A cloud server service scheduling method and device

By building weighted call graphs and filtering the optimal edge cloud or central cloud in the cloud server system, the data path is optimized, and the inefficiency and resource consumption of mobile cloud computing and mobile edge computing under massive service requests and high service quality requirements is solved. , realize more efficient service scheduling and resource utilization.

CN116866356BActive Publication Date: 2024-11-22SHANDONG ZHITUO BIG DATA CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311081128.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-08-25
Publication Date
2024-11-22
Estimated Expiration
2043-08-25

AI Technical Summary

Technical Problem

When existing mobile cloud computing and mobile edge computing face massive service requests and high service quality requirements, there are problems of increased transmission costs and delays, and consume additional computing resources and wireless bandwidth resources during task offloading, which is difficult to Services that adapt to task requests and low latency requirements for a large number of devices.

Method used

Provide a cloud server service scheduling method, by soot conversion and building weighted call graphs for terminal-initiated applications, filtering the optimal local edge cloud, and calculating processing costs based on distance, energy consumption data and time request arrival quantity, selecting the optimal Edge cloud or central cloud is used to optimize data paths and reduce transmission and processing time overhead.

Benefits of technology

It improves the efficiency of cloud servers in data upload, processing and download, reduces network load and resource consumption, and adapts to the task requests of massive devices and service requirements of low latency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116866356B_ABST
    Figure CN116866356B_ABST
Patent Text Reader

Abstract

The present application relates to the field of cloud server management technology, which is a cloud server management technology, and specifically to a cloud server service scheduling method and device; the method includes: performing soot conversion on an application initiated by a terminal and constructing a weighted call graph of the application; obtaining the distance from a microservice to multiple local edge clouds, energy consumption data of multiple local edge clouds, and the average time request arrival number of multiple local edge cloud deployed microservices, and obtaining multiple processing costs of multiple local edge cloud deployed microservices based on the distance and the arrival number, and comparing the minimum of multiple processing costs to obtain the optimal edge cloud; obtaining multiple time costs of the optimal edge cloud and multiple remote edge clouds and a central cloud, comparing the minimum of multiple time costs to obtain a target cloud node, and using the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud as the data path of the service, and using the data paths corresponding to multiple microservices as the scheduling path.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of cloud server management technology, which is a cloud server management technology, and specifically to a cloud server service scheduling method and device. Background Art

[0002] Mobile Cloud Computing (MCC) has been widely regarded as an important key basic technology to solve the current problems of serious occupation and restriction of data resources of mobile terminal devices due to its powerful mobile data collection computing and information storage technology. One of the core service ideas of MCC is to offload the high-intensity tasks existing on smart phone mobile terminal devices and place them on new cloud computing servers for task execution, so as to effectively solve the service problem of limited computing resources. However, when faced with massive service requests and requests with high service quality requirements, MCC presents some shortcomings. Since the physical communication distance between smart mobile devices and cloud servers is long, the transmission cost and delay of service requests increase. In addition, the massive requests arriving at the centralized cloud server will cause a surge in the load of the core network, resulting in network congestion. Therefore, the existing MCC cannot be applied to the task requests of massive devices and services with low latency requirements such as unmanned driving.

[0003] In MEC, in order to improve their own computing power, smart mobile devices will offload workloads to edge servers for processing. In recent years, Mobile Edge Computing Offloading (MECO) has been regarded as an effective solution to deal with its problems. By offloading complex tasks to edge servers with stronger processing power and higher utility, smart mobile devices can complete task processing efficiently. Although MEC can reduce the burden of smart mobile devices in processing complex tasks, it consumes additional computing resources and additional wireless bandwidth resources in the edge server during the task offloading process. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a cloud server service scheduling method and device, which can solve the problem of low efficiency of cloud servers in data uploading, processing and downloading.

[0005] In order to achieve the above purpose, the technical solution adopted in the embodiment of the present application is as follows:

[0006] In a first aspect, a cloud server service scheduling method is provided, the method is applied to a cloud service system, the cloud service system communicates with a terminal, the cloud service system includes multiple edge clouds and a central cloud, the terminal communicates with the edge cloud, the multiple edge clouds communicate with the central cloud, the multiple edge clouds include multiple local edge clouds and multiple remote edge clouds, the local edge cloud is used to schedule an application initiated by the terminal to the remote edge cloud or the central cloud, and send the processing result of the remote edge cloud or the central cloud to the terminal; the method includes: performing soot conversion on the application initiated by the terminal and constructing a weighted call graph of the application, the weighted call graph is used to characterize multiple microservices in the application relationship between; obtaining the distance from the microservice to multiple local edge clouds, the energy consumption data of multiple local edge clouds, and the average time request arrival number of multiple local edge clouds deploying microservices, and obtaining multiple processing costs of multiple local edge clouds deploying microservices based on the distance and the arrival number, and comparing the smallest of the multiple processing costs to obtain the optimal edge cloud; obtaining multiple time costs of the optimal edge cloud and multiple remote edge clouds and central clouds, and comparing the smallest of the multiple time costs to obtain the target cloud node, and using the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud as the data path of the microservice, and using the data paths corresponding to the multiple microservices as the scheduling paths.

[0007] Furthermore, the application initiated by the terminal is soot converted and a weighted call graph of the application is constructed, including: converting the contents of multiple classes in the application to obtain multiple class representations, and obtaining a graph model based on the transmission relationship of the class representations; and assigning weights to the multiple class representations to obtain weights for each class representation, and weighting the graph model based on the weights to obtain the weighted call graph.

[0008] Furthermore, the processing cost is expressed based on the following formula: ;in is the energy consumption data of the local edge cloud, The average time for request arrival of the microservice deployed on the local edge cloud, The distance from the microservice to the multiple local edge clouds.

[0009] Furthermore, the energy consumption data of the local edge cloud is determined based on the following formula: , where v represents the exponential parameter and v ≥ 1, (1-α) represents the standard energy consumption of the idle local edge cloud and ,in Represents the scheduling function, which is expressed by the following formula: ; The optimization function is expressed based on the following formula: ,in The queue backlog at time t is expressed based on the following formula: , N represents the number of queue backlogs; where i represents the i-th microservice, and j represents the j-th local edge cloud.

[0010] Furthermore, the average time request arrival number of the local edge cloud deployed microservice is determined based on the following formula: , where T is the unit time number of the time series, and j represents the local edge cloud; The relationship between microservice type i and the local edge cloud is expressed based on the following formula: .

[0011] Furthermore, the time cost includes transmission time cost, processing time cost and feedback time cost.

[0012] Furthermore, the time cost of the optimal edge cloud and the multiple remote edge clouds is expressed based on the following formula: , where te represents the remote edge cloud, For input data, is the data transmission rate relative to the remote edge cloud; the time cost of the optimal edge cloud and the central cloud is expressed based on the following formula: ,in represents the central cloud, For input data, is the data transmission rate relative to the central cloud.

[0013] Furthermore, the processing time cost of the multiple remote edge clouds is determined based on the following formula: ,in represents the computational workload, Represents the computing power of the remote edge cloud; the processing time cost of the central cloud is determined based on the following formula: ,in Indicates the computing power of the central cloud.

[0014] Furthermore, the feedback time cost of the remote edge cloud is determined based on the following formula: ,in Indicates the amount of data transferred. is the data transmission rate relative to the remote edge cloud; the feedback time cost of the central cloud is determined based on the following formula: ,in is the data transmission rate relative to the central cloud.

[0015] In a second aspect, a cloud server service scheduling device is provided, the device is applied to a cloud service system, the cloud service system communicates with a terminal, the cloud service system includes multiple edge clouds and a central cloud, the terminal communicates with the edge cloud, the multiple edge clouds communicate with the central cloud, the multiple edge clouds include multiple local edge clouds and multiple remote edge clouds, the local edge cloud is used to schedule an application initiated by the terminal to the remote edge cloud or the central cloud, and send the processing result of the remote edge cloud or the central cloud to the terminal; the device includes: a splitting module, used to perform soot conversion on the application initiated by the terminal and construct a weighted call graph of the application, the weighted call graph is used to characterize the relationship between multiple microservices in the application; The optimal edge cloud determination module is used to obtain the distance from the microservice to multiple local edge clouds, the energy consumption data of multiple local edge clouds, and the average time request arrival number of multiple local edge cloud deployed microservices, and based on the distance and the arrival number, obtain multiple processing costs of multiple local edge cloud deployed microservices, and compare the smallest of the multiple processing costs to obtain the optimal edge cloud; the final scheduling module obtains multiple time costs of the optimal edge cloud and multiple remote edge clouds and central clouds, compares the smallest of the multiple time costs to obtain the target cloud node, and uses the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud as the data path of the microservice, and uses the data paths corresponding to the multiple microservices as the scheduling paths.

[0016] According to a third aspect, a terminal device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-described methods when executing the computer program.

[0017] According to a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of the above claims is implemented.

[0018] In the technical solution provided in the embodiment of the present application, the optimal local edge cloud is obtained as the local server by screening multiple local edge clouds in the server system, and the best edge cloud or central cloud corresponding to the optimal time cost is obtained by calculating the comprehensive time cost. The best edge cloud or central cloud and the local edge cloud that are finally determined constitute a scheduling path to achieve overall scheduling of the service. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] The methods, systems and / or programs in the accompanying drawings will be further described according to exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein example numbers represent similar mechanisms in the various views of the accompanying drawings.

[0021] Figure 1 It is a flow chart of the cloud server service scheduling method provided by an embodiment of the present application;

[0022] Figure 2 It is a structural diagram of a cloud server service scheduling device provided in an embodiment of the present application;

[0023] Figure 3 It is a schematic diagram of the cloud server scheduling device structure provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0025] In the following detailed description, numerous specific details are set forth by way of example in order to provide a comprehensive understanding of the relevant guidance. However, it will be apparent to those skilled in the art that the present application may be practiced without these details. In other cases, well-known methods, procedures, systems, compositions and / or circuits have been described at a relatively high level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.

[0026] Flowcharts are used in the present application to illustrate the execution process performed by the system according to the embodiment of the present application. It should be clearly understood that the execution process of the flowchart may not be performed in order. On the contrary, these execution processes may be performed in reverse order or simultaneously. In addition, at least one other execution process may be added to the flowchart. One or more execution processes may be deleted from the flowchart.

[0027] Before further describing the embodiments of the present invention in detail, the nouns and terms involved in the embodiments of the present invention are described. The nouns and terms involved in the embodiments of the present invention are subject to the following interpretations.

[0028] (1) In response to, it is used to indicate the conditions or states on which the executed operation depends. When the dependent conditions or states are met, one or more operations executed may be in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed are executed.

[0029] (2) Based on: used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed may be in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed are executed.

[0030] The cloud server service scheduling method and device provided in the embodiment of the present application are applied to a server system, wherein the server system communicates with a terminal, and the terminal sends a service or application to the server system for data processing. The server system used in the embodiment of the present application includes a local edge cloud, a remote edge cloud, and a central cloud, wherein the local edge cloud is used to schedule an application initiated by the terminal to the remote edge cloud or the central cloud, and send the processing result of the remote edge cloud or the central cloud to the terminal.

[0031] The basic concept of Mobile Edge Computing (MEC). As a new paradigm, MEC has been increasingly used to adapt to these needs. The core idea of ​​MEC is to move computing resources to the logical edge of the network, so that mobile users can access the platform through WiFi or cellular networks. This provides edge users with fast and powerful computing, energy efficiency, storage capacity, mobility, location and context-aware support. Today, such edge service provision scenarios are no longer just a concept, but a reality.

[0032] In MEC, smart mobile devices offload tasks to nearby access nodes through WiFi or cellular networks, and then the access nodes schedule the corresponding tasks to the appropriate MEC servers for processing, and return the processing results to the smart mobile devices after the processing is completed. Among them, how to offload user tasks more safely and efficiently and schedule tasks to appropriate MEC servers has become a hot topic in MEC research. In terms of security, trust plays an important role in all commercial cloud environments, and trust management is an indispensable part of MEC business. In MEC, service providers of the same type will deploy their services on different edge servers for users to choose. Generally speaking, users of smart mobile devices tend to choose services with higher quality of service (QoS). A dishonest service provider may provide false QoS values ​​to attract more service requests and thus obtain higher revenue. Therefore, it is necessary to design a scheduling algorithm with trust guarantee. In terms of efficiency, in MEC, smart mobile devices offload tasks to access nodes, and then the access nodes schedule them to the corresponding edge servers according to demand. In this case, there may be a situation where task offloading and task scheduling are not coordinated. Generally speaking, WiFi has a higher transmission rate and a corresponding lower overhead. Therefore, if WiFi is available, smart mobile devices prefer to offload tasks through WiFi. However, due to the characteristics of user channels and mobility, continuous connection to WiFi is not feasible. Therefore, in order to maximize the offloading utility of the system and use WiFi as much as possible, some task loads will be backlogged in the queue. However, in task scheduling, since some tasks are backlogged in the offloading queue, the computing resources of the MEC server are redundant, resulting in resource waste.

[0033] In MEC, in order to improve their own computing power, smart mobile devices will offload workloads to edge servers for processing. In recent years, Mobile Edge Computing Offloading (MECO) has been regarded as an effective solution to deal with its problems. By offloading complex tasks to edge servers with stronger processing power and higher utility, smart mobile devices can complete task processing efficiently. Although MEC can reduce the burden of smart mobile devices in processing complex tasks, it will consume additional computing resources and additional wireless bandwidth resources in the edge server during the task offloading process. In response to these resource consumption issues, existing research has optimized them from the aspects of minimizing energy consumption, minimizing latency, and balancing energy consumption and latency.

[0034] Therefore, in order to balance the different impacts of the application service's computing program being executed in different locations, a reasonable application service scheduling strategy must be used to determine whether the service is executed on a local server or on a mobile or remote server. In the past research on edge computing service scheduling, a coarse-grained service scheduling method was often used. This coarse-grained scheduling takes the entire mobile application as the scheduling object, does not divide the application according to its function, but directly schedules the entire application service to the target server. This method usually does not consider the transmission delay and transmission consumption of the entire application. However, with the continuous development of distributed program architecture, current applications and services are no longer limited to monolithic architectures, nor are they limited to running on a single server. Instead, an application is decoupled into multiple microservices and data dependencies. The divided microservices require less computational complexity and data transmission when executing, and can also be executed on different servers at the same time, which can well alleviate the computing pressure of a single server. The emergence of this architecture also provides a new idea for fine-grained scheduling for the scheduling of application services. Some or all microservices can be scheduled to multiple mobile servers or remote servers for processing, saving computing time and transmission time, and improving the resource utilization of computing server clusters.

[0035] At present, in the application partition research based on microservice architecture, most of them are considered in the cloud computing environment. When microservices are divided into different partitions, they are all executed inside the cloud center. Compared with the scheduling in edge computing, the time loss of data transmission and communication between microservices during execution is often ignored. At present, most of the research based on edge computing is still based on coarse-grained scheduling. In the selection of scheduling targets, only the cloud center and local edge devices are considered. The coarse-grained scheduling method is also not suitable for new edge computing scenarios that introduce a large number of lightweight computing nodes, such as drone-assisted edge computing scenarios. Based on the actual situation of edge computing and the microservice architecture, this paper studies how to reduce the time overhead of services and improve service efficiency through reasonable application partitioning algorithms and scheduling strategies.

[0036] In view of the above background information, an embodiment of the present application provides a cloud server service scheduling method, which specifically includes the following methods:

[0037] Step S110: Perform soot conversion on the application initiated by the terminal and construct a weighted call graph of the application.

[0038] In an embodiment of the present application, for service scheduling, applications initiated by the terminal are first classified, and a weighted call graph is constructed according to the classification results, wherein the weighted call graph is used to characterize the relationship between multiple microservices in the application.

[0039] The process of obtaining the weighted call graph is mainly implemented through partitioning. The contents of multiple classes in the application are converted to obtain multiple class representations, and a graph model is obtained based on the transmission relationship of the class representations. The multiple class representations are assigned weights for each class representation, and the graph model is weighted based on the weights to obtain the weighted call graph.

[0040] Step S120. Obtain the distance from the microservice to the multiple local edge clouds, the energy consumption data of the multiple local edge clouds, and the average time request arrival number of the multiple local edge clouds deploying microservices, and obtain multiple processing costs of the multiple local edge clouds deploying microservices based on the distance and the arrival number, and compare the minimum of the multiple processing costs to obtain the optimal edge cloud.

[0041] This step mainly involves screening local edge clouds to select the optimal local edge cloud for processing the service.

[0042] The processing cost is determined based on the following formula:

[0043] ;in is the energy consumption data of the local edge cloud, The average time for request arrival of the microservice deployed on the local edge cloud, The distance from the microservice to the multiple local edge clouds.

[0044] Specifically, the energy consumption data for the local edge cloud is determined based on the following formula:

[0045] , where v represents the exponential parameter and v ≥ 1, (1-α) represents the standard energy consumption of the idle local edge cloud and ,in Represents the scheduling function, which is expressed by the following formula:

[0046] ; The optimization function is expressed based on the following formula:

[0047] ,in The queue backlog at time t is expressed based on the following formula:

[0048] , N represents the number of queue backlogs;

[0049] Among them, i represents the i-th microservice, and j represents the j-th local edge cloud.

[0050] The specific average time request arrival number for local edge cloud deployment microservices is determined based on the following formula:

[0051] , where T is the unit time number of the time series, and j represents the local edge cloud;

[0052] The relationship between microservice type i and the local edge cloud is expressed based on the following formula:

[0053] .

[0054] Step S130. Obtain multiple time costs of the optimal edge cloud and multiple remote edge clouds and central clouds, compare the multiple time costs and select the target cloud node with the smallest one, and use the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud as the data path of the microservice, and use the data paths corresponding to the multiple microservices as scheduling paths.

[0055] After the optimal local edge cloud is obtained in step S120, the subsequent edge processing is performed by the remote edge cloud or the central cloud, and how to obtain the optimal processing unit in multiple remote edge clouds and the central cloud is compared by comparing the time cost of the remote edge cloud and the central cloud. In the embodiment of the present application, the time cost includes the transmission time cost, the processing time cost and the feedback time cost.

[0056] During the processing process, it is the transmission time cost of data transmission from the optimal local edge cloud to the remote edge cloud or central cloud, the processing time cost of the remote edge cloud or central cloud, and the transmission time cost of data transmission from the remote edge cloud or central cloud to the optimal local edge cloud.

[0057] The specific acquisition of the above time cost is determined based on the following formula:

[0058] The time cost for the optimal edge cloud and the multiple remote edge clouds is expressed based on the following formula:

[0059] , where te represents the remote edge cloud, For input data, is the data transmission rate relative to the remote edge cloud.

[0060] The time cost for the optimal edge cloud and the central cloud is expressed based on the following formula:

[0061] ,in represents the central cloud, For input data, is the data transmission rate relative to the central cloud.

[0062] The processing time cost for the remote edge cloud is determined based on the following formula:

[0063] ,in represents the computational workload, Indicates the computing power of the remote edge cloud.

[0064] The processing time cost for the central cloud is determined based on the following formula:

[0065] ,in Indicates the computing power of the central cloud.

[0066] The feedback time cost for the remote edge cloud is determined based on the following formula:

[0067] ,in Indicates the amount of data transferred. is the data transmission rate relative to the remote edge cloud.

[0068] The feedback time cost for the central cloud is determined based on the following formula:

[0069] ,in is the data transmission rate relative to the central cloud.

[0070] The minimum cost among the above time costs is the optimal edge cloud unit. Using time as the judgment dimension reduces the difficulty of evaluation and improves the efficiency of evaluation.

[0071] See also Figure 2 , a cloud server service scheduling device 200 is provided, wherein the device comprises:

[0072] A splitting module 210 is used to perform soot conversion on the application initiated by the terminal and construct a weighted call graph of the application, wherein the weighted call graph is used to characterize the relationship between multiple microservices in the application;

[0073] The optimal edge cloud determination module 220 is used to obtain the distance from the microservice to the multiple local edge clouds, the energy consumption data of the multiple local edge clouds, and the average time request arrival number of the multiple local edge clouds deploying microservices, and obtain multiple processing costs of the multiple local edge clouds deploying microservices based on the distance and the arrival number, and compare the minimum of the multiple processing costs to be the optimal edge cloud;

[0074] The final scheduling module 230 obtains multiple time costs of the optimal edge cloud and multiple remote edge clouds and central clouds, compares the minimum of the multiple time costs as the target cloud node, and uses the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud as the data path of the microservice, and uses the data paths corresponding to the multiple microservices as the scheduling paths.

[0075] See also Figure 3 , the above method can also be integrated into the provided cloud server service scheduling device 300. In view of the fact that the device may have relatively large differences due to different configurations or performances, it can include one or more processors 301 and memory 302, and the memory 302 can store one or more storage applications or data. Among them, the memory 302 can be a short-term storage or a persistent storage. The application stored in the memory 302 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the cloud server service scheduling device. Furthermore, the processor 301 can be configured to communicate with the memory 302, and the cloud server service scheduling device executes a series of computer executable instructions in the memory 302. The cloud server service scheduling device may also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input / output interfaces 305, one or more keyboards 306, etc.

[0076] In a specific embodiment, the cloud server service scheduling device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions in the cloud server service scheduling device, and the one or more programs are configured to be executed by one or more processors, including computer executable instructions for performing the following:

[0077] Performing soot conversion on the application initiated by the terminal and constructing a weighted call graph of the application;

[0078] Obtain the distance from the microservice to the multiple local edge clouds, the energy consumption data of the multiple local edge clouds, and the average time request arrival number of the multiple local edge clouds deploying microservices, and obtain multiple processing costs of the multiple local edge clouds deploying microservices based on the distance and the arrival number, and compare the multiple processing costs to find the optimal edge cloud;

[0079] Obtain multiple time costs of the optimal edge cloud and multiple remote edge clouds and central clouds, compare the multiple time costs and take the target cloud node as the smallest one, and use the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud as the data path of the microservice, and use the data paths corresponding to the multiple microservices as the scheduling paths.

[0080] The following is a detailed introduction to the various components of the processor:

[0081] In this embodiment, the processor is an application specific integrated circuit (ASIC), or is configured to implement one or more integrated circuits of the embodiments of the present application, such as one or more microprocessors (digital signal processor, DSP), or one or more field programmable gate arrays (field programmable gate array, FPGA).

[0082] Optionally, the processor may execute various functions by running or executing a software program stored in the memory and calling data stored in the memory, such as executing the above-mentioned Figure 1 The method shown.

[0083] In a specific implementation, as an embodiment, the processor may include one or more microprocessors.

[0084] Among them, the memory is used to store the software program that executes the solution of the present application, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0085] Optionally, the memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be integrated with the processor, or may exist independently and be coupled to the processing unit through the interface circuit of the processor, which is not specifically limited in the embodiments of the present application.

[0086] It should be noted that the structure of the processor shown in this embodiment does not constitute a limitation on the device, and the actual device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0087] In addition, the technical effects of the processor can refer to the technical effects of the method described in the above method embodiment, which will not be repeated here.

[0088] It should be understood that the processor in the embodiments of the present application may be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0089] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0090] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0091] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0092] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0093] 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, or a combination of computer software and electronic hardware. 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.

[0094] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0095] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0096] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0097] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0098] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0099] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A cloud server service scheduling method, characterized in that: The method is applied to a cloud service system, the cloud service system communicates with a terminal, the cloud service system includes multiple edge clouds and a central cloud, the terminal communicates with the edge cloud, the multiple edge clouds communicate with the central cloud, the multiple edge clouds include multiple local edge clouds and multiple remote edge clouds, the local edge cloud is used to schedule an application initiated by the terminal to the remote edge cloud or the central cloud, and send the processing result of the remote edge cloud or the central cloud to the terminal; the method includes: Performing soot conversion on the application initiated by the terminal and constructing a weighted call graph of the application, wherein the weighted call graph is used to characterize the relationship between multiple microservices in the application; Obtaining the distance from the microservice to the multiple local edge clouds, the energy consumption data of the multiple local edge clouds, and the average time request arrival number of the multiple local edge clouds deploying microservices, and obtaining the processing cost of the multiple local edge clouds deploying microservices based on the distance and the arrival number, and confirming the local edge cloud with the smallest processing cost as the optimal edge cloud; The processing cost is expressed based on the following formula: ;in is the energy consumption data of the local edge cloud, The average time for the local edge cloud to deploy microservices is the number of requests arriving, is the distance from the microservice to the local edge cloud; wherein i represents the i-th microservice, and j represents the j-th local edge cloud; Obtaining the time cost of the optimal edge cloud and the multiple remote edge clouds and the central cloud, wherein the time cost includes a transmission time cost, a processing time cost, and a feedback time cost; The remote edge cloud or central cloud with the lowest time cost is confirmed as the target cloud node, and the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud are used as the data path of the microservice, and the data paths corresponding to multiple microservices are used as scheduling paths.

2. The cloud server service scheduling method according to claim 1, characterized in that: Performing soot conversion on the application initiated by the terminal and constructing a weighted call graph of the application includes: Converting the contents of multiple classes in the application to obtain multiple class representations, and obtaining a graph model based on the transmission relationship of the class representations; And assigning values ​​to the plurality of class representations to obtain weights for each class representation, and weighted processing of the graph model based on the weights to obtain the weighted call graph.

3. The cloud server service scheduling method according to claim 2, characterized in that: The energy consumption data of the local edge cloud is determined based on the following formula: , where v represents the exponential parameter and v ≥ 1, (1-α) represents the standard energy consumption of the idle local edge cloud and ,in Represents the scheduling function, which is expressed by the following formula: ; The optimization function is expressed based on the following formula: ,in The queue backlog at time t is expressed based on the following formula: , N represents the number of queue backlogs; Among them, i represents the i-th microservice, and j represents the j-th local edge cloud.

4. The cloud server service scheduling method according to claim 3, characterized in that: The average time request arrival number of the local edge cloud deployed microservice is determined based on the following formula: , where T is the unit time number of the time series, and j represents the jth local edge cloud; The relationship between microservice type i and the local edge cloud is expressed based on the following formula: 。 5. The cloud server service scheduling method according to claim 1, characterized in that: The transmission time cost of the optimal edge cloud and the multiple remote edge clouds is expressed based on the following formula: , where te represents the remote edge cloud, For input data, is the data transmission rate relative to the remote edge cloud; The transmission time cost between the optimal edge cloud and the central cloud is expressed based on the following formula: ,in represents the central cloud, For input data, is the data transmission rate relative to the central cloud.

6. The cloud server service scheduling method according to claim 1, characterized in that: The processing time cost of the multiple remote edge clouds is determined based on the following formula: ,in represents the computational workload, Indicates the computing power of the remote edge cloud; The processing time cost of the central cloud is determined based on the following formula: ,in Indicates the computing power of the central cloud.

7. The cloud server service scheduling method according to claim 1, characterized in that: The feedback time cost of the remote edge cloud is determined based on the following formula: ,in Indicates the amount of data transferred. is the data transmission rate relative to the remote edge cloud; The feedback time cost of the central cloud is determined based on the following formula: ,in is the data transmission rate relative to the central cloud.

8. A cloud server service scheduling device, characterized in that: The device is applied to a cloud service system, the cloud service system communicates with a terminal, the cloud service system includes multiple edge clouds and a central cloud, the terminal communicates with the edge cloud, the multiple edge clouds communicate with the central cloud, the multiple edge clouds include multiple local edge clouds and multiple remote edge clouds, the local edge cloud is used to schedule an application initiated by the terminal to the remote edge cloud or the central cloud, and send the processing result of the remote edge cloud or the central cloud to the terminal; the device includes: A splitting module, used to perform soot conversion on the application initiated by the terminal and construct a weighted call graph of the application, wherein the weighted call graph is used to characterize the relationship between multiple microservices in the application; The optimal edge cloud determination module is used to obtain the distance from the microservice to the multiple local edge clouds, the energy consumption data of the multiple local edge clouds, and the average time request arrival number of the multiple local edge clouds deploying microservices, and obtain the processing cost of the multiple local edge clouds deploying microservices based on the distance and the arrival number, and confirm the local edge cloud with the smallest processing cost as the optimal edge cloud; The processing cost is expressed based on the following formula: ;in is the energy consumption data of the local edge cloud, The average time for the local edge cloud to deploy microservices is the number of requests arriving, is the distance from the microservice to the local edge cloud; wherein i represents the i-th microservice, and j represents the j-th local edge cloud; The final scheduling module obtains the time cost of the optimal edge cloud and the multiple remote edge clouds and the central cloud, wherein the time cost includes transmission time cost, processing time cost and feedback time cost; The remote edge cloud or central cloud with the lowest time cost is confirmed as the target cloud node, and the cloud corresponding to the target cloud node and the communication path with the optimal edge cloud are used as the data path of the microservice, and the data paths corresponding to multiple microservices are used as scheduling paths.