target time period; if the edge computing node cluster meets the node wake-up condition, the edge computing node to be awakened is determined in the edge computing node cluster according to the total idle computing power resource information and the total change computing power resource information, and the awakened edge computing node to be awakened is processed within the target time period. By adopting the present application, the operating cost of the edge computing node can be reduced while meeting the computing power demand.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a data processing method, device and readable storage medium based on edge computing. Background Art

[0002] Cloud gaming refers to the process of running a game on a remote server, compressing and encoding the rendered game screen, and then sending it to the terminal through the network in the form of audio and video streams. Cloud gaming does not need to consider terminal configuration, which completely solves the technical problem of insufficient terminal performance and inability to run heavy games. However, cloud gaming has very high requirements for network latency. In order to provide more stable network conditions for the object, cloud gaming servers are generally closer to the object through large-scale deployment of edge computing nodes.

[0003] However, the online number of game objects is quite tidal. In order to provide a better experience for the objects, the online number of game objects is usually preset (the online number is set high), and then computing resources are prepared for cloud gaming applications according to this online number. However, during non-peak hours, the actual number of online game objects is far less than the preset number, so the computing resources prepared by the edge computing nodes will be idle during non-peak hours.

[0004] That is to say, during non-peak hours, edge computing nodes will run at a higher frequency to provide greater computing resources. Not only will the computing resources of the edge computing nodes be wasted, but edge computing nodes running at a higher frequency throughout the day will also increase power consumption (i.e., power loss of edge computing nodes), greatly increase the failure rate, and thus increase operating costs. Summary of the invention

[0005] The embodiments of the present application provide a data processing method, device, and readable storage medium based on edge computing, which can reduce the operating costs of edge computing nodes while meeting computing power requirements.

[0006] On the one hand, an embodiment of the present application provides a data processing method based on edge computing, including:

[0007] Obtain the total idle computing power resource information of the edge computing node cluster; the total idle computing power resource information is the computing power resource information of the edge computing node cluster that is not occupied by the target application when the target application is running;

[0008] Predict the total change in computing power resources of the edge computing node cluster for the target application within the target time period;

[0009] If the edge computing node cluster meets the node sleep condition, the edge computing nodes to be dormant are determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, and the dormant edge computing nodes are put into sleep mode within the target time period;

[0010] If the edge computing node cluster meets the node wake-up condition, the edge computing node to be awakened is determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, and the edge computing node to be awakened is awakened within the target time period.

[0011] On the one hand, an embodiment of the present application provides a data processing device based on edge computing, including:

[0012] The idle resource acquisition module is used to obtain the total idle computing power resource information of the edge computing node cluster; the total idle computing power resource information is the computing power resource information of the edge computing node cluster that is not occupied by the target application when the target application is running;

[0013] The change resource prediction module is used to predict the total change computing power resource information of the edge computing node cluster for the target application within the target time period;

[0014] A sleep processing module is used to determine the edge computing nodes to be dormant in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node dormancy condition;

[0015] The sleep processing module is also used to perform sleep processing on the sleep edge computing nodes within a target time period;

[0016] A wake-up processing module is used to determine the edge computing node to be awakened in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node wake-up condition;

[0017] The wake-up processing module is also used to perform wake-up processing on the edge computing node to be awakened within a target time period.

[0018] In one embodiment, the sleep processing module includes:

[0019] A sleep quantity determination unit, configured to determine the predicted total idle computing power resource information of the edge computing node cluster within a target time period according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node sleep condition;

[0020] The sleep quantity determination unit is further used to determine the sleep quantity of nodes according to the predicted total idle computing power resource information and the unit computing power resource information; the unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster;

[0021] The dormant node determination unit is used to determine the edge computing nodes to be dormant in the edge computing node cluster according to the number of dormant nodes.

[0022] In one embodiment, the sleepy node determination unit includes:

[0023] The dormant node acquisition subunit is used to acquire the dormant edge computing nodes in the edge computing node cluster; the dormant edge computing nodes are edge computing nodes in the edge computing node cluster that are currently in a dormant state;

[0024] The polling table acquisition subunit is used to obtain the node sleep polling table corresponding to the edge computing node cluster; the node sleep polling table includes the sleep polling order of each edge computing node in the edge computing node cluster;

[0025] The dormant node determination subunit is used to obtain the edge computing nodes to be dormant from the node dormant polling table in sequence according to the position of the dormant edge computing nodes in the node dormant sorting table and the number of dormant nodes; the edge computing nodes to be dormant are currently in normal operating state.

[0026] In one embodiment, the sleepy node determination unit includes:

[0027] The normal node acquisition subunit is used to acquire N normally operating edge computing nodes in the edge computing node cluster; the N normally operating edge computing nodes are currently in a normal operating state; N is a positive integer;

[0028] A sorting subunit is used to obtain node idle computing resource information corresponding to each of the N normally operating edge computing nodes;

[0029] The sorting subunit is further used to sort the idle computing resource information of N nodes in order of size to obtain a node idle computing resource information sequence;

[0030] The dormant node acquisition subunit is used to sequentially acquire the idle computing power resource information of the target node in the node idle computing power resource information sequence according to the number of dormant nodes, and determine the normally operating edge computing node corresponding to the idle computing power resource information of the target node as the computing node to be dormant.

[0031] In one embodiment, the sleep processing module includes:

[0032] A freezing processing unit, used to freeze the running state of the edge computing node to be dormant within a target time period to obtain a frozen computing node;

[0033] A quantity statistics unit, used to count the online quantity of node online business objects corresponding to the frozen computing nodes;

[0034] A data migration unit is used to obtain the operating data of the frozen computing node when the number of online nodes is less than the online threshold, and migrate the operating data to the target edge computing node; the operating state of the target edge computing node in the target time period is a normal operating state;

[0035] The sleep processing unit is used to freeze the computing node for sleep processing when the running data is successfully migrated to the target edge computing node.

[0036] In one embodiment, the wake-up processing module includes:

[0037] A wake-up quantity determination unit is used to determine the predicted excess computing power resource information of the edge computing node cluster within the target time period according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node wake-up condition;

[0038] The wake-up quantity determination unit is further used to determine the node wake-up quantity according to the predicted over-limit computing power resource information and the unit computing power resource information; the unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster;

[0039] The awakening node acquisition unit is used to acquire a dormant edge computing node in the edge computing node cluster; the dormant edge computing node is an edge computing node in the edge computing node cluster that is currently in a dormant state;

[0040] The awakening node acquisition unit is further used to acquire the edge computing nodes to be awakened from the dormant edge computing nodes according to the number of awakened nodes.

[0041] In one embodiment, the edge computing-based data processing device further includes:

[0042] A change trend determination module is used to determine the change trend of computing power resources corresponding to the edge computing node cluster according to the total change computing power resource information;

[0043] A condition determination module is used to determine the node processing conditions satisfied by the edge computing node cluster according to the total idle computing resource information and the total changed computing resource information if the computing resource change trend is an increasing trend; the node processing conditions include node wake-up conditions and node sleep conditions;

[0044] The condition determination module is also used to determine whether the edge computing node cluster meets the node sleep condition if the computing power resource change trend is a reduction trend.

[0045] In one embodiment, the condition determination module includes:

[0046] A resource comparison unit, used to compare the total idle computing power resource information with the total changed computing power resource information;

[0047] A condition determination unit, configured to determine that the edge computing node cluster meets the node sleep condition if the total idle computing power resource information is greater than the total changed computing power resource information;

[0048] The condition determination unit is also used to determine whether the edge computing node cluster meets the node wake-up condition if the total idle computing power resource information is less than the total changed computing power resource information.

[0049] In one embodiment, the edge computing node cluster includes at least two edge computing nodes; the at least two edge computing nodes include edge computing node M i ; i is a positive integer;

[0050] The idle resource acquisition module includes:

[0051] The occupied resource determination unit is used to obtain the target application in the edge computing node M i Q pieces of current operation information; one piece of current operation information includes an application activity scenario where an online business object is located, and information about the resources occupied by the business object in the application activity scenario;

[0052] The occupied resource determination unit is further used to obtain the occupied resource information of the business object corresponding to each current operation information, and obtain the occupied resource information of Q business objects;

[0053] The occupied resource determination unit is further used to determine the sum of the occupied resource information of Q business objects as the target application for the edge computing node M i The occupied node computing resource information;

[0054] The idle resource determination unit is used to determine the total idle computing power resource information of the edge computing node cluster according to the occupied node computing power resource information corresponding to the at least two edge computing nodes respectively when determining the occupied node computing power resource information of each edge computing node of the target application for at least two edge computing nodes.

[0055] In one embodiment, the idle resource determination unit is further specifically used to obtain the maximum computing power resource information corresponding to each edge computing node in the at least two edge computing nodes, and obtain at least two maximum computing power resource information;

[0056] The idle resource determination unit is further specifically used to determine the total computing power resource information corresponding to at least two maximum computing power resource information, and determine the total occupied computing power resource information corresponding to at least two occupied node computing power resource information;

[0057] The idle resource determination unit is also specifically used to determine the absolute value of the resource difference between the total computing power resource information and the total occupied computing power resource information as the total idle computing power resource information of the edge computing node cluster.

[0058] In one embodiment, the edge computing node cluster includes at least two edge computing nodes; the at least two edge computing nodes include edge computing node M i ; i is a positive integer;

[0059] The change resource prediction module includes:

[0060] The object prediction unit is used to predict the target application for the edge computing node M within the target time period. i New business objects, offline business objects, and scene switching business objects;

[0061] The change resource determination unit is used to determine the target application within the target time period for the edge computing node M according to the number of new business objects added, the number of offline business objects offline, and the number of switching business objects switched by the scene. i Node change computing power resource information;

[0062] The change resource determination unit is also used to determine the sum of the node change computing power resource information corresponding to at least two edge computing nodes as the total change computing power resource information of the edge computing node cluster when determining the node change computing power resource information of each edge computing node of the target application for at least two edge computing nodes.

[0063] In one embodiment, the target application includes one or more application activity scenarios;

[0064] The change resource determination unit is further specifically used to predict the initial login scenario corresponding to the newly added business object; the one or more application activity scenarios include the initial login scenario;

[0065] The change resource determination unit is further specifically used to determine the application activity scenario in which the offline business object is located when the offline behavior occurs as the offline application activity scenario;

[0066] The change resource determination unit is further specifically used to determine the application activity scene in which the node scene switching object is located before the scene switching is performed as the initial application activity scene, and determine the application activity scene in which the scene switching object is located after the scene switching is performed as the target application activity scene;

[0067] The change resource determination unit is further specifically used to obtain the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, and determine the target application within the target time period for the edge computing node M according to the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, as well as the switching quantity, the offline quantity, and the new quantity. i The node changes computing power resource information.

[0068] In one embodiment, the change resource determination unit is further specifically used to determine the first change computing power resource information corresponding to the newly added business object of the node according to the newly added quantity and the average computing power requirement information corresponding to the initial login scenario;

[0069] The change resource determination unit is further specifically used to determine the second change computing power resource information corresponding to the node offline business object according to the offline number and the average computing power demand information corresponding to the online application activity scenario;

[0070] The change resource determination unit is further specifically used to determine the third change computing power resource information corresponding to the node scene switching object according to the average computing power requirement information corresponding to the initial application activity scene, the average computing power requirement information corresponding to the target application activity scene, and the switching quantity;

[0071] The change resource determination unit is further specifically used to determine the target application within the target time period according to the first change computing power resource information, the second change computing power resource information and the third change computing power resource information, for the edge computing node M i The node changes computing power resource information.

[0072] An embodiment of the present application provides a computer device, including: a processor and a memory;

[0073] The memory stores a computer program, and when the computer program is executed by the processor, the processor executes the method in the embodiment of the present application.

[0074] On the one hand, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the method in the embodiment of the present application is executed.

[0075] In one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method provided in one aspect of the embodiments of the present application.

[0076] In an embodiment of the present application, after determining the current total idle computing power resource information of the edge computing node cluster and the total change computing power resource information for the edge computing node cluster in the target time period of the target application, the edge computing node to be dormant can be determined based on the total idle computing power resource information and the total change computing power resource information when the edge computing node meets the node sleep condition, and the dormant edge computing node to be dormant can be processed in the target time period; at the same time, the present application can determine the edge computing node to be awakened based on the total idle computing power resource information and the total change computing power resource information when the edge computing node meets the node wake-up condition, and the awakened edge computing node to be awakened can be processed in the target time period. It should be understood that the total change computing power resource information is the computing power demand change information of the target application for the edge computing node cluster in the target time period, and the dormancy or wake-up processing is performed in the target time period according to the total change computing power resource information, that is, dynamically dormant or awakened some edge computing nodes in the edge computing node cluster according to the real-time computing power demand change information. This not only allows the computing power resources provided by the edge computing node cluster to meet the computing power demand change information, but also allows the edge computing nodes in the edge computing node cluster to sleep according to the real-time computing power demand, reducing the number of running nodes, that is, reducing the running time, thereby reducing the redundant power consumption loss of the edge computing nodes, greatly reducing the failure rate of the nodes, and then greatly reducing the operating costs. In summary, the present application can dynamically sleep or wake up the edge computing nodes according to the real-time total change computing power resource information, so that the edge computing nodes can reduce the operating costs while meeting the computing power demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0078] Figure 1 This is a schematic diagram of a network architecture provided by an embodiment of the present application;

[0079] Figure 2a-2b This is a schematic diagram of a scenario in which node processing is performed on an edge computing node, provided in an embodiment of the present application;

[0080] Figure 3 It is a flowchart of a data processing method based on edge computing provided in an embodiment of the present application;

[0081] Figure 4It is a schematic diagram of a process for determining total change computing power resource information of an edge computing node cluster provided by an embodiment of the present application;

[0082] Figure 5 It is a flow chart of a method for determining node processing conditions satisfied by an edge computing node cluster provided in an embodiment of the present application;

[0083] Figure 6 It is a logic flow chart for performing node processing provided by an embodiment of the present application;

[0084] Figure 7 This is a schematic diagram of a process for migrating operation data provided by an embodiment of the present application;

[0085] Figure 8 It is a system architecture diagram provided by an embodiment of the present application;

[0086] Fig. 9 It is a structural schematic diagram of a data processing device based on edge computing provided in an embodiment of the present application;

[0087] Fig.10 It is a structural diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0088] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0089] This application involves technologies such as cloud computing, cloud gaming, and edge computing. The following will first explain the relevant concepts such as cloud computing, cloud gaming, and edge computing.

[0090] Cloud computing refers to the delivery and use model of IT infrastructure, which means obtaining required resources through the network in an on-demand and easily scalable manner; in a broad sense, cloud computing refers to the delivery and use model of services, which means obtaining required services through the network in an on-demand and easily scalable manner. This service can be related to IT and software, the Internet, or other services. Cloud computing is the product of the development and integration of traditional computer and network technologies such as grid computing, distributed computing, parallel computing, utility computing, network storage technologies, virtualization, and load balancing.

[0091] With the development of the Internet, real-time data streams, and the diversification of connected devices, as well as the demand for search services, social networks, mobile commerce, and open collaboration, cloud computing has developed rapidly. Different from the previous parallel distributed computing, the emergence of cloud computing will promote revolutionary changes in the entire Internet model and enterprise management model from a conceptual perspective.

[0092] Cloud gaming, also known as gaming on demand, is an online gaming technology based on cloud computing technology. Cloud gaming technology enables thin clients with relatively limited graphics processing and data computing capabilities to run high-quality games. In a cloud gaming scenario, the game is not played on the player's game terminal, but on a cloud server, which renders the game scene into a video and audio stream and transmits it to the player's game terminal over the network. The player's game terminal does not need to have powerful graphics computing and data processing capabilities, but only needs to have basic streaming media playback capabilities and the ability to obtain player input commands and send them to the cloud server.

[0093] Edge computing refers to the use of an open platform that integrates network, computing, storage, and application core capabilities to provide the nearest end service at the side close to the object or data source. Its application is initiated at the edge side, resulting in faster network service response, meeting basic needs in terms of real-time business, application intelligence, security, and privacy protection.

[0094] Cloud gaming edge computing nodes are nodes used for edge computing, and are generally composed of multiple servers with graphics processing unit (GPU) computing capabilities. A single server can be called a computing node.

[0095] Computing power, as the name implies, refers to the computing power of a device, ranging from mobile phones and computers to supercomputers. Computing power exists in various hardware devices. Computing power resources are the hardware or network resources required by a device to perform computing tasks, which can usually include central processing unit (CPU) computing power resources, GPU computing power resources, memory resources, network bandwidth resources, and disk resources.

[0096] The solution provided in the embodiments of the present application relates to cloud computing and cloud gaming technology in the field of cloud technology, and the specific process is explained through the following embodiments.

[0097] See also Figure 1 , Figure 1 Schematic diagram of a network architecture provided by an embodiment of the present application. Figure 1 As shown, the network architecture may include a management server 100 and edge nodes 11, 12, ..., and 1n, wherein the edge node 11 may include multiple computing servers such as computing servers 11a and 11b, and the edge node 12 may include multiple computing servers such as computing servers 12a and 12b. Figure 1 As shown, computing servers 11a, 11b and other computing servers in edge node 11 can communicate with each other, computing servers 12a, 12b and other computing servers in edge node 12 can communicate with each other, and any computing server in edge node 11, any computing server in edge node 12, ..., any computing server in edge node 1n can respectively connect to the management server 100 through the network, so that each computing server can exchange data with the management server 100 through the network connection, so that each computing server can receive management data from the management server 100. It can be understood that computing servers in edge nodes are usually deployed in the same area, while different edge nodes are usually deployed in different areas.

[0098] like Figure 1As shown, the computing servers in the above edge nodes can all correspond to the terminal device cluster, and each terminal device in the terminal device cluster can be integrated and installed with a target application. When the target application runs in each terminal device, data can be exchanged with the computing server assigned to it by the management server 100. Among them, the target application can include one or more applications in applications such as game applications, video editing applications, social applications, instant messaging applications, live broadcast applications, short video applications, video applications, music applications, shopping applications, novel applications, payment applications, browsers, etc. that have the function of displaying text, images, audio, and video data information. The computing server provides corresponding functional services for the target application running in the terminal device, but at the same time consumes corresponding computing resources. The computing resources of a computing server can correspond to different terminal devices at the same time. When the terminal device connected to the computing server runs a target application (for example, a cloud game application), the target application will occupy the computing resources of the computing server.

[0099] As can be seen from the above, each edge node (such as edge node 11) may include multiple computing servers (such as edge node 11 may include computing server 11a, computing server 11b, etc.), each computing server may also be referred to as an edge computing node, and the present application may refer to an edge node (such as edge node 11) including multiple computing servers as an edge computing node cluster. Hereinafter, a computing server (such as computing server 11a) is referred to as an edge computing node, and an edge node (such as edge node 11) where a computing server is located is referred to as an edge computing node cluster.

[0100] An edge computing node cluster can provide functional services (provide computing resources) for a target application. In order for the objects using the target application to run the target application smoothly, the edge computing node cluster will usually prepare computing resources for the target application according to the maximum number of online users of the target application (the number of objects logged into the target application) (each edge computing node in the edge computing node cluster will prepare computing resources according to the maximum number of online users). For example, taking the edge computing node cluster as edge node 11 as an example, for the computing server 11a in the edge node 11, the preset maximum number of online users of the target application is 500, then the operating frequency of the computing server 11a (that is, the frequency of the computing module (such as CPU, GPU, etc.) in the computing server 11a when it is running) can be set according to the maximum number of online users 500, and computing resources can be prepared according to the operating frequency. Among them, the operating frequency may refer to the frequency of the computing module (such as CPU, GPU, etc.) in the computing server 11a when it is running; when the preset maximum number of online objects of the target application is 500, then the operating frequency set by the computing server 11a for the target application is the frequency that can provide functional services for 500 online objects, and the prepared computing power resources are the computing power resources corresponding to the operating frequency (the computing power resources corresponding to the operating frequency can be called the maximum computing power resource information of the computing server 11a). Similarly, each other computing server (edge ​​computing node) in the edge node 11 can also prepare computing power resources according to the maximum number of online objects of each object in the same way as the computing server 11a prepares computing power resources. As a result, the entire edge node 11 will have a total maximum number of online objects, and will also have a total computing power resource prepared for the target application (that is, the sum of the maximum computing power resource information of all edge computing nodes in the edge computing node cluster can be called the total computing power resource information).

[0101] However, if within a period of time, the actual number of online objects of the target application is far less than the preset maximum online number, then the total computing power resource information prepared by the edge computing node cluster is actually too large, and the actual required computing power resources may be smaller. At this time, each edge computing node in the edge computing node cluster will still continue to run at an excessively high operating frequency, which will cause a waste of computing power resources of each edge computing node, and will also cause excessive power consumption of the edge computing nodes. Continuous operation will also greatly damage the edge computing nodes, causing failures and affecting the life of the edge computing nodes, thereby increasing operating costs. Similarly, if within a period of time, the actual number of online objects of the target application is far higher than the preset maximum online number, then the total computing power resource information prepared by the edge computing node cluster is actually too small. At this time, the functional services that the edge computing node cluster can provide may not be sufficient to support the operation of the target application and cannot meet the target application's demand for computing power resources.

[0102] In order to enable the edge computing node cluster to better meet the target application's demand for computing resources while reducing the operating cost of the edge computing node cluster, the management server 100 will interact with each edge computing node to obtain the current operating frequency of each edge computing node, so as to obtain the computing resource information corresponding to the current operating frequency (i.e., the maximum computing resource information); the management server 100 can also obtain the occupied node computing resource information of each edge computing node (i.e., the computing resources of each edge computing node occupied by the target application when it is running, such as the CPU computing resources of the computing server 11a occupied by the target application when running the target application), then The management server 100 can obtain the current idle computing resource information of each edge computing node (i.e., the computing resources of each edge computing node that are not occupied by the target application during the current operation, such as the CPU computing resources that are not used in the computing server 11a when providing functional services for the target application) through the maximum computing resource information of each edge computing node and the occupied node computing resource information; after obtaining the current idle computing resource information of each edge computing node, the total idle computing resource information of the edge computing node cluster where the edge computing node is located can also be determined (i.e., the sum of the current idle computing resource information of all edge computing nodes in the edge computing node cluster). Among them, the current may refer to the moment when the occupied node computing resource information of each edge computing node is counted.

[0103] It should be understood that after obtaining the occupied node computing power resource information of each edge computing node, the total occupied computing power resource information of the edge computing node cluster can also be obtained (i.e., the sum of the occupied node computing power resource information of all edge computing nodes in the edge computing node cluster). Further, the management server 100 can predict the total change computing power resource information of the target application for the edge computing node cluster within the target time period (i.e., the computing power resources that the target application is expected to add or release for the edge computing node cluster compared with the total occupied computing power resource information within the target time period, such as the CPU computing power resources of the edge computing node cluster that the target application may add when the target application is running within the target time period); subsequently, the management server 100 can determine the computing power resource change trend corresponding to the edge computing node cluster based on the predicted total change computing power resource information. Among them, the computing power resource change trend may refer to whether the computing power resources required by the target application during the target time period have increased, decreased, or remained unchanged with the total occupied computing power resource information. The computing power resource change trend may include an increasing trend (corresponding to an increase in the required computing power resources), a decreasing trend (corresponding to a decrease in the required computing power resources), and a stable trend (corresponding to the required computing power resources remaining unchanged with the total occupied computing power resource information). The total change computing power resource information can be compared with the value of 0. If the total change computing power resource information is greater than 0 (that is, the total change computing power resource information is a positive value), it can be determined that within the target time period, the computing power resources of the edge computing node cluster that the target application is expected to occupy will increase, and the computing power resource change trend corresponding to the edge computing node cluster can be an increasing trend; if the total change computing power resource information is less than 0 (that is, the total change computing power resource information is a negative value), it can be determined that within the target time period, the computing power resources of the edge computing node cluster that the target application is expected to occupy will decrease, and the computing power resource change trend corresponding to the edge computing node cluster can be a reducing trend; if the total change computing power resource information is equal to 0, it can be determined that within the target time period, the computing power resources of the edge computing node cluster that the target application is expected to occupy will not change, and the computing power resource change trend corresponding to the edge computing node cluster can be a stable trend.

[0104] When the computing power resource change trend of the edge computing node cluster is an increasing change trend, it can be determined according to the total idle computing power resource information and the total changed computing power resource information whether the node processing condition satisfied by the edge computing node cluster is a node wake-up condition (e.g., the current total idle computing power resource information of the edge computing node cluster cannot satisfy the total changed computing power resource information, then the edge computing node cluster satisfies the node wake-up condition at this time) or a node sleep condition (e.g., the current total idle computing power resource information of the edge computing node cluster is sufficient, and there is remaining idle computing power resource information while satisfying the total changed computing power resource information, then the edge computing node cluster satisfies the node sleep condition at this time). When the node processing condition of the edge computing node cluster is When the node processing condition of the edge computing node cluster is a node sleep condition, the management server 100 can determine the edge computing nodes to be awakened in the edge computing node cluster (that is, the edge computing nodes that will no longer run within the target time period, and these edge computing nodes can be set to sleep from the normal operating state within the target time period), and perform sleep processing on the edge computing nodes to be awakened within the target time period; and when the node processing condition of the edge computing node cluster is a node wake-up condition, the management server 100 can determine the edge computing nodes to be awakened in the edge computing node cluster (that is, the edge computing nodes that need to be converted from the sleep state to the normal operating state within the target time period), and perform wake-up processing on the edge computing nodes to be awakened within the target time period.

[0105] When the computing power resource change trend of the edge computing node cluster is a decreasing trend, it can be directly determined that the edge computing node cluster meets the node sleep condition, and then the edge computing nodes to be dormant can be determined based on the total idle computing power resource information and the total changed computing power resource information, and the dormant edge computing nodes to be dormant can be put into dormancy within the target time period. Among them, for the specific implementation method of determining whether the node processing condition satisfied by the edge computing node cluster is the node wake-up condition or the node sleep condition based on the total idle computing power resource information and the total changed computing power resource information, please refer to the subsequent Figure 3 The description in the corresponding embodiment.

[0106] It should be understood that the total change computing power resource information in the target time period is the computing power demand change information of the target application in the target time period. When the computing power demand of the target application in the target time period increases, some edge computing nodes in a dormant state can be awakened so that they can provide functional services for the target application, thereby meeting the new computing power demand of the target application; when the computing power demand of the target application in the target time period decreases, some edge computing nodes in a normal operating state can be dormant so that they can "rest" and no longer continue to run in the target time period, thereby reducing the running time and increasing the lifespan. In other words, this application can dynamically sleep or wake up edge computing nodes according to real-time computing power requirements, thereby reducing operating costs while meeting computing power requirements.

[0107] It is understandable that the above processing process can be executed by the management server alone, by the computing server alone, or by the management server and the computing server together. The specific implementation can be adjusted according to actual needs and is not limited here.

[0108] It is understandable that the method provided in the embodiments of the present application can be executed by a computer device, including but not limited to a terminal device, a computing server or a management server. Among them, the management server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0109] It is understandable that the above-mentioned devices (such as the above-mentioned management server 100, computing server 11a, computing server 11b, computing server 12a, ..., computing server 12b, etc.) can be a node in a distributed system, wherein the distributed system can be a blockchain system, and the blockchain system can be a distributed system formed by connecting the multiple nodes through network communication. Among them, the nodes can form a peer-to-peer (P2P, Peer To Peer) network, and the P2P protocol is an application layer protocol running on the Transmission Control Protocol (TCP, Transmission Control Protocol) protocol. In a distributed system, any form of computer equipment, such as servers, terminal devices and other electronic devices, can become a node in the blockchain system by joining the peer-to-peer network.

[0110] Among them, the terminal devices in the above-mentioned terminal device cluster may include mobile phones, tablet computers, laptops, PDAs, smart speakers, mobile Internet devices (MID), POS (Point Of Sales), wearable devices (such as smart watches, smart bracelets, etc.), vehicle-mounted devices, etc.

[0111] For easier understanding, see Figure 2a-2b , Figure 2a-2b Schematic diagram of a scenario of performing node processing on an edge computing node provided by an embodiment of the present application. Figure 2a-2bThe scenario shown is described by taking the above-mentioned management server 100 performing node processing on computing servers 11a, 11b, 11c, and 11d in edge node 11 (i.e., edge computing node cluster) within the target time period as an example. Hereinafter, edge node 11 is referred to as edge computing node cluster 11, and computing servers 11a, 11b, 11c, and 11d are referred to as edge computing nodes.

[0112] like Figure 2a As shown, business object A1, business object A2, business object A3, business object A4, business object A5, business object A6, business object A7, business object A8, business object A9, business object A10, business object A11, and business object A12 are all online business objects; wherein, the business object may refer to the bound account of the business user who uses the terminal device to run the target application in the target application, and the business user can use the bound account to log in to the target application, and the target application can also determine whether the business user is logged in through the bound account. When the business user uses his bound account to log in to the target application, the business object corresponding to the business user (i.e., the bound account) is in an online state, and at this time, these business objects can be called online business objects. In other words, the above-mentioned business objects A1, business objects A2, ..., and business objects A12 are all currently in an online state.

[0113] like Figure 2a The terminal device 10a shown is the terminal corresponding to the business object A1, the terminal device 10b is the terminal corresponding to the business object A2, the terminal device 10c is the terminal corresponding to the business object A3, the terminal device 10d is the terminal corresponding to the business object A4, the terminal device 10e is the terminal corresponding to the business object A5, the terminal device 10f is the terminal corresponding to the business object A6, the terminal device 10g is the terminal corresponding to the business object A7, the terminal device 10h is the terminal corresponding to the business object A8, the terminal device 10i is the terminal corresponding to the business object A9, the terminal device 10j is the terminal corresponding to the business object A10, the terminal device 10k is the terminal corresponding to the business object A11, and the terminal device 10m is the terminal corresponding to the business object A12. At this time, edge computing node 11a provides functional services (provides computing power resources) for terminal devices 10a, 10b, and 10c, edge computing node 11b provides functional services (provides computing power resources) for terminal devices 10d, 10e, and 10f, edge computing node 11c provides functional services (provides computing power resources) for terminal devices 10g, 10h, and 10i, and edge computing node 11d provides functional services (provides computing power resources) for terminal devices 10j, 10k, and 10m.

[0114] The management server 100 can obtain the current idle computing resource information of the edge computing node 11a, the edge computing node 11b, the edge computing node 11c, and the edge computing node 11d (i.e., the computing resources of each edge computing node that are not occupied when the target application is currently running). The following will take the example of the management server 100a obtaining the idle computing resource information of the edge computing node 11a to explain its specific process.

[0115] The management server 100 can obtain the computing power resources of the edge computing node 11a occupied when the target application is run on the terminal device 10a, the terminal device 10b, and the terminal device 10c (hereinafter referred to as the occupied node computing power resource information), wherein the occupied node computing power resource information may refer to the quantitative index information of the computing power resources occupied when the edge computing node 11a provides functional services for the target application. Among them, the quantitative index information may include one or more index information of multiple index information such as central processing unit computing power information (CPU), graphics processing unit computing power information (GPU), memory usage information, network bandwidth usage information, disk read and write capability information, etc. Here, the quantitative index information including CPU computing power and GPU computing power will be taken as an example, and the occupied computing power resource information is the statistics of the CPU computing power resources and GPU computing power resources of the edge computing node 11a occupied when running the target application.

[0116] Subsequently, the management server 100 can obtain the current operating frequency of the edge computing node 11a, so as to determine the current maximum computing power resource information of the edge computing node 11a. According to the maximum computing power resource information and the occupied computing power resource information of the edge computing node 11a, the idle computing power resource information of the edge computing node can be determined (e.g., using the maximum computing power resource information minus the occupied computing power resource information). Similarly, the management server 100 can also determine the idle computing power resource information corresponding to the edge computing node 11b, the edge computing node 11c, and the edge computing node 11d. Subsequently, the management server 100 can determine the total idle computing power resource information corresponding to the edge computing node cluster 11 (e.g., adding the idle computing power resource information corresponding to the edge computing node 11a, the edge computing node 11b, the edge computing node 11c, and the edge computing node 11d).

[0117] Furthermore, the management server 100 can predict the total change computing power resource information of the target application for the edge computing node cluster 11 within the target time period, and the total change computing power resource information is the computing power resources that the target application is expected to increase (or decrease, or remain unchanged) during the target time period. In other words, the total change computing power resource information can be used to characterize whether the computing power resources that the target application is expected to occupy will increase, decrease, or remain unchanged during the target time period compared to the current total occupied computing power resource information of the edge computing node cluster 11. A specific method for the management server 100 to determine the total change computing power resource information of the edge computing node cluster 11 may be: the management server 100 first determines the node change computing power resource information of each edge computing node in the edge computing node cluster 11 (including edge computing node 11a, edge computing node 11b, edge computing node 11c and edge computing node 11d) (that is, the target application is expected to increase (or decrease, or remain unchanged) the computing power resources occupied by a certain edge computing node during the target time period, that is, the computing power demand change information for a certain edge computing node), and then the sum of the node change computing power resources of all edge computing nodes can be determined, and this sum can be used as the total change computing power resource information of the edge computing node cluster 11. For the specific implementation method of determining the node change computing power resource information of each edge computing node, please refer to the subsequent Figure 3 The description in the corresponding embodiment.

[0118] like Figure 2b As shown, the node change computing power resource information of the edge computing node 11a determined by the management server 100 is the node change computing power resource information a, the node change computing power resource information of the edge computing node 11b is the node change computing power resource information b, the node change computing power resource information of the edge computing node 11c is the node change computing power resource information c, and the node change computing power resource information of the edge computing node 11d is the node change computing power resource information d. Subsequently, the management server 100 can add the node change computing power resource information a, the node change computing power resource information b, the node change computing power resource information c, and the node change computing power resource information d, so as to obtain the total change computing power resource information corresponding to the edge computing node cluster 11.

[0119] Further, according to the total change computing power resource information, the computing power resource change trend of the edge computing node cluster 11 can be determined. Among them, the computing power resource change trend is compared with the current total occupied computing power resource information, for the edge computing node cluster 11, whether the total computing power resources required by the target application in the target time period will be added, released (reduced), or the total occupied computing power resource information will remain unchanged. When the total change computing power resource information is greater than the value 0 (that is, the total change computing power resource information is a positive value), it can be explained that the computing power resources required by the target application in the target time period will be added, and the computing power resource change trend is an increasing change trend; when the total change computing power resource information is less than the value 0 (that is, the total change computing power resource information is a negative value), it can be explained that the computing power resources required by the target application in the target time period will be reduced, and the computing power resource change trend is a reducing change trend; when the total change computing power resource information is equal to the value 0, it can be explained that the computing power resources required by the target application in the target time period have not changed, and the computing power resource change trend is a stable change trend.

[0120] It should be understood that according to the computing power resource change trend, the total change computing power resource information and the total idle computing power resource information, the edge computing nodes in the edge computing node cluster 11 can be processed (such as node sleep processing, node wake-up processing). For example, when the computing power resource change trend is an increasing trend, it indicates that the computing power resources required by the target application will be increased within the target time period. At this time, the total idle computing power resource information of the edge computing node cluster 11 can be compared with the total change computing power resource information, thereby judging whether the total idle computing power resource information can meet the newly added computing power resources required by the target application. If the total idle computing power resources are greater than the total change computing power resource information, then the total idle computing power resources can meet the newly added computing power resources required by the target application, and after the total change computing power resource information is met, there is still excess idle computing power resource information, then some edge computing nodes in the edge computing node cluster 11 can be put into sleep mode (i.e., the functions of these edge computing nodes are turned off, and these edge computing nodes will not run within the target time period), thereby reducing the running time of the edge computing nodes, improving their lifespan, and saving costs; and if the total idle computing power resources are greater than the total change computing power resource information, then the total idle computing power resource information cannot meet the needs of the target application, then some edge computing nodes in the edge computing node cluster 11 that are in a dormant state can be awakened to run normally to provide functional services for the target application and meet the needs of the target application. It should be noted that if there are no dormant edge computing nodes in the edge computing node cluster 11 at this time, new computing servers can be launched, and these new computing servers can be added to the edge computing node cluster 11 to provide functional services for the target application. If the total idle computing power resources are equal to the total changed computing power resource information, there is no need to perform sleep or wake-up processing on the edge computing node cluster 11.

[0121] The above is only an example of the growth trend of computing power resource changes to illustrate the node processing method of the edge computing node cluster 11. The computing power resource change trend can also include a reduction trend and a stable trend. For the specific method of performing node processing on the edge computing node cluster 11 according to the specific computing power resource change trend, the total change computing power resource information and the total idle computing power resource information, please refer to the subsequent Figure 3 The description in the corresponding embodiment.

[0122] For further information, see Figure 3 , Figure 3 is a flow chart of a data processing method based on edge computing provided by an embodiment of the present application. Figure 1 The computer device (such as the management server 100) in the corresponding embodiment executes, that is, it can be executed by Figure 1 The management server 100 in is executed.

[0123] like Figure 3 As shown, the data processing method based on edge computing may include the following steps S101 to S104:

[0124] Step S101, obtaining total idle computing power resource information of the edge computing node cluster; the total idle computing power resource information is the computing power resource information of the edge computing node cluster that is not occupied by the target application when the target application is running.

[0125] In this application, edge computing nodes may refer to computer devices that can provide computing or application services, such as servers (for example, the above Figure 1 The computing servers 11a, 11b, 12a, or 12b shown in the figure). Multiple edge computing nodes can form an edge computing node cluster, and the edge computing node cluster can be as described above. Figure 1 The edge nodes 11 and 12 shown in the figure are shown in the figure. The target application may refer to an application for which the edge computing node needs to complete related computing tasks. For example, the target application may be a cloud gaming application. Based on cloud computing technology, cloud gaming is usually run on a remote server. The terminal device only needs to receive the audio and video stream sent by the remote server, and then decode and play it. At this time, the remote server can be an edge computing node.

[0126] It should be understood that when the target application runs on the terminal device, each edge computing node in the edge computing node cluster can provide corresponding computing services for it, and when the edge computing node provides corresponding computing services for it, the target application will occupy the computing power resources of the edge computing node. Among them, computing power can refer to the computing power of the edge computing node. In this application, the measurement of the computing power of the edge computing node can usually be measured by CPU computing power and GPU computing power. Among them, CPU computing power is generally measured by the number of operations per second (OPS); and GPU computing power can have a variety of measurement indicators according to the type of calculation, generally measured from two indicators of computing power (according to the type of operation, the number of floating-point operations per second (FLOPS), OPS, half-precision peak computing power and double-precision peak computing power) and data read throughput. For computing power resources, in addition to CPU computing power resources and CPU computing power resources, of course, other computing power resources can also be included, such as memory resources, network bandwidth resources, disk resources, etc. This application does not limit the content included in computing power resources. The following will take computing resources including CPU computing resources and GPU computing resources as an example for explanation.

[0127] Specifically, the target application may include one or more application activity scenarios. The application activity scenario may refer to the scenario type to which the screen displayed on the terminal device belongs when the target application is running on the terminal device. An application activity scenario may provide corresponding application functions for the target application. For example, when the target application is a social application, the application activity scenario may include a voice call scenario, a video call scenario, and a text chat scenario. When a business user logs in to the social application through the bound account of the social application (which may be called a business object), in the voice call scenario, the social application provides a voice communication function between different business objects (i.e., business users can communicate with each other by voice); in the video call scenario, the social application provides a video communication function between different business objects (i.e., business users can communicate with each other by video); in the text chat scenario, the social application provides a text communication function between different business objects (i.e., business users can communicate with each other by text). For example, when the target application is a cloud gaming application, the application activity scenarios may include a game homepage lobby scenario (i.e., when a business user logs in to or opens a cloud gaming application, the cloud gaming application usually presents a default screen, which can be used for game character presentation, character costume change, competitive mode selection, etc. The default presentation screen is usually called a game homepage lobby), a single-player competitive scenario (i.e., a single-player combat mode is available in the cloud gaming application, and the single-player competitive scenario is used for single-player skydiving, single-player level-breaking, etc.), and a multiplayer competitive scenario (i.e., a multiplayer combat mode is available in the cloud gaming application, and the multiplayer competitive scenario can be used for multiplayer team-up to fight monsters, improve game levels, multiplayer team-up to break through levels, etc.). In the game homepage lobby scenario, the cloud gaming application provides business with functions such as game background introduction, game character introduction, information display, and competitive mode selection; in the single-player competitive scenario, the cloud gaming application provides business objects with single-player combat competitive functions; in the multiplayer competitive scenario, the cloud gaming application provides business objects with multiplayer combat competitive functions, multiplayer communication functions, etc.

[0128] It is understandable that when the application activity scenarios of the business object in the target application are different, the computing power resources required are usually different. For example, when the target application is a cloud gaming application, the application activity scenarios include the game homepage lobby scene, the single-player competition scene, and the multiplayer competition scene. Usually, in the game homepage lobby scene, it is necessary to present simple pictures and controls such as the game background, the game character introduction of the business object, and the game competition mode selection to the business object. Then the computing power resources required for the game homepage lobby scene will also be relatively small; in the single-player competition scene, the business object can play a single-player competitive game (for example, a single player operates a game character to increase game experience or obtain game coins, etc.), which is different from Compared with the game homepage lobby scene, the business objects in the single-player competition scene will have more operation instructions, and the corresponding calculations will also be more, so more computing power resources will be required in the single-player competition scene; in the multiplayer competition scene, the business objects can perform a large number of game operations (such as sliding the direction wheel, releasing the skills of the game character, and clicking the retreat control). In addition, business objects can also communicate with each other through voice and text. In other words, the multiplayer competition scene needs to provide computing services for communication for business objects in addition to satisfying the game operation instructions of the business objects. Therefore, compared with the single-player competition scene, the multiplayer competition scene requires more computing power resources.

[0129] Therefore, the present application can obtain the current online business objects (when the business user uses his bound account to log in to the target application, the business object corresponding to the business user (i.e., the bound account) is online, and these business objects can be called online business objects) in each edge computing node. The application activity scenario in the target application, and obtain the computing power resources occupied by each online business object in the application activity scenario, thereby statistically obtaining the computing power resources of the edge computing nodes occupied by all online business objects. The computing power resources of the edge computing nodes occupied by all online business objects of an edge computing node can be called the occupied node computing power resource information of an edge computing node. When the occupied node computing power resource information of each edge computing node in the edge computing node cluster is determined, the total occupied computing power resource information of the edge computing node cluster can be determined, and then the total idle computing power resource information of the edge computing node cluster can be determined. Among them, the total idle computing power resource information is the computing power resources of the edge computing node cluster that are not occupied by the target application when running the target application.

[0130] The edge computing node cluster includes at least two edge computing nodes, and the at least two edge computing nodes include edge computing node M i (i is a positive integer) as an example, the specific method for determining the total idle computing power resource information of the edge computing node cluster can be: the target application can be obtained on the edge computing node M iQ (Q is a positive integer) pieces of current operation information in the target application; wherein, one piece of current operation information may include an application activity scenario where an online business object is located, and the business object occupied resource information in the application activity scenario; then, the business object occupied resource information corresponding to each piece of current operation information may be obtained, thereby obtaining Q pieces of business object occupied resource information; then, the sum of the Q pieces of business object occupied resource information may be determined as the target application for the edge computing node M. i When the occupied node computing power resource information of the target application for each of the at least two edge computing nodes is determined, the total idle computing power resource information of the edge computing node cluster can be determined based on the occupied node computing power resource information corresponding to the at least two edge computing nodes.

[0131] Among them, the specific method for determining the total idle computing power resource information of the edge computing node cluster based on the occupied node computing power resource information corresponding to at least two edge computing nodes can be: the maximum computing power resource information corresponding to each edge computing node in the at least two edge computing nodes can be obtained to obtain at least two maximum computing power resource information; then, the total computing power resource information corresponding to the at least two maximum computing power resource information can be determined, and the total occupied computing power resource information corresponding to the at least two occupied node computing power resource information can be determined; the absolute value of the resource difference between the total computing power resource and the total occupied computing power resource information can be determined as the total idle computing power resource information of the edge computing node cluster.

[0132] It should be understood that the above-mentioned business object resource occupation information can be understood as the computing power resources of a certain edge computing node occupied by an online business object when it is in a certain application activity scenario; after each business user logs in to the target application through a bound account (hereinafter referred to as a business object), the login status of its business object in the target application is the online status, and the business object can also be called an online business object at this time. The computer device will obtain the application activity scenario of each online business object in the target application in real time, and count the computing power resources occupied by the online business object when it is in a certain application activity scenario. The computing power resources occupied by an online business object in the current application activity scenario and in the application activity scenario can constitute a current operation information. Because a current operation information includes an application activity scenario where an online business object is located, and the computing power resources occupied when it is in the application activity scenario; then Q current operation information can include the application activity scenarios where Q online business objects are located, and the computing power resources occupied by Q online business objects when they are in the corresponding application activity scenarios. The total computing power resources (i.e., the occupied node computing power resource information) of a certain edge computing node occupied by the Q online business objects are added and summed. When the occupied node computing power resource information of each edge computing node is determined, these occupied node computing power resource information can be added, thereby obtaining the total occupied computing power resource information corresponding to the edge computing node cluster (i.e., the total occupied computing power resource information corresponding to at least two occupied node computing power resource information).

[0133] Furthermore, the current maximum computing power resource information corresponding to each edge computing node can be obtained, and the maximum computing power resource information of all edge computing nodes in the edge computing node cluster can be added together to obtain the total computing power resource information corresponding to the edge computing node cluster (i.e., the total computing power resource information corresponding to at least two maximum computing power resource information). Subsequently, the total computing power resource information of the edge computing node cluster is subtracted from the total occupied computing power resource information of the edge computing node cluster to obtain the total idle computing power resource information corresponding to the edge computing node cluster.

[0134] It should be understood that computing resources may include CPU computing resources and GPU computing resources. Therefore, when counting each edge computing node (such as edge computing node M i ) can count the edge computing nodes M occupied by Q online business objects. i The CPU computing resources and GPU computing resources occupied by the edge computing node M i The occupied node computing resource information.

[0135] Step S102: predict the total change computing power resource information of the edge computing node cluster for the target application within the target time period.

[0136] In this application, the target application includes one or more application activity scenarios, and the computing resources required by the business objects in different application activity scenarios will also be different. Then for an edge computing node, the addition, offline, and scene switching of business objects will also cause changes in the required computing resources. Then this application can predict the new business objects of each edge computing node in the edge computing node cluster (that is, the bound accounts corresponding to the business users who log in or open the target application within the target time period, such as edge computing node M i For example, the newly added business object does not belong to the Q current online business objects mentioned above), the offline business object (that is, the bound account corresponding to the business user who exits the cloud gaming application within the target time period in the current online business object, such as the edge computing node M i For example, offline business objects, i.e., the bound accounts corresponding to the business users who are expected to exit the target application within the target time period among the Q current online business objects mentioned above) and scene switching business objects (i.e., the business objects that will switch the application activity scene within the target time period among the current online business objects of the edge computing node, and the application activity scenes of these scene switching business objects are different from the application activity scenes in the target time period (for example, for cloud gaming applications, an online business object is in a single-player competitive scene at the current moment, and will exit the single-player competitive scene within the target time period and enter the game homepage lobby scene. The online business object can be called a scene switching object)). According to the newly added business objects, offline business objects, and scene switching business objects, the node change computing power resource information of each edge computing node can be determined (i.e., for a certain edge computing node, after the business object is added, offline, or the scene is switched, whether the computing power resources required for the target application are added, reduced, or unchanged). After determining the node change computing power resource information of each edge computing node in the edge computing node cluster, the node change computing power resource information of all edge computing nodes in the edge computing node cluster is added together to determine the total change computing power resource information of the edge computing node cluster.

[0137] For the specific implementation method of determining the total change computing power resource information of the edge computing node cluster, please refer to the subsequent Figure 4 The description in the corresponding embodiment.

[0138] Step S103: If the edge computing node cluster meets the node sleep condition, the edge computing nodes to be sleep are determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, and the edge computing nodes to be sleep are sleep processed within the target time period.

[0139] In the present application, after determining the total idle computing power resource information and the total change computing power resource information of the edge computing node cluster, the computing power resource change trend corresponding to the edge computing node cluster can be determined according to the total change computing power resource information; the computing power resource change trend includes an increasing change trend, a decreasing change trend, and a stable change trend. When the total change computing power resource information is greater than the value 0, the computing power resource change trend can be an increasing change trend; when the total change computing power resource information is less than the value 0, the computing power resource change trend can be a decreasing change trend; when the total change computing power resource information is equal to the value 0, the computing power resource change trend can be a stable change trend.

[0140] Subsequently, based on the total idle computing power resource information, the total changed computing power resource information, and the computing power resource change trend, the node processing conditions corresponding to the edge computing node cluster can be determined, where the node processing conditions can include node wake-up conditions and node sleep conditions. For the specific implementation method of determining the node processing conditions of edge computing nodes, please refer to the subsequent Figure 5 The description in the corresponding embodiment.

[0141] Furthermore, when the edge computing node cluster meets the node sleep condition, the edge computing node cluster can be put into sleep within the target time period. The specific method can be: the edge computing nodes to be put into sleep can be determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information; then, the edge computing nodes to be put into sleep can be put into sleep within the target time period. Among them, the specific method for determining the edge computing nodes to be put into sleep in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information can be: if the edge computing node cluster meets the node sleep condition, the predicted total idle computing power resource information of the edge computing node cluster within the target time period can be determined according to the total idle computing power resource information and the total changed computing power resource information; the number of nodes to be put into sleep can be determined according to the predicted total idle computing power resource information and the unit computing power resource information; wherein, the unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster; the edge computing nodes to be put into sleep can be determined in the edge computing node cluster according to the number of nodes to be put into sleep.

[0142] It should be understood that the unit computing power resource information may refer to the total computing power resource information corresponding to an edge computing node in the edge computing node cluster. In one feasible manner, the total computing power resource information corresponding to an edge computing node may be the average value of the sum of the maximum computing power resource information of all edge computing nodes (all edge computing nodes currently in normal operation) in the edge computing node cluster, that is, the average maximum computing power resource information; in another feasible manner, the total computing power resource information corresponding to an edge computing node may be a manually specified value (e.g., a value determined based on manual experience).

[0143] If the edge computing node cluster meets the node sleep condition, it can be said that the total change computing power resource information is a negative value, and the computing power resources required by the target application in the time period have been reduced. The edge computing node cluster can release some computing power resources in the target time period, and some edge computing nodes in the edge computing node cluster can also be put into sleep mode in the target time period. By subtracting the total change computing power resource information from the total idle computing power resource information (that is, adding the computing power resources to be released to the total idle computing power resource information), the predicted total idle computing power resource information in the target time period can be obtained. By dividing the predicted total idle computing power resource information by the above-mentioned unit computing power resource information, the number of edge computing nodes that can be put into sleep mode (that is, the number of sleeping nodes) can be obtained.

[0144] Furthermore, the edge computing nodes to be dormant can be determined in the edge computing node cluster according to the number of dormant nodes, and the specific method may be: the dormant edge computing nodes in the edge computing node cluster can be obtained; wherein the dormant edge computing nodes are edge computing nodes in the edge computing node cluster that are currently in a dormant state; then, the node dormant polling table corresponding to the edge computing node cluster can be obtained; the node dormant polling table includes the dormant polling order of each edge computing node in the edge computing node cluster; according to the position of the dormant edge computing nodes in the node dormant sorting table and the number of dormant nodes, the edge computing nodes to be dormant can be obtained in sequence from the node dormant polling table; the edge computing nodes to be dormant are currently in normal operation.

[0145] Taking the edge computing node cluster including edge computing node 1, edge computing node 2, edge computing node 3, edge computing node 4, and edge computing node 5, and the node sleep polling table is {edge computing node 1, edge computing node 2, edge computing node 3, edge computing node 4, edge computing node 5} as an example, the edge computing node 1 and edge computing node 2 are dormant edge computing nodes; the number of dormant nodes is 2, then the edge computing nodes to be dormant can be obtained in sequence as edge computing node 3 and edge computing node 4. When the target time period is reached, edge computing node 3 and edge computing node 4 can be put to sleep.

[0146] Optionally, the specific method for determining the edge computing nodes to be dormant in the edge computing node cluster according to the number of dormant nodes can also be: N normally operating edge computing nodes in the edge computing node cluster can be obtained; wherein the N normally operating edge computing nodes are currently in normal operating state; N is a positive integer; the node idle computing power resource information corresponding to each normally operating edge computing node in the N normally operating edge computing nodes is obtained; the N node idle computing power resource information is sorted in order of size to obtain a node idle computing power resource information sequence; the target node idle computing power resource information is obtained in order in the node idle computing power resource information sequence according to the number of dormant nodes, and the normal operating edge computing node corresponding to the target node idle computing power resource information is determined as the computing node to be dormant. In other words, after determining the number of dormant nodes, the to-be-dormant edge computing nodes corresponding to the number of dormant nodes can be obtained in the normal operating edge computing nodes in the order of the node idle computing power resource information from large to small.

[0147] Further, after determining the edge computing nodes to be put into sleep, the edge computing nodes to be put into sleep can be put into sleep within the target time period, and the specific method can be as follows: within the target time period, the operating status of the edge computing nodes to be put into sleep is frozen to obtain frozen computing nodes; then, the online number of online business objects of the nodes corresponding to the frozen computing nodes can be counted; when the online number is less than the online threshold, the operating data of the frozen computing nodes is obtained, and the operating data is migrated to the target edge computing node; the operating status of the target edge computing node within the target time period is a normal operating status; when the operating data is successfully migrated to the target edge computing node, the frozen computing node is put into sleep. Optionally, when the online number is equal to the online threshold, the operating data can also be migrated (i.e., the operating data of the frozen computing node is obtained and the operating data is migrated to the target edge computing node), and after successful migration, the frozen computing node is put into sleep.

[0148] That is to say, within the target time period, the node status of the edge computing node to be dormant can be frozen, and the newly launched business objects will no longer be allocated to the edge computing node to be dormant in the frozen state (i.e., frozen computing node); in this process, you can wait for the online business objects in the edge computing node to be dormant in the frozen state (i.e., frozen computing node) to go offline naturally. After a period of time, most online business objects will go offline naturally after closing the target application, but a small number of online business objects will still be online. At this time, the running data of the edge computing node to be dormant in the frozen state (i.e., frozen computing node) can be migrated to other edge computing nodes that are running normally.

[0149] Step S104: If the edge computing node cluster meets the node wake-up condition, the edge computing node to be awakened is determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, and the edge computing node to be awakened is awakened within the target time period.

[0150] In this application, if the edge computing node cluster meets the node wake-up condition, it can be explained that the total change computing power resource information is a positive value, the computing power resources required by the target application in the time period have increased, and the total idle computing power resource information cannot meet the newly added demand computing power resources. At this time, it is necessary to wake up the dormant edge computing nodes to provide computing power resources for the target application. By subtracting the total change computing power resource information from the total idle computing power resource information (that is, subtracting the newly added demand computing power resources from the total idle computing power resource information), the predicted excess computing power resource information in the target time period can be obtained (which may refer to the newly added computing power resource information that exceeds the total computing power resource information of the edge computing node cluster, that is, the computing power resource information that exceeds the maximum computing power resource that the edge computing node can provide and is also required in addition), and the predicted excess computing power resource information is divided by the above-mentioned unit computing power resource information to obtain the number of edge computing nodes that need to be awakened (that is, the number of node awakenings); according to the number of node awakenings, the edge computing nodes to be awakened can be obtained from the dormant edge computing nodes. The specific method may be as follows: if the edge computing node cluster meets the node wake-up condition, the predicted excess computing power resource information of the edge computing node cluster within the target time period may be determined according to the total idle computing power resource information and the total changed computing power resource information; the number of node wake-ups may be determined according to the predicted excess computing power resource information and the unit computing power resource information; wherein the unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster; subsequently, the dormant edge computing nodes in the edge computing node cluster may be obtained; the dormant edge computing nodes are edge computing nodes in the edge computing node cluster that are currently in a dormant state; according to the number of node wake-ups, the edge computing nodes to be awakened may be obtained from the dormant edge computing nodes.

[0151] It should be understood that, taking the number of node wake-ups as 2 as an example, after determining that the number of node wake-ups is 2, 2 dormant edge computing nodes can be randomly obtained from the dormant edge computing nodes, and they can be awakened within the target time period and added to the edge computing node cluster to provide functional services for the target application. Optionally, if there are no 2 dormant edge computing nodes among the dormant edge computing nodes, then a new edge computing node (computing server that has not been capped) can be added to the edge computing node cluster (i.e., a new computing server is online) to provide functional services for the target application within the target time period.

[0152] In an embodiment of the present application, after determining the current total idle computing power resource information of the edge computing node cluster and the total change computing power resource information for the edge computing node cluster of the target application within the target time period, it can be determined whether the edge computing node cluster meets the node sleep condition or the node wake-up condition based on the total idle computing power resource information and the total change computing power resource information. When the edge computing node meets the node wake-up condition, the awakened edge computing node can be awakened within the target time period; when the node sleep condition is met, the dormant edge computing node can be put into sleep within the target time period. The total change computing power resource information is the computing power demand change information of the target application for the edge computing node cluster within the target time period. The sleep or wake-up processing is performed within the target time period according to the total change computing power resource information, that is, some edge computing nodes in the edge computing node cluster are dynamically put into sleep or wake-up according to the real-time computing power demand change information. This not only allows the computing power resources provided by the edge computing node cluster to meet the computing power demand change information, but also allows the edge computing nodes in the edge computing node cluster to sleep according to the real-time computing power demand, reducing the number of running nodes, that is, reducing the running time, thereby reducing the redundant power consumption loss of the edge computing nodes, greatly reducing the failure rate of the nodes, and then greatly reducing the operating costs. In summary, the present application can dynamically sleep or wake up the edge computing nodes according to the real-time total change computing power resource information, so that the edge computing nodes can reduce the operating costs while meeting the computing power demand.

[0153] For further information, see Figure 4 , Figure 4 This is a schematic diagram of a process for determining the total change computing power resource information of an edge computing node cluster provided by an embodiment of the present application. The process may correspond to the above Figure 3 The process of determining the total change computing power resource information in step S102 in the corresponding embodiment is based on the edge computing node cluster including at least two edge computing nodes, and the at least two edge computing nodes including edge computing node M i (i is a positive integer) as an example. Figure 4 As shown, the process may include at least the following steps S401 to S403:

[0154] Step S401: predict the target application for edge computing node M within the target time period. i New business objects, offline business objects, and scene switching business objects.

[0155] Specifically, the computer device will obtain the target application for the edge computing node M iThe historical business behavior data of the target application will be obtained, and then the operation activity information of the target application in the target time period will be obtained; according to the operation activity information and historical business behavior data in the target time period, the edge computing node M can be determined i New business objects for the target application within the target time period; according to the application activity scenario where the business object is located, the operation activity information within the target time period, and the historical business behavior data, the edge computing node M can be determined. i Offline business objects and scene switching business objects for the target application within the target time period. The historical business behavior data may include the relevant behavior data of the historical online business objects, historical online business objects, historical offline business objects, and historical scene switching business objects of the target application at each time node within the historical time period, etc. The relevant behavior data may include the application activity scene, operation behavior, application running time, etc.

[0156] It should be understood that the operational activity information within the target time period may refer to special activities launched in the target application during holidays (such as May Day, Chinese Valentine's Day, Dragon Boat Festival), specific holidays, version release dates and other time periods. For example, taking the target application as a cloud gaming application, during the Dragon Boat Festival, new Dragon Boat Festival limited-time activities (such as game character racing activities) and character costume limited-time purchase activities will be launched in the cloud gaming application. Business objects are likely to choose to log in to the cloud gaming application within the target time period based on these operational activity information to participate in this Dragon Boat Festival limited-time activity, purchase character costumes, etc. These business objects that did not log in to the cloud gaming application when counting the occupied node computing power resource information (that is, they do not belong to the current online business objects of the edge computing node) but are predicted to log in to the cloud gaming application within the target time period can be called new business objects. At the same time, the computer equipment can predict, based on the operational activity information and historical business behavior data, which business objects running the target application will be shut down within the target time period among the current online business objects of the edge computing node. These business objects can be called offline business objects. The computer equipment can also predict, based on the operational activity information and historical business behavior data, which business objects will switch scenes among the current online business objects of the edge computing node. These business objects can be called scene switching business objects.

[0157] Optionally, the solution provided in the embodiment of the present application may involve machine learning technology of artificial intelligence. Machine Learning (ML) is a multi-disciplinary cross-disciplinary subject involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in studying how computers simulate or realize human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all fields of artificial intelligence. Machine learning and deep learning generally include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning. It is specifically described by the following embodiments: When a computer device predicts the offline business user of a certain edge computing node for a target application within a target time period, it can generate the application activity scenario of the business user, the operation activity information corresponding to the target application, and the offline behavior characteristics corresponding to the offline behavior of the business user according to the offline prediction model, and then output the predicted offline label corresponding to the offline behavior feature in the offline prediction model, and then determine the offline business user according to the predicted offline label. Among them, the offline prediction model is a machine learning model trained based on historical business behavior data. It is used to simulate the offline behavior of business users in different time periods and infer the application activity scenarios, time nodes and other behavior states when users will go offline. Correspondingly, the computer equipment can also predict the scene switching business users and new business users of a certain edge computing node through the corresponding machine learning model.

[0158] Step S402: Determine the number of new business objects added, the number of offline business objects offline, and the number of switching business objects for scene switching within the target time period for the edge computing node M. i The node changes computing power resource information.

[0159] Specifically, the initial login scenario corresponding to the newly added business object can be predicted; wherein one or more application activity scenarios include the initial login scenario; the application activity scenario in which the offline business object is located when the offline behavior occurs can be determined as the offline application activity scenario; the application activity scenario in which the node scene switching object is located before the scene switching is determined as the initial application activity scenario, and the application activity scenario in which the scene switching object is located after the scene switching is determined as the target application activity scenario; the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario is obtained, and according to the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, as well as the switching quantity, the offline quantity, and the new addition quantity, the target application is determined within the target time period for the edge computing node M. iThe node changes computing power resource information.

[0160] Among them, according to the average computing power demand information corresponding to the initial login scenario, offline application activity scenario, initial application activity scenario, and target application activity scenario, as well as the number of switches, offline numbers, and new additions, the target application is determined within the target time period for the edge computing node M. i The specific method for determining the node change computing power resource information may be as follows: according to the average computing power requirement information corresponding to the newly added quantity and the initial login scenario, the first change computing power resource information corresponding to the node newly added business object may be determined; according to the average computing power requirement information corresponding to the offline quantity and the online application activity scenario, the second change computing power resource information corresponding to the node offline business object may be determined; according to the average computing power requirement information corresponding to the initial application activity scenario, the average computing power requirement information corresponding to the target application activity scenario, and the switching quantity, the third change computing power resource information corresponding to the node scenario switching object may be determined; according to the first change computing power resource information, the second change computing power resource information, and the third change computing power resource information, the target application may be determined within the target time period for the edge computing node M. i The node changes computing power resource information.

[0161] It should be understood that when a business object logs in or opens a target application, it usually enters a default interface. The scene corresponding to this default interface can be called the initial login scene; wherein, the one or more application activity scenes included in the target application include the initial login scene. For example, when a business object logs in to a cloud gaming application, it usually enters the game homepage lobby scene, which can be called the initial login scene. The computer device can obtain the average computing power requirement information corresponding to the initial login scene (that is, the average computing power resources required by each online business object in the initial login scene), and according to the newly added number of business objects and the average computing power requirement information corresponding to the initial login scene (for example, multiplying the newly added number by the average computing power requirement information corresponding to the initial login scene), the newly added computing power resource information corresponding to the newly added business objects can be determined (such as the product of the newly added number and the average computing power requirement information corresponding to the initial login scene), and the newly added computing power resource information can be the edge computing node M. i When providing functional services for new business objects, additional computing resources are required; the additional computing resource information can be referred to as the first changed computing resource information.

[0162] It should be understood that the computer device can also obtain the application activity scenario in which the offline business object is located when the offline behavior occurs (hereinafter referred to as the offline application activity scenario), and determine the released computing power resource information corresponding to the offline business object based on the average computing power demand information corresponding to the offline application activity scenario (that is, the computing power resources required by each online business object on average in the offline application activity scenario) and the offline number of offline business objects. In other words, when these offline business users close the target application, the edge computing node M i There is no need to provide functional services for it, and the corresponding computing resources can be released. These released computing resources can be called second change computing resource information. The second change computing resource information can be the product of the number of offline applications and the average computing demand information corresponding to the offline application activity scenario.

[0163] It should be understood that because different application activity scenarios usually require different computing resources, when the business object changes the scenario, the required computing resources will also change (may increase, may decrease, or may remain unchanged). Then, the computer device can predict the edge computing node M i Among the current online business objects, the business objects that will switch scenes within the target time period are predicted, and the application activity scenes of these scene switching objects before the scene switching (which can be called the initial application activity scene) and the application activity scenes after the scene switching (which can be called the target application activity scene) are predicted; for the initial application activity scene, the edge computing node can no longer continue to provide corresponding computing resources for these scene switching users, and the corresponding computing resources should be released; for the target application activity scene, the edge computing node needs to provide it with corresponding computing resources, and new computing resources should be added. Then the average computing power requirement information corresponding to the initial application activity scenario (the computing power resources required by each online business object on average in the initial application activity scenario) and the average computing power requirement information corresponding to the target application activity scenario (the computing power resources required by each online business object on average in the target application activity scenario) can be obtained. The product of the average computing power requirement information corresponding to the initial application activity scenario and the number of switches of the scene switching user can be determined as the computing power resources that should be released corresponding to the initial application activity scenario; the product of the average computing power requirement information corresponding to the target application activity scenario and the number of switches of the scene switching user can be determined as the computing power resources that should be added corresponding to the target application activity scenario. The computing power resources that should be released corresponding to the initial application activity scenario and the computing power resources that should be added corresponding to the target application activity scenario can be added together, and the result can be used as the third change computing power resource information corresponding to the scene switching user (which may be a positive value or a negative value).

[0164] Furthermore, the newly added computing resources can be added together, and the computing resources that should be released can be subtracted from the result of the addition to obtain the final node change computing resource information. That is, the first change computing resource information and the third change computing resource information can be added together, and then the second change computing resource information can be subtracted from the result of the addition to obtain the target application in the target time period for the edge computing node M. i The node changes computing power resource information.

[0165] Step S403, when the node change computing power resource information of each edge computing node of at least two edge computing nodes of the target application is determined, the sum of the node change computing power resource information corresponding to the at least two edge computing nodes is determined as the total change computing power resource information of the edge computing node cluster.

[0166] It should be understood that each edge computing node can use the above-mentioned method to determine the edge computing node M. i After determining the node change computing power resource information of each edge computing node in the edge computing node cluster, the sum of the node change computing power resource information of all edge computing nodes in the edge computing node cluster can be counted, thereby determining the total change computing power resource information corresponding to the edge computing node cluster.

[0167] For further information, see Figure 5 , Figure 5 This is a flow chart of a method for determining node processing conditions satisfied by an edge computing node cluster provided in an embodiment of the present application. The process may correspond to the above Figure 3 The process of determining the node processing conditions of the edge computing node cluster in step S103 in the corresponding embodiment. Figure 5 As shown, the process may include the following steps S201-S203:

[0168] Step S201, determining the computing power resource change trend corresponding to the edge computing node cluster according to the total changed computing power resource information.

[0169] Specifically, the computing power resource change trend may include an increasing change trend, a decreasing change trend, and a stable change trend (i.e., the total change computing power resource information is 0). It should be understood that the total change computing power resource information can be compared with the value 0, so as to determine whether the target application's demand for computing power resources has changed within the target time period, and if so, whether the change is an increasing change or a decreasing change. If the total change computing power resource information is greater than the value 0, it is an increasing change, and the computing power resource change trend can be an increasing change trend; if the total change computing power resource information is less than the value 0, it is a decreasing change, and the computing power resource change trend can be a decreasing change trend; if the total change computing power resource information is equal to the value 0, the target application's demand for computing power resources has not changed within the target time period, and the computing power resource change trend can be a stable change trend.

[0170] Step S202: If the computing power resource change trend is an increasing trend, the node processing conditions satisfied by the edge computing node cluster are determined according to the total idle computing power resource information and the total changed computing power resource information; the node processing conditions include node wake-up conditions and node sleep conditions.

[0171] Specifically, if the computing power resource change trend is an increasing trend, the total idle computing power resource information can be compared with the total changed computing power resource information; if the total idle computing power resource information is greater than the total changed computing power resource information, it can be determined that the edge computing node cluster meets the node sleep condition; if the total idle computing power resource information is less than the total changed computing power resource information, it can be determined that the edge computing node cluster meets the node wake-up condition.

[0172] It can be understood that if the total change computing power resource information is greater than the value 0, it can be determined that the computing power demand of the target application for the edge computing node cluster is an increasing change, and the computing power resource change trend corresponding to the edge computing node cluster is an increasing change trend; at this time, the total idle computing power resource information can be compared with the total change computing power resource information to determine whether the total idle computing power resource information can meet the newly added computing power resources. If the total idle computing power resource information is greater than the total change computing power resource information, it can be determined that after the total idle computing power resource information meets the newly added computing power resources, there is still excess idle computing power resource information. At this time, in order to reduce the operating power loss of the edge computing node cluster and reduce the operating time of the edge computing nodes, some edge computing nodes of the edge computing node cluster can be put into sleep mode within the target time period, that is, the edge computing node cluster meets the node sleep condition at this time; and if the total idle computing power resource information If the source information is less than the total changed computing power resource information, it can be determined that the total idle computing power resource information cannot meet the newly added computing power resources. At this time, it can be determined that the edge computing node cluster meets the node wake-up conditions, and the edge computing nodes in the edge computing node cluster that are in a dormant state can be woken up within the target time period to enable them to operate normally and provide computing power resources for the target application (optionally, if there are no dormant edge computing nodes in the edge computing node cluster, new edge computing nodes can be added within the target time period and deployed to the edge computing node cluster to provide computing power resources for the target application); if the total idle computing power resource information is equal to the total changed computing power resource information, it can be determined that the total idle computing power resource information can meet the newly added computing power resources, and there is no redundant idle computing power resource information, then the edge computing node cluster does not need to be put to sleep or woken up within the target time period.

[0173] Step S203: If the computing power resource change trend is a reduction trend, it is determined that the edge computing node cluster meets the node sleep condition.

[0174] Specifically, if the total changed computing power resource information is less than the value of 0, it can be determined that the computing power demand of the target application for the edge computing node cluster is a reduced change, and the computing power resource change trend corresponding to the edge computing node cluster is a reduced change trend; at this time, the edge computing node cluster can release part of the computing power resources, then some edge computing nodes of the edge computing node cluster can be directly put into sleep mode within the target time period, that is, the edge computing node cluster meets the node sleep conditions at this time.

[0175] It can be understood that if the total change computing power resource information is equal to the value 0, it can be determined that the computing power demand of the target application for the edge computing node cluster has not changed, and at this time, the edge computing node can be subjected to node sleep processing or node wake-up processing. Optionally, when the total change computing power resource information is equal to the value 0, it can be determined whether to perform sleep processing on the edge computing node cluster based on the size of the total idle computing power resource information. If the total idle computing power resource information is sufficient (for example, the total idle computing power resource information is greater than the idle computing power threshold (which can be a manually specified value)), then some edge computing nodes in the edge computing node cluster can be subjected to sleep processing within the target time period. Optionally, when the total change computing power resource information is less than the value 0, if the value of the total change computing power resource information is very small (that is, the computing power resource demand of the target application changes very little, and the total change computing power resource information is a value slightly greater than 0), then the state of the edge computing node cluster can be kept unchanged at this time, and the edge computing nodes in the edge computing node cluster are not subjected to sleep processing or wake-up processing.

[0176] In an embodiment of the present application, after determining the current total idle computing power resource information of the edge computing node cluster and the total change computing power resource information for the edge computing node cluster of the target application within the target time period, it can be determined whether the edge computing node cluster meets the node sleep condition or the node wake-up condition based on the total idle computing power resource information and the total change computing power resource information. When the edge computing node meets the node wake-up condition, the awakened edge computing node can be awakened within the target time period; when the node sleep condition is met, the dormant edge computing node can be put into sleep within the target time period. The total change computing power resource information is the computing power demand change information of the target application for the edge computing node cluster within the target time period. The sleep or wake-up processing is performed within the target time period according to the total change computing power resource information, that is, some edge computing nodes in the edge computing node cluster are dynamically put into sleep or wake-up according to the real-time computing power demand change information. This not only allows the computing power resources provided by the edge computing node cluster to meet the computing power demand change information, but also allows the edge computing nodes in the edge computing node cluster to sleep according to the real-time computing power demand, reducing the number of running nodes, that is, reducing the running time, thereby reducing the redundant power consumption loss of the edge computing nodes, greatly reducing the failure rate of the nodes, and then greatly reducing the operating costs. In summary, the present application can dynamically sleep or wake up the edge computing nodes according to the real-time total change computing power resource information, so that the edge computing nodes can reduce the operating costs while meeting the computing power demand.

[0177] For further understanding, please see Figure 6 , Figure 6 is a logic flow chart for performing node processing provided by an embodiment of the present application, such as Figure 6As shown, the logic flow may include at least the following steps S51-S512:

[0178] Step S51, predicting the change trend of computing power resources.

[0179] Specifically, the computing power resource change trend may include an increasing change trend, a decreasing change trend, and a stable change trend.

[0180] Step S52: Determine whether the required computing power of the target application increases according to the computing power resource change trend.

[0181] Specifically, if the required computing power of the target application increases, the process proceeds to the subsequent step S53; if the required computing power of the target application does not increase (ie, decreases or remains unchanged), the process proceeds to the subsequent step S54.

[0182] Step S53, determine whether the idle computing power can meet the newly added required computing power.

[0183] Specifically, when the required computing power of the target application increases, if the total idle computing power resource information of the edge computing node cluster can meet the newly required computing power, then the subsequent step S55 can be entered; if the total idle computing power resource information of the edge computing node cluster cannot meet the newly required computing power, then the subsequent step S56 can be entered.

[0184] Step S54, determining whether the required computing power remains unchanged.

[0185] Specifically, when the target application's required computing power does not increase, it can be determined whether the target application's required computing power remains unchanged or decreases. If the target application's required computing power remains unchanged, the process proceeds to the subsequent step S59; if the target application's required computing power decreases, the process proceeds to the subsequent step S510.

[0186] Step S55, keeping the edge computing node cluster in a dormant or awake state unchanged.

[0187] Specifically, the dormant or awakening state of the edge computing node cluster is kept unchanged, that is, the edge computing node cluster is not put to sleep or awakened within the target time period.

[0188] Step S56, calculating the wake-up quantity.

[0189] Specifically, as described in the above step S53, if the total idle computing power resource information of the edge computing node cluster cannot meet the newly added required computing power, the number of node wake-ups can be calculated.

[0190] Step S57, waking up each device one by one.

[0191] Specifically, after calculating the number of awakened nodes, edge computing nodes corresponding to the awakened number may be awakened one by one among the dormant edge computing nodes.

[0192] Step S58, joining the cluster.

[0193] Specifically, the awakened edge computing node can be added to the edge computing node cluster to provide computing resources for the target application within the target time period.

[0194] Step S59, keep the sleep or wake-up state of the edge computing node cluster unchanged.

[0195] Step S510, calculating the sleep quantity.

[0196] Specifically, as described in the above step S54, if the required computing power of the target application is reduced, the number of dormant nodes can be calculated.

[0197] Step S511, freezing the dormant nodes.

[0198] Specifically, after calculating the number of dormant nodes, the edge computing nodes to be dormant can be obtained and frozen.

[0199] Step S512: instance migration.

[0200] Specifically, instance migration is to migrate the running data of the edge computing node to be dormant. Figure 7 , Figure 7 This is a schematic diagram of a process for migrating operation data provided by an embodiment of the present application. Figure 7 As shown, the process may include the following steps S5121-S5125:

[0201] Step S5121, the running status is saved.

[0202] Specifically, all running status data of the target application instance can be saved on the edge computing node to be dormant.

[0203] Step S5122, data transmission.

[0204] Specifically, the above operating status data can be transmitted to the target edge computing node (any edge computing node that is in normal operating state within the target time period).

[0205] Step S5123, pull up the instance.

[0206] Specifically, in the target edge computing node, a new application instance can be pulled up and the above-mentioned running status data can be loaded.

[0207] Step S5124, disconnect or reconnect.

[0208] Specifically, the connection between the edge computing node to be dormant and the target application can be disconnected, and the connection between the target edge computing node and the target application can be established at the same time. After the connection is successfully established, the application instance is destroyed in the edge computing node to be dormant.

[0209] Step S5125, restore the application process.

[0210] Specifically, the target application continues to be rendered in the edge computing node to be dormant, and the application process continues using the above connection.

[0211] It should be noted that the migration process of running data requires seamless and fast migration, and the time of each link of data migration should be strictly controlled to avoid obvious perception of business objects; because there are special scenarios such as scene switching and scene loading (such as game loading in cloud gaming applications) in the target application, the business objects will wait for the application to load resources at this time. In this application, this kind of waiting time for the application to load resources can be selected to migrate the running data, which will only increase the loading waiting time very little and will not affect the actual operation of the business objects in the target application.

[0212] Step S513, sleep processing.

[0213] Specifically, after the running data is successfully migrated, the edge computing node to be dormant can be put into dormancy.

[0214] For the specific implementation of steps S51 to S513, please refer to the above Figure 3 The description of step S101 to step S104 in the corresponding embodiment will not be repeated here.

[0215] For further information, see Figure 8 , Figure 8 This is a system architecture diagram provided by an embodiment of the present application. Figure 8 As shown, the system as a whole may include a management server and an edge computing node, wherein the edge computing node may include an instance management module, a computing power information collection module, and a machine control module. For ease of understanding, the functions corresponding to each module will be described below.

[0216] The computing power information collection module is mainly used to collect the current computing power usage and report it to the management server. It can mainly include CPU computing power information collection and GPU computing power information collection. Among them, CPU computing power information collection mainly collects CPU model, current maximum computing power that can be achieved, current actual computing power (i.e. occupied computing power), etc.; while GPU computing power information collection mainly collects GPU model, current maximum computing power that can be achieved, and current actual computing power.

[0217] The instance management module is mainly used for instance migration and instance reception. Instance migration mainly includes: when the local machine needs to be fully repaired, the local application instance (such as a cloud game instance) is migrated to another edge computing node; instance reception includes: receiving application instances migrated from the edge computing node to be dormant.

[0218] The machine control module is mainly used to perform sleep or wake-up operations according to the instructions of the management server.

[0219] The management server may include a data analysis module, a computing power prediction module, an instance scheduling module, and a node scheduling module. For ease of understanding, the functions of each module will be explained below.

[0220] Data analysis module: used to receive data reports from edge computing nodes and perform preliminary analysis for use by other modules.

[0221] Computing power prediction module: used to predict the idle computing power or computing power gap of edge computing nodes (that is, the computing power lacking in the edge computing node cluster), and calculate the number of machines that need to sleep or wake up based on the maximum computing power of each edge computing node.

[0222] The node scheduling module is used to determine the edge computing nodes that are dormant or awakened according to the calculation results of the computing power prediction module, ensuring that all edge computing nodes can take turns to dormant, so as to keep the operating time of the edge computing nodes basically consistent and achieve consistency between performance loss and service life.

[0223] The instance scheduling module is used to schedule application instances when edge computing nodes need to sleep (including no longer allocating new requests to edge computing nodes to be dormant, running data migration, etc.).

[0224] In an embodiment of the present application, after determining the current total idle computing power resource information of the edge computing node cluster and the total change computing power resource information for the edge computing node cluster of the target application within the target time period, it can be determined whether the edge computing node cluster meets the node sleep condition or the node wake-up condition based on the total idle computing power resource information and the total change computing power resource information. When the edge computing node meets the node wake-up condition, the awakened edge computing node can be awakened within the target time period; when the node sleep condition is met, the dormant edge computing node can be put into sleep within the target time period. The total change computing power resource information is the computing power demand change information of the target application for the edge computing node cluster within the target time period. The sleep or wake-up processing is performed within the target time period according to the total change computing power resource information, that is, some edge computing nodes in the edge computing node cluster are dynamically put into sleep or wake-up according to the real-time computing power demand change information. This not only allows the computing power resources provided by the edge computing node cluster to meet the computing power demand change information, but also allows the edge computing nodes in the edge computing node cluster to sleep according to the real-time computing power demand, reducing the number of running nodes, that is, reducing the running time, thereby reducing the redundant power consumption loss of the edge computing nodes, greatly reducing the failure rate of the nodes, and then greatly reducing the operating costs. In summary, the present application can dynamically sleep or wake up the edge computing nodes according to the real-time total change computing power resource information, so that the edge computing nodes can reduce the operating costs while meeting the computing power demand.

[0225] For further information, see Fig. 9 , Fig. 9 : is a schematic diagram of the structure of a data processing device based on edge computing provided in an embodiment of the present application. The data processing device based on edge computing can be a computer program (including program code) running in a computer device, for example, the data processing device based on edge computing is an application software; the data processing device based on edge computing can be used to execute Figure 3 As shown in the method. Fig. 9 As shown, the data processing device 1 based on edge computing may include: an idle resource acquisition module 11, a change resource prediction module 12, a sleep processing module 13 and a wake-up processing module 14.

[0226] The idle resource acquisition module 11 is used to obtain the total idle computing power resource information of the edge computing node cluster; the total idle computing power resource information is the computing power resource information of the edge computing node cluster that is not occupied by the target application when the target application is running;

[0227] The change resource prediction module 12 is used to predict the total change computing power resource information of the edge computing node cluster for the target application within the target time period;

[0228] The sleep processing module 13 is used to determine the edge computing nodes to be dormant in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node sleep condition;

[0229] The sleep processing module 13 is further used to perform sleep processing on the sleep edge computing node within the target time period;

[0230] A wake-up processing module 14 is used to determine the edge computing node to be awakened in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node awakening condition;

[0231] The wake-up processing module 14 is further configured to perform wake-up processing on the edge computing node to be awakened within a target time period.

[0232] The specific implementation of the idle resource acquisition module 11, the change resource prediction module 12, the sleep processing module 13 and the wake-up processing module 14 can be found in the above Figure 3 The description of step S101 to step S104 in the corresponding embodiment will not be repeated here.

[0233] In one embodiment, the sleep processing module 13 may include: a sleep quantity determining unit 131 and a sleep node determining unit 132 .

[0234] The sleep quantity determining unit 131 is used to determine the predicted total idle computing power resource information of the edge computing node cluster within the target time period according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node sleep condition;

[0235] The sleep quantity determination unit 131 is further used to determine the sleep quantity of nodes according to the predicted total idle computing power resource information and the unit computing power resource information; the unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster;

[0236] The sleep node determination unit 132 is used to determine the edge computing nodes to be sleep in the edge computing node cluster according to the number of sleep nodes.

[0237] The specific implementation of the sleep number determination unit 131 and the sleep node determination unit 132 can be found in the above Figure 3 The description of step S103 in the corresponding embodiment will not be repeated here.

[0238] In one embodiment, the sleepy node determination unit 132 may include: a sleepy node acquisition subunit 1321 , a polling table acquisition subunit 1322 , and a sleepy node determination subunit 1323 .

[0239] The dormant node acquisition subunit 1321 is used to acquire a dormant edge computing node in the edge computing node cluster; the dormant edge computing node is an edge computing node in the edge computing node cluster that is currently in a dormant state;

[0240] The polling table acquisition subunit 1322 is used to acquire a node sleep polling table corresponding to the edge computing node cluster; the node sleep polling table includes a sleep polling order of each edge computing node in the edge computing node cluster;

[0241] The sleep node determination subunit 1323 is used to obtain the edge computing nodes to be dormant from the node sleep polling table in sequence according to the position of the dormant edge computing nodes in the node sleep sorting table and the number of dormant nodes; the edge computing nodes to be dormant are currently in normal operation.

[0242] The specific implementation of the dormant node acquisition subunit 1321, the polling table acquisition subunit 1322, and the dormant node determination subunit 1323 can be found in the above Figure 3 The description of step S103 in the corresponding embodiment will not be repeated here.

[0243] In one embodiment, the sleepy node determination unit 132 may further include: a normal node acquisition subunit 1324 , a sorting subunit 1325 , and a sleepy node acquisition subunit 1326 .

[0244] The normal node acquisition subunit 1324 is used to acquire N normally operating edge computing nodes in the edge computing node cluster; the N normally operating edge computing nodes are currently in a normal operating state; N is a positive integer;

[0245] The sorting subunit 1325 is used to obtain node idle computing resource information corresponding to each of the N normally operating edge computing nodes;

[0246] The sorting subunit 1325 is further used to sort the idle computing resource information of the N nodes in order of size to obtain a node idle computing resource information sequence;

[0247] The dormant node acquisition subunit 1326 is used to sequentially acquire the target node idle computing resource information in the node idle computing resource information sequence according to the number of dormant nodes, and determine the normally operating edge computing node corresponding to the target node idle computing resource information as the computing node to be dormant.

[0248] The specific implementation of the normal node acquisition subunit 1324, the sorting subunit 1325 and the dormant node acquisition subunit 1326 can be found in the above Figure 3The description of step S103 in the corresponding embodiment will not be repeated here.

[0249] In one embodiment, the sleep processing module 13 may include: a freeze processing unit 133 , a quantity counting unit 134 , a data migration unit 135 and a sleep processing unit 136 .

[0250] A freezing processing unit 133 is used to freeze the running state of the edge computing node to be dormant within a target time period to obtain a frozen computing node;

[0251] A quantity counting unit 134, used to count the online quantity of node online business objects corresponding to the frozen computing nodes;

[0252] The data migration unit 135 is used to obtain the operation data of the frozen computing node when the number of online nodes is less than the online threshold, and migrate the operation data to the target edge computing node; the operation state of the target edge computing node in the target time period is a normal operation state;

[0253] The sleep processing unit 136 is used to freeze the computing node and perform sleep processing when the running data is successfully migrated to the target edge computing node.

[0254] The specific implementation of the freezing processing unit 133, the number counting unit 134, the data migration unit 135 and the sleep processing unit 136 can be found in the above Figure 3 The description of step S103 in the corresponding embodiment will not be repeated here.

[0255] In one embodiment, the wake-up processing module 14 may include: a wake-up quantity determining unit 141 and a wake-up node acquiring unit 142 .

[0256] The wake-up quantity determination unit 141 is used to determine the predicted excess computing power resource information of the edge computing node cluster within the target time period according to the total idle computing power resource information and the total changed computing power resource information if the edge computing node cluster meets the node wake-up condition;

[0257] The wake-up quantity determination unit 141 is further used to determine the node wake-up quantity according to the predicted over-limit computing power resource information and the unit computing power resource information; the unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster;

[0258] The awakening node acquisition unit 142 is used to acquire a dormant edge computing node in the edge computing node cluster; the dormant edge computing node is an edge computing node in the edge computing node cluster that is currently in a dormant state;

[0259] The awakened node acquisition unit 142 is further configured to acquire the edge computing nodes to be awakened from the dormant edge computing nodes according to the number of awakened nodes.

[0260] The specific implementation of the wake-up number determination unit 141 and the wake-up node acquisition unit 142 can be found in the above Figure 3 The description of step S104 in the corresponding embodiment will not be repeated here.

[0261] In one embodiment, the edge computing-based data processing device 1 may further include: a change trend determination module 15 and a condition determination module 16 .

[0262] A change trend determination module 15 is used to determine the change trend of computing power resources corresponding to the edge computing node cluster according to the total change computing power resource information;

[0263] The condition determination module 16 is used to determine the node processing conditions satisfied by the edge computing node cluster according to the total idle computing resource information and the total changed computing resource information if the computing resource change trend is an increasing change trend; the node processing conditions include node wake-up conditions and node sleep conditions;

[0264] The condition determination module 16 is also used to determine whether the edge computing node cluster meets the node sleep condition if the computing power resource change trend is a reduction trend.

[0265] The specific implementation of the change trend determination module 15 and the condition determination module 16 can be found in the above Figure 3 The description of step S102 in the corresponding embodiment will not be repeated here.

[0266] In one embodiment, the condition determination module 16 may include: a resource comparison unit 161 and a condition determination unit 162 .

[0267] A resource comparison unit 161, configured to compare the total idle computing power resource information with the total changed computing power resource information;

[0268] A condition determination unit 162, configured to determine that the edge computing node cluster meets the node sleep condition if the total idle computing resource information is greater than the total changed computing resource information;

[0269] The condition determination unit 162 is further used to determine whether the edge computing node cluster meets the node wake-up condition if the total idle computing power resource information is less than the total changed computing power resource information.

[0270] The specific implementation of the resource comparison unit 161 and the condition determination unit 162 can be found in the above Figure 3 The description of step S102 in the corresponding embodiment will not be repeated here.

[0271] In one embodiment, the edge computing node cluster includes at least two edge computing nodes; the at least two edge computing nodes include edge computing node M i ; i is a positive integer;

[0272] The idle resource acquisition module 11 may include: an occupied resource determination unit 111 and an idle resource determination unit 112 .

[0273] The occupied resource determination unit 111 is used to obtain the target application in the edge computing node M i Q pieces of current operation information; one piece of current operation information includes an application activity scenario where an online business object is located, and information about the resources occupied by the business object in the application activity scenario;

[0274] The occupied resource determination unit 111 is further used to obtain the occupied resource information of the business object corresponding to each current operation information, and obtain the occupied resource information of Q business objects;

[0275] The occupied resource determination unit 111 is further configured to determine the sum of the occupied resource information of the Q business objects as the target application for the edge computing node M. i The occupied node computing resource information;

[0276] The idle resource determination unit 112 is used to determine the total idle computing power resource information of the edge computing node cluster according to the occupied node computing power resource information corresponding to the at least two edge computing nodes respectively when determining the occupied node computing power resource information of each edge computing node of the target application for at least two edge computing nodes.

[0277] The specific implementation of the occupied resource determination unit 111 and the idle resource determination unit 112 can be found in the above Figure 3 The description of step S101 in the corresponding embodiment will not be repeated here.

[0278] In one embodiment, the idle resource determination unit 112 is further specifically configured to obtain the maximum computing power resource information corresponding to each edge computing node in the at least two edge computing nodes, and obtain at least two maximum computing power resource information;

[0279] The idle resource determination unit 112 is further specifically configured to determine total computing power resource information corresponding to at least two maximum computing power resource information, and determine total occupied computing power resource information corresponding to at least two occupied node computing power resource information;

[0280] The idle resource determination unit 112 is further specifically used to determine the absolute value of the resource difference between the total computing power resource information and the total occupied computing power resource information as the total idle computing power resource information of the edge computing node cluster.

[0281] In one embodiment, the edge computing node cluster includes at least two edge computing nodes; the at least two edge computing nodes include edge computing node M i ; i is a positive integer;

[0282] The change resource prediction module 12 may include: an object prediction unit 121 and a change resource determination unit 122 .

[0283] The object prediction unit 121 is used to predict the target application for the edge computing node M within the target time period. i New business objects, offline business objects, and scene switching business objects;

[0284] The change resource determination unit 122 is used to determine the target application within the target time period for the edge computing node M according to the number of new business objects added, the number of offline business objects offline, and the number of switching business objects switched by the scene. i Node change computing power resource information;

[0285] The change resource determination unit 122 is also used to determine the sum of the node change computing power resource information corresponding to at least two edge computing nodes as the total change computing power resource information of the edge computing node cluster when determining the node change computing power resource information of each edge computing node of the target application for at least two edge computing nodes.

[0286] The specific implementation of the object prediction unit 121 and the change resource determination unit 122 can be found in the above Figure 3 The description of step S102 in the corresponding embodiment will not be repeated here.

[0287] In one embodiment, the target application includes one or more application activity scenarios;

[0288] The changed resource determination unit 122 is further specifically used to predict the initial login scenario corresponding to the newly added business object; the one or more application activity scenarios include the initial login scenario;

[0289] The changed resource determination unit 122 is further specifically configured to determine the application activity scenario in which the offline business object is located when the offline behavior occurs as the offline application activity scenario;

[0290] The change resource determination unit 122 is further specifically used to determine the application activity scene in which the node scene switching object is located before the scene switching is performed as the initial application activity scene, and determine the application activity scene in which the scene switching object is located after the scene switching is performed as the target application activity scene;

[0291] The change resource determination unit 122 is further specifically used to obtain the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, and determine the target application within the target time period for the edge computing node M according to the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, as well as the switching quantity, the offline quantity, and the new quantity. i The node changes computing power resource information.

[0292] In one embodiment, the change resource determination unit 122 is further specifically used to determine the first change computing power resource information corresponding to the newly added business object of the node according to the newly added quantity and the average computing power requirement information corresponding to the initial login scenario;

[0293] The change resource determination unit 122 is further specifically configured to determine the second change computing power resource information corresponding to the node offline business object according to the offline quantity and the average computing power demand information corresponding to the online application activity scenario;

[0294] The change resource determination unit 122 is further specifically used to determine the third change computing power resource information corresponding to the node scene switching object according to the average computing power requirement information corresponding to the initial application activity scene, the average computing power requirement information corresponding to the target application activity scene, and the switching quantity;

[0295] The change resource determination unit 122 is further specifically configured to determine the target application within the target time period according to the first change computing power resource information, the second change computing power resource information, and the third change computing power resource information, for the edge computing node M i The node changes computing power resource information.

[0296] In an embodiment of the present application, after determining the current total idle computing power resource information of the edge computing node cluster and the total change computing power resource information for the edge computing node cluster of the target application within the target time period, it can be determined whether the edge computing node cluster meets the node sleep condition or the node wake-up condition based on the total idle computing power resource information and the total change computing power resource information. When the edge computing node meets the node wake-up condition, the awakened edge computing node can be awakened within the target time period; when the node sleep condition is met, the dormant edge computing node can be put into sleep within the target time period. The total change computing power resource information is the computing power demand change information of the target application for the edge computing node cluster within the target time period. The sleep or wake-up processing is performed within the target time period according to the total change computing power resource information, that is, some edge computing nodes in the edge computing node cluster are dynamically put into sleep or wake-up according to the real-time computing power demand change information. This not only allows the computing power resources provided by the edge computing node cluster to meet the computing power demand change information, but also allows the edge computing nodes in the edge computing node cluster to sleep according to the real-time computing power demand, reducing the number of running nodes, that is, reducing the running time, thereby reducing the redundant power consumption loss of the edge computing nodes, greatly reducing the failure rate of the nodes, and then greatly reducing the operating costs. In summary, the present application can dynamically sleep or wake up the edge computing nodes according to the real-time total change computing power resource information, so that the edge computing nodes can reduce the operating costs while meeting the computing power demand.

[0297] For further information, see Fig.10 , Fig.10 Schematic diagram of the structure of a computer device provided in an embodiment of the present application. Fig.10 As shown above Fig. 9 The device 1 in the corresponding embodiment can be applied to the above-mentioned computer device 1000, and the above-mentioned computer device 1000 may include: a processor 1001, a network interface 1004 and a memory 1005. In addition, the above-mentioned computer device 1000 also includes: an object interface 1003, and at least one communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. Among them, the object interface 1003 may include a display screen (Display), a keyboard (Keyboard), and the object interface 1003 may optionally include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or it may be a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 1005 may optionally also be at least one storage device located away from the aforementioned processor 1001. Fig.10As shown, the memory 1005 as a computer-readable storage medium may include an operating system, a network communication module, an object interface module, and a device control application program.

[0298] exist Fig.10 In the computer device 1000 shown, the network interface 1004 can provide a network communication function; the object interface 1003 is mainly used to provide an input interface for the object; and the processor 1001 can be used to call the device control application stored in the memory 1005 to achieve:

[0299] Obtain the total idle computing power resource information of the edge computing node cluster; the total idle computing power resource information is the computing power resource information of the edge computing node cluster that is not occupied by the target application when the target application is running;

[0300] Predict the total change in computing power resources of the edge computing node cluster for the target application within the target time period;

[0301] If the edge computing node cluster meets the node sleep condition, the edge computing nodes to be dormant are determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, and the dormant edge computing nodes are put into sleep mode within the target time period;

[0302] If the edge computing node cluster meets the node wake-up condition, the edge computing node to be awakened is determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, and the edge computing node to be awakened is awakened within the target time period.

[0303] It should be understood that the computer device 1000 described in the embodiment of the present application can execute the above Figures 3 to 5 The description of the data processing method based on edge computing in the corresponding embodiment can also be performed as described above. Fig. 9 The description of the data processing device 1 based on edge computing in the corresponding embodiment will not be repeated here. In addition, the description of the beneficial effects of adopting the same method will not be repeated here.

[0304] In addition, it should be pointed out here that: the embodiment of the present application also provides a computer-readable storage medium, and the computer-readable storage medium stores a computer program executed by the computer device 1000 for data processing mentioned above, and the computer program includes program instructions. When the processor executes the program instructions, the computer program can execute the above-mentioned data processing. Figures 3 to 5The description of the above-mentioned data processing method based on edge computing nodes in the corresponding embodiment will not be repeated here. In addition, the description of the beneficial effects of the same method will not be repeated. For technical details not disclosed in the computer-readable storage medium embodiment involved in this application, please refer to the description of the method embodiment of this application.

[0305] The computer-readable storage medium may be an edge computing-based data processing device provided in any of the aforementioned embodiments or an internal storage unit of the computer device, such as a hard disk or memory of a computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (smart media card, SMC), a secure digital (secure digital, SD) card, a flash card (flash card), etc. equipped on the computer device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the computer device and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0306] In one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method provided in one aspect of the embodiments of the present application.

[0307] The terms "first", "second", etc. in the description, claims, and drawings of the embodiments of the present application are used to distinguish different objects, rather than to describe a specific order. In addition, the term "comprising" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, device, product, or equipment that includes a series of steps or units is not limited to the listed steps or modules, but optionally includes steps or modules that are not listed, or optionally includes other step units inherent to these processes, methods, devices, products, or equipment.

[0308] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0309] The method and related apparatus provided by the embodiment of the present application are described with reference to the method flow chart and / or structural diagram provided by the embodiment of the present application. Specifically, each process and / or box in the method flow chart and / or structural diagram, as well as the combination of the processes and / or boxes in the flow chart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the process. Figure 1 A process or multiple processes and / or structures Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or structures Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A flow or multiple flows and / or structures illustrate the steps of the functions specified in one block or multiple blocks.

[0310] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A data processing method based on edge computing, characterized in that: include: Get the total idle computing power resource information of the edge computing node cluster; The total idle computing power resource information is the computing power resource information of the edge computing node cluster that is not occupied by the target application when the target application is running; Predicting total change computing power resource information of the target application for the edge computing node cluster within a target time period; If the edge computing node cluster meets the node sleep condition, the predicted total idle computing power resource information of the edge computing node cluster within the target time period is determined according to the total idle computing power resource information and the total changed computing power resource information, the number of node sleeps is determined according to the predicted total idle computing power resource information and the unit computing power resource information, the edge computing nodes to be sleep are determined in the edge computing node cluster according to the number of node sleeps, and the edge computing nodes to be sleep are sleep processed within the target time period; The unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster; the node sleep condition refers to the condition that the computing power resource change trend indicated by the total change computing power resource information is an increasing change trend and the total idle computing power resource information is greater than the total change computing power resource information, or the computing power resource change trend indicated by the total change computing power resource information is an increasing change trend; If the edge computing node cluster meets the node wake-up condition, the edge computing node to be awakened is determined in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, and the edge computing node to be awakened is awakened within the target time period; the node wake-up condition refers to the condition that the computing power resource change trend indicated by the total changed computing power resource information is an increasing change trend, and the total idle computing power resource information is less than the total changed computing power resource information.

2. The method according to claim 1, characterized in that The step of determining the edge computing node to be dormant in the edge computing node cluster according to the number of dormant nodes includes: Acquire a dormant edge computing node in the edge computing node cluster; the dormant edge computing node is an edge computing node in the edge computing node cluster that is currently in a dormant state; Obtaining a node sleep polling table corresponding to the edge computing node cluster; the node sleep polling table includes a sleep polling order of each edge computing node in the edge computing node cluster; According to the position of the dormant edge computing node in the node dormancy ranking table and the number of dormant nodes, the to-be-dormant edge computing nodes are sequentially acquired from the node dormancy polling table; the to-be-dormant edge computing nodes are currently in a normal operating state.

3. The method according to claim 1, characterized in that The step of determining the edge computing node to be dormant in the edge computing node cluster according to the number of dormant nodes includes: Obtain N normally operating edge computing nodes in the edge computing node cluster; the N normally operating edge computing nodes are currently in a normal operating state; N is a positive integer; Obtain node idle computing power resource information corresponding to each of the N normally operating edge computing nodes; Sort the idle computing resource information of N nodes in order of size to obtain a node idle computing resource information sequence; The idle computing power resource information of the target node is obtained in sequence in the idle computing power resource information sequence according to the number of dormant nodes, and the normally operating edge computing node corresponding to the idle computing power resource information of the target node is determined as the edge computing node to be dormant.

4. The method according to claim 1, characterized in that: The performing a sleep process on the edge computing node to be sleep within the target time period includes: Within the target time period, freezing the running state of the edge computing node to be dormant to obtain a frozen computing node; Counting the online number of node online business objects corresponding to the frozen computing node; When the online number is less than the online threshold, the operating data of the frozen computing node is obtained, and the operating data is migrated to the target edge computing node; the operating state of the target edge computing node in the target time period is a normal operating state; When the running data is successfully migrated to the target edge computing node, the frozen computing node is put into hibernation.

5. The method according to claim 1, characterized in that If the edge computing node cluster meets the node wake-up condition, determining the edge computing node to be awakened in the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information, including: If the edge computing node cluster meets the node wake-up condition, determining the predicted excess computing power resource information of the edge computing node cluster within the target time period according to the total idle computing power resource information and the total changed computing power resource information; Determine the number of node wake-ups according to the predicted excess computing power resource information and the unit computing power resource information; the unit computing power resource information is the total computing power resource information corresponding to an edge computing node in the edge computing node cluster; Acquire a dormant edge computing node in the edge computing node cluster; the dormant edge computing node is an edge computing node in the edge computing node cluster that is currently in a dormant state; According to the node awakening number, the edge computing node to be awakened is obtained from the dormant edge computing nodes.

6. The method according to claim 1, characterized in that The method further comprises: Determine the computing power resource change trend corresponding to the edge computing node cluster according to the total changed computing power resource information; If the computing power resource change trend is an increasing change trend, determining the node processing conditions satisfied by the edge computing node cluster according to the total idle computing power resource information and the total changed computing power resource information; the node processing conditions include the node wake-up condition and the node sleep condition; If the computing power resource change trend is a reduction trend, it is determined that the edge computing node cluster meets the node sleep condition.

7. The method according to claim 6, characterized in that The determining, according to the total idle computing power resource information and the total changed computing power resource information, a node processing condition satisfied by the edge computing node cluster includes: Compare the total idle computing power resource information with the total changed computing power resource information; If the total idle computing power resource information is greater than the total changed computing power resource information, it is determined that the edge computing node cluster meets the node sleep condition; If the total idle computing power resource information is less than the total changed computing power resource information, it is determined that the edge computing node cluster meets the node wake-up condition.

8. The method according to claim 1, characterized in that The edge computing node cluster includes at least two edge computing nodes; the at least two edge computing nodes include edge computing nodes M i ; i is a positive integer; The obtaining of total idle computing power resource information of the edge computing node cluster includes: Get the target application in the edge computing node M i Q current operation information in; one current operation information includes an application activity scene where an online business object is located, and the business object occupied resource information in the application activity scene; Q is a positive integer; Obtain the resource occupation information of the business object corresponding to each current operation information, and obtain the resource occupation information of Q business objects; The sum of the resource occupied by the Q business objects is determined as the target application for the edge computing node M i The occupied node computing resource information; When the occupied node computing power resource information of each edge computing node of at least two edge computing nodes for the target application is determined, the total idle computing power resource information of the edge computing node cluster is determined according to the occupied node computing power resource information respectively corresponding to the at least two edge computing nodes.

9. The method according to claim 8, characterized in that The determining, according to the occupied node computing resource information respectively corresponding to the at least two edge computing nodes, the total idle computing resource information of the edge computing node cluster includes: Obtaining maximum computing power resource information corresponding to each edge computing node of the at least two edge computing nodes, to obtain at least two pieces of maximum computing power resource information; Determine the total computing power resource information corresponding to the at least two maximum computing power resource information, and determine the total occupied computing power resource information corresponding to the at least two occupied node computing power resource information; The absolute value of the resource difference between the total computing power resource information and the total occupied computing power resource information is determined as the total idle computing power resource information of the edge computing node cluster.

10. The method according to claim 1, characterized in that The edge computing node cluster includes at least two edge computing nodes; the at least two edge computing nodes include edge computing nodes M i ; i is a positive integer; The predicted total change computing power resource information of the edge computing node cluster for the target application within the target time period includes: Predict the target application to target the edge computing node M within the target time period i New business objects, offline business objects, and scene switching business objects; According to the number of new business objects, the number of offline business objects, and the number of switching business objects of the scene switching, it is determined that the target application is used for the edge computing node M within the target time period. i Node change computing power resource information; When the node change computing power resource information of the target application for each of the at least two edge computing nodes is determined, the sum of the node change computing power resource information corresponding to the at least two edge computing nodes is determined as the total change computing power resource information of the edge computing node cluster.

11. The method according to claim 10, characterized in that The target application includes one or more application activity scenarios; The target application is determined within the target time period for the edge computing node M according to the number of new business objects added, the number of offline business objects offline, and the number of switching of the scene switching business objects. i The node change computing resource information includes: Predicting an initial login scenario corresponding to the newly added business object; the one or more application activity scenarios including the initial login scenario; Determine the application activity scenario in which the offline business object is located when the offline behavior occurs as the offline application activity scenario; Determine the application activity scene in which the scene switching object is located before the scene switching is performed as the initial application activity scene, and determine the application activity scene in which the scene switching object is located after the scene switching is performed as the target application activity scene; Obtain the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, respectively; and determine the target application within the target time period for the edge computing node M according to the average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, as well as the switching quantity, the offline quantity, and the new quantity. i The node changes computing power resource information.

12. The method according to claim 11, characterized in that The average computing power demand information corresponding to the initial login scenario, the offline application activity scenario, the initial application activity scenario, and the target application activity scenario, as well as the switching quantity, the offline quantity, and the new quantity, determines the target application within the target time period for the edge computing node M. i The node change computing resource information includes: Determine first changed computing resource information corresponding to the newly added business object according to the newly added quantity and the average computing power requirement information corresponding to the initial login scenario; Determine the second changed computing power resource information corresponding to the offline business object according to the offline quantity and the average computing power demand information corresponding to the offline application activity scenario; Determine the third changed computing power resource information corresponding to the scene switching object according to the average computing power requirement information corresponding to the initial application activity scene, the average computing power requirement information corresponding to the target application activity scene, and the switching quantity; Determine the target application within the target time period according to the first change computing power resource information, the second change computing power resource information, and the third change computing power resource information, and i The node changes computing power resource information.

13. A computer device, characterized in that: include: Processor, memory, and network interface; The processor is connected to the memory and the network interface, wherein the network interface is used to provide a network communication function, the memory is used to store program code, and the processor is used to call the program code so that the computer device executes the method described in any one of claims 1-12.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing the method according to any one of claims 1 to 12.

Images