A connection resource scheduling method, cloud platform and electronic device

Abstract

The present application relates to the technical field of cloud platforms, and provides a connection resource scheduling method, a cloud platform and an electronic device. The method is applied to the cloud platform and includes the steps of: in response to a connection resource scheduling request, obtaining a scheduling delay factor; wherein the connection resource scheduling request is used to trigger the reallocation of connection resources between connected terminals and to-be-connected terminals; determining a target resource allocation algorithm from multiple resource allocation algorithms based on the scheduling delay factor; forcibly disconnecting the connection of the connected terminal and listening for a reconnection request sent by the connected terminal; selecting a target terminal from the connected terminal and the to-be-connected terminal based on the attribute information of the reconnection request and the target resource allocation algorithm; and allocating the connection resource to the target terminal, thereby realizing the reallocation of the connection resource, ensuring that the utility generated by the resource is maximized, and ensuring the smooth progress of the service.

Description

Technical Field

[0001] This application relates to the technical field of cloud platforms, and in particular, to a connection resource scheduling method, a cloud platform, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the rapid development of Internet technology and Internet of Things technology, in some cloud platform application scenarios, such as the centralized management platform of Internet of Things (IoT) devices and cloud game platforms, a large number of terminal devices need to access the cloud platform for real-time communication, resulting in frequent occurrence of high-concurrency connection requests, which poses higher requirements for the connection resource management and scheduling of the cloud platform.

[0003] Existing connection resource management solutions usually adopt a static allocation mechanism. A common approach is to estimate the maximum concurrency at the design stage and configure enough connection channels accordingly to meet the requirements at peak times. This static resource configuration method, especially in the case of sudden high concurrency, the preset connection capacity will soon be occupied, and some of these occupied connection resources are ineffective occupations. For example, the terminal device does not upload data, or the terminal is not in the game, resulting in a delay in the response to new connection requests, or even a situation where access cannot be achieved for a long time, seriously affecting the progress of the business. Summary of the Invention

[0004] In order to achieve effective scheduling of connection resources in high-concurrency scenarios and ensure the smooth progress of the business, an embodiment of this application provides a connection resource scheduling method. The method is applied to a cloud platform and includes the steps of: in response to a connection resource scheduling request, obtaining a scheduling delay factor; wherein, the connection resource scheduling request is used to trigger the reallocation of connection resources between connected terminals and to-be-connected terminals; determining a target resource allocation algorithm from multiple resource allocation algorithms based on the scheduling delay factor; forcibly disconnecting the connection of the connected terminal and listening for a reconnection request sent by the connected terminal; selecting a target terminal from the connected terminal and the to-be-connected terminal based on the attribute information of the reconnection request and the target resource allocation algorithm; and allocating the connection resource to the target terminal.

[0005] Based on the above technical solution, it is possible to reallocate connection resources in a high-concurrency scenario, thereby optimizing the utility of connection resources. It is possible to allocate the occupied but invalid connection resources to the to-be-connected terminals to a certain extent, relieve the queuing pressure, and achieve resource flow. At the same time, by determining the delay factor and selecting the resource allocation algorithm based on the delay factor, it is ensured that the reallocation process of resources will not have a negative impact on the connected terminals in the target terminal, and at the same time, the optimization processing duration is effectively controlled, so as to achieve user imperceptibility. Furthermore, by obtaining the reconnection request and selecting the target terminal based on the attributes of the reconnection request, the accuracy of data analysis can be guaranteed, and the actual network state can be approximated.

[0006] In one implementation, the method for obtaining the scheduling delay factor includes: determining the minimum response time for reconnection after disconnection according to the network state of the connected terminal as the scheduling delay factor.

[0007] In one implementation, determining the minimum response time for reconnection after disconnection according to the network state of the connected terminal includes: correcting the first transmission delay based on the state correction parameter to obtain the second transmission delay, and selecting the minimum value among the second transmission delays as the minimum response time for reconnection after disconnection; wherein, the first transmission delay is detected by the resource scheduling process for the network connections between the connected terminals and the cloud platform when determining that the number of to-be-connected terminals reaches the second threshold; the state correction parameter is determined by the resource scheduling process based on the data transmission volume per unit time of the network connections between the connected terminals and the cloud platform.

[0008] Based on the scheduling delay factor determined by the above technical solution, it is ensured that after forced disconnection, the reconnection request of the target terminal can be normally responded to maintain the network connection and reduce the impact on the services of the connected terminals.

[0009] In one implementation, the method for determining the target resource allocation algorithm from multiple resource allocation algorithms based on the scheduling delay factor includes: determining the total number of the connected terminals and the to-be-connected terminals; predicting the corresponding prediction calculation duration for each resource allocation algorithm based on the total number; selecting the resource allocation algorithm with a prediction calculation duration less than and the smallest difference from the scheduling delay factor as the target resource allocation algorithm.

[0010] In one implementation, the calculation intensities of the resource allocation algorithms are different.

[0011] In one implementation, selecting a target terminal from the connected terminals and the to-be-connected terminals based on the attribute information of the reconnection request and the target resource allocation algorithm includes: evaluating the activity status of each of the connected terminals based on the respective attribute information to obtain corresponding evaluation results; calculating the initial survival values of each of the connected terminals and each of the to-be-connected terminals based on the target resource allocation algorithm; correcting the initial survival values of each of the connected terminals according to the evaluation results to obtain the final survival values of each of the connected terminals; correcting the corresponding initial survival values based on the active status of each of the to-be-connected terminals to obtain the final survival values of each of the to-be-connected terminals; and selecting a target number of target terminals from the connected terminals and the to-be-connected terminals based on the respective final survival values.

[0012] Based on the above technical solution, the activity status of the connected terminals and the active status of the to-be-connected terminals are evaluated, and the evaluation results are used to correct the initial survival values calculated by the resource scheduling algorithm, so that the service utility and real-time activity status of the terminals are considered simultaneously in the process of determining the survival values of the terminals, which can largely ensure the effective utilization of connection resources.

[0013] In one implementation, allocating the available connection resources to the target terminals includes: responding to the reconnection requests of the connected terminals in the target terminals to restore the connections; and responding to the login requests of the to-be-connected terminals in the target terminals to establish connections with the to-be-connected terminals in the target terminals.

[0014] Based on the same inventive concept, an embodiment of the present application further provides a cloud platform, and the cloud platform schedules the connection resources of the platform based on the above method.

[0015] In addition, an embodiment of the present application further provides an electronic device, which includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, the above method is implemented.

[0016] Another embodiment of the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above method is implemented. Description of the Drawings

[0017] The drawings constituting a part of this application are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation to this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0019] Figure 1 The flowchart of the connection resource scheduling method provided by the embodiment of the present application is shown.

[0020] Figure 2 The flowchart of the method for obtaining the scheduling delay factor in the embodiment of the present application is shown.

[0021] Figure 3 The flowchart of the method for determining the target resource allocation algorithm in the embodiment of the present application is shown.

[0022] Figure 4 The flowchart of the method for selecting the target terminal in the embodiment of the present application is shown. Specific embodiments

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0024] In the description of the embodiments of the present application, unless otherwise specified, the meaning of "multiple" refers to two or more. The "first", "second" and various numerical numbers are only for the convenience of description and do not limit the scope of the embodiments of the present application.

[0025] The features, structures or characteristics in the present application can be combined in one or more embodiments in any suitable manner. In various embodiments of the present application, the size of the serial numbers of the processes does not mean the sequence of execution. The execution sequence of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0026] Some optional features in the embodiments of the present application, in some scenarios, can be implemented independently without relying on other features, solve the corresponding technical problems, achieve the corresponding effects, and can also be combined with other features according to requirements in some scenarios.

[0027] In this application, unless otherwise specified, the same or similar parts among various embodiments can be referred to each other. In each embodiment of this application, if there is no special specification and logical conflict, the terms and / or descriptions among different embodiments are consistent and can be cited mutually, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships. The embodiments of this application do not constitute a limitation on the protection scope of this application.

[0028] The embodiments of this application will be described in detail below with reference to the drawings.

[0029] The method provided by the embodiments of this application is applied to a cloud platform. In a specific example, the cloud platform can be a cloud management platform for managing Internet of Things terminal devices, or a cloud game platform, or other cloud service platforms that also need to effectively manage high-concurrency requests.

[0030] Please refer to Figure 1 , the connection resource scheduling method provided by the embodiments of this application specifically includes the following steps.

[0031] S101, in response to a connection resource scheduling request, obtain a scheduling delay factor.

[0032] Among them, the connection resource scheduling request is used to trigger the reallocation of connection resources between the connected terminals and the to-be-connected terminals.

[0033] Specifically, the cloud platform maintains a connection resource pool for managing connection resources. Among them, the number of connection resources in the resource pool can be the maximum number of connection resources that the cloud platform can provide determined according to the system load, that is, the maximum connection number, to avoid the situation of insufficient system resources after accessing too many terminals.

[0034] The cloud platform creates a connection processing process and a resource scheduling process.

[0035] When the connection processing process receives a connection request sent by a terminal, it first determines whether there is an idle connection resource in the resource pool. If there is, it directly responds to the connection request and allocates a connection resource for the connection request to establish a network connection with the terminal.

[0036] If there is no idle connection resource currently, the connection processing process writes the connection request into the queuing queue and waits until a connection resource is released, and then takes out the connection request from the queue for processing.

[0037] In a high-concurrency scenario, the number of connection requests in the queuing queue may surge. When the number of connection requests in the queuing queue exceeds the first threshold, the connection processing process automatically generates a connection resource scheduling request and sends it to the resource scheduling process to request the reallocation of connection resources.

[0038] After receiving a connection resource scheduling request, the resource scheduling process first obtains a scheduling delay factor, which is used to indicate the disconnection duration of the connected terminal, that is, after disconnecting the connection, a reconnection request can be responded to within the disconnection duration to restore the connection.

[0039] In one example, please refer to Figure 2 , and the method for obtaining the scheduling delay factor includes the following steps.

[0040] S201, obtain the connected terminals.

[0041] In implementation, the resource scheduling process can access the resource pool to obtain the allocation objects of each connection resource, and these allocation objects are the connected terminals. It can be understood that the number of connected terminals is the same as the number of connection resources in the resource pool.

[0042] S202, determine the network status of each connected terminal.

[0043] In implementation, the resource scheduling process monitors the queuing queue in real time. When it is determined that the number of connection requests in the queuing queue reaches the second threshold, the network status of each connected terminal is evaluated.

[0044] In one implementation, the resource scheduling process can detect the network transmission delay between the connected terminal and the cloud platform in real time, and determine the network status of the connected terminal based on the first transmission delay obtained by detection. Among them, the first transmission delay is determined based on the difference between the sending time of the detection message and the receiving time of the detection response message.

[0045] In another implementation, the resource scheduling process can detect the network connection between the connected terminal and the cloud platform to obtain the first transmission delay; obtain the historical interaction messages between the connected terminal and the cloud platform, count the total data volume of each interaction message and the connected duration to obtain a status correction parameter; correct the first transmission delay based on the status correction parameter to obtain the second transmission delay. Among them, the second transmission delay is used to characterize the network status of the connected terminal. The second threshold is less than the first threshold.

[0046] In a specific example, the resource scheduling process can calculate the data transmission volume per unit time based on the total data volume and the connected duration, and determine the status correction parameter based on the data transmission volume per unit time. Among them, the larger the data transmission volume per unit time, the larger the status correction parameter, and the correction parameter is greater than or equal to 1 and less than or equal to 1.2. Then, by dividing the first transmission delay by the status correction parameter, the second transmission delay is obtained, that is, for the connected terminal with a larger data transmission volume per unit time, the corresponding second transmission delay is smaller.

[0047] S203. Determine the minimum response time for reconnecting after disconnection based on the network status of the connected terminal, and use it as the scheduling delay factor.

[0048] Correspondingly, in one example, select the minimum value among each first transmission delay as the value of the scheduling delay factor.

[0049] In another example, select the minimum value among each second transmission delay as the value of the scheduling delay factor. Since the second transmission delay is the result of correcting the first transmission delay based on the status correction parameter, and when the data transmission volume per unit time is larger, the second transmission delay is reduced proportionally. In this way, it can be ensured to a certain extent that the more active connected terminals will not be unable to reconnect normally due to unreasonable values of the scheduling delay factor.

[0050] S102. Determine the target resource allocation algorithm from multiple resource allocation algorithms based on the scheduling delay factor.

[0051] Specifically, the resource allocation algorithm is used to calculate the survival value of each terminal to be calculated, so that the resource scheduling process can determine the resource allocation object according to the survival value. Among them, the survival value is used to reflect the utility that the resource allocation object can generate after being allocated connection resources. It can be understood that in different demand scenarios, the definition of utility is different.

[0052] For example, in the application of the Internet of Things, the relevant terminals of the key monitoring objects can generate greater utility. In the application of cloud games, the active players can generate greater utility. Therefore, the determination method of utility can be defined according to the actual application requirements, so as to set the corresponding resource allocation algorithm.

[0053] The cloud platform is pre-configured with multiple resource allocation algorithms. The differences between the resource allocation algorithms include different parameter ranges, different parameter acquisition difficulties, and different algorithm complexities. Therefore, the calculation intensities of the resource allocation algorithms are different. That is to say, under the same number of terminals to be calculated, the required calculation time is different.

[0054] In one example, the cloud platform is configured with two resource allocation algorithms. One resource allocation algorithm directly determines the initial survival value of each terminal to be calculated based on the priority of the terminal to be calculated. Among them, the priority is pre-configured according to the terminal attributes. For example, in the cloud game application scenario, the corresponding priority can be determined according to the user's membership level. In the Internet of Things application scenario, the corresponding priority can be determined according to the device type or monitoring task attributes. Therefore, the resource scheduling process can directly obtain the priorities of each terminal to be calculated, and based on the corresponding relationship between the priority and the survival value, output the initial survival value of each terminal to be calculated.

[0055] Another resource allocation algorithm includes obtaining parameter values of multiple terminals to be calculated in the same dimension, and respectively calculating corresponding initial survival values based on the comprehensive parameters in the same dimension. For example, in the application scenario of cloud games, the corresponding initial survival values can be calculated comprehensively based on parameters such as the historical activity, payment information, interaction records with other players, and game progress of players. In the application scenario of the Internet of Things, the corresponding initial survival values can be calculated comprehensively based on the operating parameters of terminal devices, service types, whether it is currently in the data reporting cycle, the amount of data to be sent, etc. Since the acquisition and comprehensive calculation of each parameter take a long time, there is an obvious difference in the calculation duration of the two algorithms. In a high-concurrency scenario, the difference in the total calculation duration will be amplified.

[0056] It can be understood that the specific algorithm can be selected according to the actual situation of the application scenario, and the present application is not limited thereto.

[0057] Please refer to Figure 3 , the method for determining the target resource allocation algorithm from multiple resource allocation algorithms based on the scheduling delay factor includes:

[0058] S301, determine the total number of connected terminals and terminals to be connected.

[0059] In implementation, the total number of connected terminals and terminals to be connected can be determined according to the number of connection resources and the number of connection requests in the queuing queue respectively.

[0060] S302, predict the corresponding predicted calculation duration in each resource allocation algorithm based on the total number.

[0061] In one implementation, the calculation duration of each resource allocation algorithm under different preset data amounts can be obtained in advance through multiple tests, so as to establish an association table to record the calculation duration of each preset data amount under different resource allocation algorithms. The value range of the preset data amount includes the sum of the connection resource number and the first threshold as the minimum value, the sum of the connection resource number and the maximum capacity of the queuing queue as the maximum value, and the intermediate node value is determined according to the historical highest queuing data. Or, the intermediate node value is determined according to a fixed step size.

[0062] The resource scheduling process can match the total number with the preset numbers in the association table, match the target preset number that is not less than the total number and has the smallest difference from the total number, and use the calculation duration corresponding to the target preset data under different resource allocation algorithms as the predicted calculation duration.

[0063] S303, select the resource allocation algorithm whose predicted calculation duration is less than and has the smallest difference from the scheduling delay factor as the target resource allocation algorithm.

[0064] The target resource allocation algorithm determined based on the above method can ensure that the calculation is completed in advance within the time range limited by the scheduling extension factor.

[0065] S103. Forcefully disconnect the connection of the connected terminal and listen for the reconnection request sent by the connected terminal.

[0066] After determining the target resource allocation algorithm, simultaneously forcefully disconnect the network connections of all connected terminals from the cloud platform and monitor each connection port to receive the reconnection requests sent by each connected terminal.

[0067] S104. Select a target terminal from the connected terminals and the to-be-connected terminals based on the attribute information of the reconnection request and the target resource allocation algorithm.

[0068] Among them, the attribute information of the reconnection request includes the sending frequency.

[0069] It should be noted that while forcefully disconnecting the connected terminal, start the calculation based on the target resource allocation algorithm. The start node of the listening period is the time of forced disconnection, and the end node is the time when the calculation is completed.

[0070] In an example, please refer to Figure 4 , the method for selecting a target terminal from the connected terminals and the to-be-connected terminals based on the attribute information of the reconnection request and the target resource allocation algorithm includes the following steps.

[0071] S401. Evaluate the activity status of each connected terminal based on each attribute information to obtain the corresponding evaluation result.

[0072] In implementation, a reconnection control service is pre-implemented on each terminal to control the number of times of sending reconnection requests when the terminal discovers that the connection is disconnected.

[0073] In an example, the method for the reconnection control service to monitor the network connection status includes determining whether the network is disconnected based on the response status of normal data packets. For example, if the heartbeat packet cannot be received normally or the cloud platform does not respond to the service request, it is determined that the network connection is disconnected. In another example, before forcefully disconnecting the connection, the resource scheduling service first sends a connection disconnection notice to each connected terminal. After the reconnection control service monitors the connection disconnection notice, it can determine that the network connection has been disconnected. In some other examples, the monitoring of the network connection status can also be implemented by combining the above two examples.

[0074] When it is monitored that the network connection is disconnected, the reconnection control service determines the sending frequency of the reconnection request based on the demand degree of the interaction between the terminal and the cloud platform. Among them, the greater the demand degree, the higher the sending frequency.

[0075] In one example, the method for determining the demand degree includes obtaining the activity of the service process interacting with the cloud platform on the terminal to determine the demand degree. The higher the activity, the higher the corresponding demand degree. Specifically, one or more of the performance metrics such as the CPU occupancy rate, memory usage, disk I / O, and network I / O of the service process can be obtained to evaluate the activity of the service process.

[0076] For example, the corresponding threshold ranges are determined in advance according to the CPU occupancy rates of the service process in different active states such as idle, normal, and busy. Then, according to the matching relationship between the real-time obtained CPU occupancy rate and the threshold ranges, the current active state of the service process is determined. Furthermore, according to the corresponding relationship between the active state and the demand degree, the current demand degree of the service process is determined. Then, the sending frequency of the reconnection request corresponding to the demand degree is set. It can be understood that the corresponding relationship between the active state and the demand degree can be pre-configured on the terminal according to the service attributes of the terminal. Similarly, the corresponding relationship between the demand degree and the sending frequency of the reconnection request is also pre-configured on the terminal.

[0077] It should be noted that the corresponding relationship between the active state and the demand degree is set personalized according to different terminal situations, and the corresponding relationship between the demand degree and the sending frequency is set uniformly based on the global devices. In this way, not only can the resource scheduling process uniformly evaluate the terminals according to the sending frequency, but also the personalized configuration of terminals with different importance levels can be realized to balance the evaluation results of the terminals. Generally speaking, for terminals with a higher importance level, the demand degree corresponding to the same active state is higher than that of terminals with a lower importance level. Based on the same corresponding relationship between the demand degree and the sending frequency, the determined sending frequency is also higher. Among them, the importance level of the terminal can be determined according to the actual application requirements. For example, in the Internet of Things application scenario, the importance level can be determined according to the device type, monitoring object, data collection period, etc. of the terminal; in the cloud game scenario, the importance level of the corresponding terminal can be determined according to the identity of the currently logged-in player, historical operation records, etc.

[0078] It can be understood that in the actual application process, the relevant corresponding relationships can be adjusted according to the business requirements, and this application is not limited thereto.

[0079] The reconnection control service sends a reconnection request to the cloud platform based on the determined sending frequency.

[0080] The resource scheduling process receives reconnection requests by listening on the corresponding port, determines the sending frequency within the listening duration, and evaluates the activity status of the connected terminals based on the sending frequency to obtain an evaluation result. Among them, the higher the sending frequency, the better the corresponding activity status, and the better the corresponding evaluation result. It should be noted that since there are differences in the connected terminals every time the resources are reallocated, when determining the activity status of each connected terminal, it is determined based on the relative relationship of each sending frequency. For example, the sending frequencies can be sorted, and the corresponding order ranking can be used as the value of the corresponding activity status, and then the evaluation result is obtained. The value setting method of the evaluation result can be set according to actual needs.

[0081] Specifically, the sending frequency can be determined according to the actually received reconnection requests.

[0082] In one example, when the number of actually received reconnection requests is greater than 1, the minimum time interval between two reconnection requests is determined as the sending frequency; when the actually received number is 1, the difference between the reception time of the reconnection request and the listening end time is first determined. If the difference is greater than the maximum time interval determined based on the actual reconnection request, the difference is used as the sending interval to calculate the corresponding sending frequency; if the difference is not greater than the above maximum time interval, the maximum time interval is used as the sending interval to calculate the corresponding sending frequency. Based on this, corresponding values can be reasonably set for the terminals whose sending frequencies cannot be determined, providing a relatively accurate data basis for the subsequent calculation process.

[0083] If the actually received reconnection request is equal to 0, the resource scheduling process directly sets the corresponding sending frequency, that is, when it is determined that due to network transmission conditions, no reconnection request can be received within the listening duration, the resource scheduling process can directly set the corresponding sending frequency according to a preset method.

[0084] In one example, the minimum value among the sending frequencies determined according to the actually received reconnection requests can be used as the sending frequency of each terminal that cannot receive a reconnection request within the listening duration, so as to uniformly deteriorate the processing of the terminals with too long first transmission delay.

[0085] In another example, the smaller value of the average and median of each actual sending frequency can be used as the base value, and the base value is reduced in proportion according to the size of each first transmission delay, and then used as the sending frequency of the corresponding connected terminal. Among them, the longer the first transmission delay, the lower the configured value obtained after compensation. In this way, a hierarchical deterioration process is performed on the terminals with too long first transmission delay.

[0086] Based on the above method, by degrading the terminals with too long first transmission delay, it is possible to ensure to a certain extent that the terminals with better communication quality obtain more resource tilts, thereby reducing the allocation of resources to terminals with uncertain evaluation results and causing unreasonable resource allocation.

[0087] In other implementations, when the actual number of reconnection requests is equal to 0, and the number of reconnection requests that should be received within the listening duration estimated based on the first transmission delay is greater than 0, it is determined that the corresponding terminal communication is abnormal, and it can be directly excluded from this reallocation to release connection resources.

[0088] S402, Calculate the initial survival values of each connected terminal and each terminal to be connected based on the target resource allocation algorithm.

[0089] S403, Modify the initial survival values of each connected terminal according to the evaluation results to obtain the final survival values of each connected terminal; modify the corresponding initial survival values based on the active states of each terminal to be connected to obtain the final survival values of each terminal to be connected.

[0090] In an implementation, the evaluation result value corresponding to the connected terminal can be used to perform a secondary calculation on the initial survival value, and the calculation result is used as the final survival value. The principle of modification is that the better the active state, the higher the value of the final survival value. In an example, the sum of the evaluation result value and the initial survival value can be used as the final survival value. It can be understood that a weight value can also be introduced during the calculation process to adjust the calculation result.

[0091] In one implementation, the resource scheduling process can also modify the corresponding initial survival values based on the active states of each terminal to be connected. The principle of modification is to determine the corresponding active state according to the number of request transmissions of each terminal to be connected. The more the number of transmissions and the more active, the higher the final survival value obtained by modification.

[0092] S404, Select a target number of target terminals from the connected terminals and the terminals to be connected based on each final survival value.

[0093] In an implementation, sorting can be performed based on the final survival value, and the top target number of terminals are selected as the target terminals, where the target number is the weight of the connection resources in the resource pool.

[0094] S105, Allocate connection resources to the target terminals.

[0095] In an implementation, the method of allocating connection resources to the target terminals includes: responding to the reconnection requests of the connected terminals in the target terminals to restore the connection; responding to the login requests of the terminals to be connected in the target terminals to establish a connection with the terminals to be connected in the target terminals.

[0096] Based on the connection resource scheduling method provided by the embodiments of the present application, by forcibly disconnecting the connected terminals and completing the reallocation of connection resources within the time limit defined by the delay factor, in a high-concurrency scenario, by reselecting the target terminals, to a certain extent, the occupied connection resources with lower utility are forcibly released for the to-be-connected terminals that can generate greater utility to use. This can not only achieve the effective flow of connection resources, relieve the queuing pressure, but also provide resource utility from the overall operation level.

[0097] Furthermore, by reasonably determining the delay factor and selecting a resource allocation algorithm with corresponding computing intensity, it can be ensured that the target terminals can be successfully reconnected after being determined, and it can adapt to high-concurrency scenarios under different network conditions.

[0098] In addition, the embodiments of the present application also provide an electronic device, which includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, it implements the method in any implementation manner in the embodiments of the present application; among them, the processor can adopt a general-purpose central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), a graphics processing unit (GPU), or one or more integrated circuits, and is used to execute relevant programs to implement the method in any implementation manner in the embodiments of the present application.

[0099] The processor can also be an integrated circuit electronic device with signal processing capabilities. In the implementation process, each step of the method in any implementation manner in the embodiments of the present application can be completed by the integrated logic circuit in the processor or the instruction in software form.

[0100] The above-mentioned processor can also be a general-purpose processor, a digital signal processor, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by the combination of the hardware and software modules in the decoding processor.

[0101] The software module may be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the functions required to be executed by the units included in the data processing device of the embodiments of the present application, or executes the method in any implementation manner of the embodiments of the present application.

[0102] Another embodiment of the present application relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the embodiments of the above method are implemented.

[0103] Those skilled in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program is stored in a storage medium and includes several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the methods of the various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks, or optical discs.

[0104] The above are all the preferred embodiments of the present application. The protection scope of the present application is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application shall be covered within the protection scope of the present application.

Claims

1. A connection resource scheduling method, characterized in that, Applied to a cloud platform, the method includes the steps of: In response to a connection resource scheduling request, obtain a scheduling delay factor; wherein, the connection resource scheduling request is used to trigger re - allocation of connection resources between connected terminals and to - be - connected terminals; Determine a target resource allocation algorithm from multiple resource allocation algorithms based on the scheduling delay factor; Forcibly disconnect the connection of the connected terminal and listen for a re - connection request sent by the connected terminal; Select a target terminal from the connected terminal and the to - be - connected terminal based on the attribute information of the re - connection request and the target resource allocation algorithm; Allocate the connection resources to the target terminal; The method for obtaining the scheduling delay factor includes: Determine the minimum response time for re - connection after disconnection according to the network state of the connected terminal as the scheduling delay factor; The determining the minimum response time for re - connection after disconnection according to the network state of the connected terminal includes: Based on a state correction parameter, correct the first transmission delay to obtain a second transmission delay, select the minimum value among the second transmission delays as the minimum response time for re - connection after disconnection; wherein, the first transmission delay is obtained by the resource scheduling process detecting the network connections between each connected terminal and the cloud platform when determining that the number of to - be - connected terminals reaches a second threshold; the state correction parameter is determined by the resource scheduling process based on the data transmission volume per unit time of each connected terminal and the cloud platform based on the network connection; The manner of determining a target resource allocation algorithm from multiple resource allocation algorithms based on the scheduling delay factor includes: Determine the total number of the connected terminal and the to - be - connected terminal; Predict the corresponding prediction calculation duration in each resource allocation algorithm based on the total number; Select the resource allocation algorithm with a prediction calculation duration less than and having the smallest difference from the scheduling delay factor as the target resource allocation algorithm.

2. The method according to claim 1, wherein The calculation intensities of the resource allocation algorithms are different.

3. The method according to claim 1, characterized in that, The selecting a target terminal from the connected terminal and the to - be - connected terminal based on the attribute information of the re - connection request and the target resource allocation algorithm includes: Based on each piece of attribute information, evaluate the activity state of each connected terminal to obtain corresponding evaluation results; Calculate the initial survival values of each connected terminal and each to - be - connected terminal based on the target resource allocation algorithm; Correct the initial survival values of each connected terminal according to the evaluation results to obtain the final survival values of each connected terminal; correct the corresponding initial survival values based on the active state of each to - be - connected terminal to obtain the final survival values of each to - be - connected terminal; Based on each of the final survival values, select a target number of target terminals from the connected terminal and the to - be - connected terminal.

4. The method according to claim 1, wherein The allocating the connection resources to the target terminal includes: Respond to the re - connection requests of the connected terminals among the target terminals to restore the connection; Respond to the login request of the terminal to be connected in the target terminal to establish a connection with the terminal to be connected in the target terminal.

5. A cloud platform, characterized in that, The cloud platform schedules the connection resources of the platform based on the method described in any one of claims 1 to 4.

6. An electronic device, characterized in that, It includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, the method described in any one of claims 1-4 is implemented.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the method described in any one of claims 1 to 4 is implemented.

Images