Server scheduling method and system under side cloud collaborative architecture

Abstract

The invention provides a server scheduling method and system under an edge-cloud collaborative architecture, and the method comprises the steps: obtaining the scheduling request information of an edge server, the scheduling request information comprising edge address information; selecting a cloud working server set capable of providing service from all the server sets based on the edge address information and a small region priority principle, wherein the cloud working server set comprises a plurality of cloud working server groups; determining a cloud end distribution work server group from the plurality of cloud end work server groups based on a load rate minimum principle, wherein the cloud end distribution work server group comprises a plurality of cloud end distribution work servers; determining a cloud distribution main server and a cloud distribution standby server from the plurality of cloud distribution working servers based on a load rate minimum principle; and the cloud distribution main server and the cloud distribution standby server are distributed to the edge server, so that the problems of relatively low reliability, relatively low communication efficiency and relatively low utilization rate of the cloud server of the existing edge-cloud collaborative overall system can be solved.

Description

Technical Field

[0001] The present invention relates to the field of Internet technology, and in particular to a server scheduling method and system under an edge-cloud collaborative architecture. Background Art

[0002] Edge-cloud collaboration is to closely integrate edge computing with cloud computing, realize the sinking of cloud computing by reasonably allocating the tasks of cloud computing and edge computing, and extend cloud computing and cloud analysis to the edge. However, at present, there is a lack of a good collaboration mechanism between cloud computing and edge computing. There are problems such as low reliability of the overall edge-cloud collaboration system, low communication efficiency, and low utilization rate of cloud servers. Summary of the invention

[0003] In response to the above technical problems, the embodiments of the present application propose a server scheduling method and system under an edge-cloud collaborative architecture, which can solve the problems of low reliability, low communication efficiency, and low cloud server utilization rate of the current edge-cloud collaborative system.

[0004] In a first aspect, an embodiment of the present application provides a server scheduling method under an edge-cloud collaborative architecture, including:

[0005] Obtaining scheduling request information of an edge server, wherein the scheduling request information includes edge address information;

[0006] Based on the edge address information and the principle of smaller area priority, a cloud working server set that can provide services is selected from all server sets, wherein the cloud working server set includes multiple cloud working server groups;

[0007] Determining a cloud allocation work server group from the plurality of cloud work server groups based on a minimum load rate principle, wherein the cloud allocation work server group includes a plurality of cloud allocation work servers;

[0008] Determining a cloud allocation main server and a cloud allocation standby server from the plurality of cloud allocation working servers based on a minimum load rate principle;

[0009] The cloud allocation primary server and the cloud allocation backup server are allocated to the edge server.

[0010] In some embodiments, the cloud working server set that can provide services is selected from all server sets based on the edge address information and the principle of smaller area priority, specifically:

[0011] Determine a target city where the edge server is located based on the edge address information;

[0012] Filtering all the server sets based on the target city to obtain a cloud city server set;

[0013] If the cloud city server set is a non-empty set, the cloud city server set is screened based on the online status to obtain the cloud working server set.

[0014] In some embodiments, the filtering of the entire server set based on the target city to obtain the cloud city server set further includes:

[0015] If the cloud city server set is an empty set, determining the target province where the target city is located based on the target city;

[0016] Filtering all server sets based on the target province to obtain a cloud province server set;

[0017] If the cloud province server set is a non-empty set, the cloud province server set is filtered based on the online status to obtain the cloud working server set.

[0018] In some embodiments, the filtering of the entire server set based on the target province to obtain the cloud province server set further includes:

[0019] If the cloud province server set is an empty set, determining the target country where the target city is located based on the target city and / or the target province;

[0020] Filtering all server sets based on the target country to obtain a cloud country server set;

[0021] If the cloud national server set is a non-empty set, the cloud national server set is screened based on the online status to obtain the cloud working server set.

[0022] In some embodiments, the cloud allocation work server group is determined from a plurality of the cloud work server groups based on the minimum load rate principle, wherein the cloud allocation work server group includes a plurality of cloud allocation work servers, specifically:

[0023] Obtain the number of groups currently allocated to edge servers of the cloud working server group;

[0024] Obtaining the allocatable number of the cloud working server group that can be allocated to the edge server;

[0025] Obtaining current group load rates of the plurality of cloud server groups based on the group allocated quantity and the group allocatable quantity;

[0026] The cloud working server group with the smallest current group load rate is selected as the cloud allocated working server group.

[0027] In some embodiments, the cloud allocation main server and the cloud allocation standby server are determined from the plurality of cloud allocation working servers based on the principle of minimum current load rate, specifically:

[0028] Obtain the number of allocated servers currently allocated to edge servers by the cloud allocation work server;

[0029] Obtaining the number of allocable servers that the cloud allocation work server can allocate to edge servers;

[0030] Obtaining current load rates of the plurality of cloud servers based on the number of allocated server groups and the number of allocatable servers;

[0031] The cloud allocation working server with the smallest current load rate is selected as the cloud allocation main server, and the other cloud allocation working servers are selected as the cloud allocation backup servers.

[0032] In some embodiments, allocating the cloud-allocated primary server and the cloud-allocated backup server to the edge server further includes:

[0033] If the connection between the edge server and the cloud allocation master server is interrupted, reconnect to the cloud allocation master server after a preset dynamic interval, wherein the value of the preset dynamic interval increases with the increase in the number of connections;

[0034] If the preset dynamic interval is greater than the preset time threshold, the cloud allocation backup server is connected.

[0035] In some embodiments, if the preset dynamic interval is greater than a preset time threshold, connecting to the cloud allocation standby server further includes:

[0036] Sending a master server reconnection request to the cloud allocation master server based on a preset master reconnection interval;

[0037] If the reconnection request of the primary server is successful, the connection with the cloud-allocated backup server is disconnected.

[0038] In some embodiments, if the preset dynamic interval is greater than a preset time threshold, connecting to the cloud allocation standby server further includes:

[0039] If the connection to the cloud-allocated backup server fails, a rescheduling request message is sent to the cloud-allocated backup server.

[0040] In a second aspect, an embodiment of the present application provides a server scheduling system under an edge-cloud collaborative architecture, including:

[0041] A cloud scheduling server is used to obtain scheduling request information of an edge server, wherein the scheduling request information includes edge address information; based on the edge address information and the principle of smaller area priority, a cloud working server set that can provide services is selected from the entire server set, wherein the cloud working server set includes multiple cloud working server groups; based on the principle of minimum load rate, a cloud allocation working server group is determined from the multiple cloud working server groups, wherein the cloud allocation working server group includes multiple cloud allocation working servers; based on the principle of minimum load rate, a cloud allocation main server and a cloud allocation backup server are determined from the multiple cloud allocation working servers; and the cloud allocation main server and the cloud allocation backup server are allocated to the edge server.

[0042] A cloud working server, connected to the cloud scheduling server, is used to provide cloud connection, communication, computing and data storage functions;

[0043] An edge server connected to the cloud scheduling server and the cloud working server;

[0044] A smart device is connected to the edge server.

[0045] The present application provides a server scheduling method and system under an edge-cloud collaborative architecture, by obtaining scheduling request information of an edge server, wherein the scheduling request information includes edge address information; based on the edge address information and the principle of smaller area priority, a cloud working server set that can provide services is selected from the entire server set, wherein the cloud working server set includes multiple cloud working server groups; based on the principle of minimum load rate, a cloud allocation working server group is determined from multiple cloud working server groups, wherein the cloud allocation working server group includes multiple cloud allocation working servers; based on the principle of minimum load rate, a cloud allocation main server and a cloud allocation backup server are determined from multiple cloud allocation working servers; the cloud allocation main server and the cloud allocation backup server are allocated to the edge server, which can solve the problems of low reliability, low communication efficiency and low cloud server utilization rate of the current edge-cloud collaborative overall system, and improve the reliability, communication efficiency and cloud server utilization rate of the edge-cloud collaborative overall system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Hereinafter, the present invention will be described in more detail based on embodiments and with reference to the accompanying drawings.

[0047] Figure 1It is a flow chart of a server scheduling method under an edge-cloud collaborative architecture provided by an embodiment of the present invention;

[0048] Figure 2 is a flow chart of reallocating cloud working servers provided by an embodiment of the present invention;

[0049] Figure 3 is a flow chart of reallocating cloud working servers provided by another embodiment of the present invention;

[0050] Figure 4 This is a schematic diagram of a server scheduling system under an edge-cloud collaborative architecture provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The present invention will be further described below in conjunction with the accompanying drawings.

[0052] With the rapid development of Internet of Things (IoT), IoT technology has been applied in various fields of our lives, from smart homes to autonomous driving, from industrial IoT to smart cities. The demand for data generation and processing in IoT technology is growing at an unprecedented rate. Although the traditional cloud computing model is powerful, it often seems powerless when faced with the demand for real-time processing of massive data. Edge computing, as an emerging computing model, has emerged. Edge computing is a distributed computing architecture that moves data processing, storage, and service functions closer to the edge where data is generated, that is, close to the data source and user location, rather than relying on centralized data centers or cloud computing platforms. The core idea is to process data close to the terminal device to reduce latency, reduce bandwidth requirements, improve data privacy, and enhance real-time performance.

[0053] With the continuous maturity of technologies such as cloud computing and the advent of the industrial Internet era, the demand for "large connections, low latency, and high bandwidth" has become increasingly strong and urgent. Edge computing has become an important part of the development of this era. Together with cloud computing, it is required by more and more industrial applications. "Cloud-edge collaboration (or edge-cloud collaboration)" has become an important direction of technological evolution in the future. Today, "cloud-edge collaboration" has been widely used in scenarios such as smart transportation, industrial Internet, energy Internet, smart medical care, smart home, security monitoring, agricultural production, and cloud games. At present, in the field of Internet of Things, "cloud-edge collaboration" is the key development direction and the optimal architecture of the current Internet of Things platform. "Cloud" and "edge" are both part of the Internet of Things platform and are distributed computing structures of the overall system. The two should be deeply integrated, with flexible division of labor and coordinated scheduling to form an integrated system that is constantly iterating and evolving together.

[0054] The current status of the Internet of Things industry is that the Internet of Things system has "cloud computing" of cloud platforms and "edge computing" of smart gateways. Cloud computing and edge computing are gradually moving towards coordinated development during the running-in process. However, at the current stage, the coordination between cloud computing and edge computing still faces many challenges. In many cases, there is a lack of a good coordination mechanism between the cloud and the edge, and each "goes its own way", failing to fully exert the effectiveness of the overall system. In particular, the reliability of the cloud computing platform is still not good enough. Once a problem occurs in the cloud, it is easy for all smart devices to become uncontrollable. There is still room for improvement in the system architecture design and server scheduling solutions, and the overall reliability and resource utilization rate need to be improved.

[0055] First, as Figure 1 As shown, in response to the above technical problems, an embodiment of the present application provides a server scheduling method under an edge-cloud collaborative architecture, including:

[0056] S101: Obtain scheduling request information of an edge server, wherein the scheduling request information includes edge address information;

[0057] It should be noted that the edge server can be an intelligent gateway. The edge server is deployed on the edge side, that is, close to the smart device. For example, if the smart device is installed in the user's home, then the edge server is installed at home; if the smart device is installed in the factory, then the edge server is installed in the factory. The edge server provides a series of functions such as device connection, data communication, device control, data storage, and status monitoring for the smart device. When the user can establish communication directly with the edge server, it can work independently and provide services without the participation of the cloud work server. The edge address information is the IP address (Internet Protocol Address) of the edge server.

[0058] S102: selecting a cloud working server set that can provide services from all server sets based on the edge address information and a smaller area priority principle, wherein the cloud working server set includes a plurality of cloud working server groups;

[0059] In some embodiments, the cloud working server set that can provide services is selected from all server sets based on the edge address information and the principle of smaller area priority, specifically:

[0060] Determine a target city where the edge server is located based on the edge address information;

[0061] Filtering all the server sets based on the target city to obtain a cloud city server set;

[0062] If the cloud city server set is a non-empty set, the cloud city server set is screened based on the online status to obtain the cloud working server set.

[0063] In some embodiments, the filtering of the entire server set based on the target city to obtain the cloud city server set further includes:

[0064] If the cloud city server set is an empty set, determining the target province where the target city is located based on the target city;

[0065] Filtering all server sets based on the target province to obtain a cloud province server set;

[0066] If the cloud province server set is a non-empty set, the cloud province server set is filtered based on the online status to obtain the cloud working server set.

[0067] In some embodiments, the filtering of the entire server set based on the target province to obtain the cloud province server set further includes:

[0068] If the cloud province server set is an empty set, determining the target country where the target city is located based on the target city and / or the target province;

[0069] Filtering all server sets based on the target country to obtain a cloud country server set;

[0070] If the cloud national server set is a non-empty set, the cloud national server set is screened based on the online status to obtain the cloud working server set.

[0071] It should be noted that if the cloud country server set is an empty set, it means that there is no cloud working server that can provide services in the target country. At this time, the entire server set (or cloud global server) can be screened based on the online status to obtain the cloud working server set. In this step, a cloud working server set that can provide services can be screened out. Among them, the smaller area priority principle means that when screening the entire server set, first screen according to smaller administrative divisions, and then screen according to larger administrative divisions (that is, cities first, provinces second, and countries third).

[0072] It should be noted that, through the IP address of the edge server, the target city, target province and target country where the edge server is located can be gradually determined from small to large, and then the cloud working server set that can provide services can be found in the target city, the target province and the target country. The cloud working server set that is closer to the edge server can be selected to shorten the distance between the cloud working server set and the edge server, thereby reducing communication delay and improving communication efficiency; among them, if the set is an empty set (that is, the cloud city server set is an empty set, the cloud province server set is an empty set, and the cloud country server set is an empty set), it means that there is no cloud working server (or cloud working server set) that can provide services in the area, and at this time it is necessary to gradually expand the scope for searching.

[0073] S103: determining a cloud allocation working server group from the plurality of cloud working server groups based on a minimum load rate principle, wherein the cloud allocation working server group includes a plurality of cloud allocation working servers;

[0074] In some embodiments, the cloud allocation work server group is determined from a plurality of the cloud work server groups based on the minimum load rate principle, wherein the cloud allocation work server group includes a plurality of cloud allocation work servers, specifically:

[0075] Obtain the number of groups currently allocated to edge servers of the cloud working server group;

[0076] Obtaining the allocatable number of the cloud working server group that can be allocated to the edge server;

[0077] Obtaining current group load rates of the plurality of cloud server groups based on the group allocated quantity and the group allocatable quantity;

[0078] The cloud working server group with the smallest current group load rate is selected as the cloud allocated working server group.

[0079] It should be noted that the cloud work server group is a logical concept, and several cloud work servers deployed in the same region (or work area) can be grouped into a group (i.e., the cloud work server group). The current group load rate of multiple cloud server groups based on the allocated number of the group and the allocatable number of the group is: "the allocated number of the group / the allocatable number of the group". The allocatable number of the group that can be allocated to the edge server by the cloud work server group can be manually set by the operation and maintenance personnel according to the hardware resources of the server and the nature and activity of the edge server, and can also be dynamically changed according to the actual operation conditions to adapt to the needs of the edge server and improve the utilization rate of server resources.

[0080] S104: determining a cloud allocation main server and a cloud allocation standby server from the plurality of cloud allocation working servers based on a minimum load rate principle;

[0081] S105: Allocate the cloud-allocated primary server and the cloud-allocated backup server to the edge server.

[0082] In some embodiments, the cloud allocation main server and the cloud allocation standby server are determined from the plurality of cloud allocation working servers based on the principle of minimum current load rate, specifically:

[0083] Obtain the number of allocated servers currently allocated to edge servers by the cloud allocation work server;

[0084] Obtaining the number of allocable servers that the cloud allocation work server can allocate to edge servers;

[0085] Obtaining current load rates of the plurality of cloud servers based on the number of allocated server groups and the number of allocatable servers;

[0086] The cloud allocation working server with the smallest current load rate is selected as the cloud allocation main server, and the other cloud allocation working servers are selected as the cloud allocation backup servers.

[0087] It should be noted that the current load rate of the plurality of cloud servers obtained based on the number of allocated server groups and the number of allocatable servers is: "number of allocated server groups / number of allocatable servers". After obtaining the current load rate, the server load rates of the plurality of cloud allocation working servers can be sorted according to the current load rate, and the cloud allocation working server with the lowest server load rate is selected as the cloud allocation main server, and the cloud allocation working server with the second lowest server load rate is selected as the cloud allocation backup server. By allocating the cloud allocation main server and the cloud allocation backup server to the edge server, the edge server can be scheduled by itself without relying on the cloud scheduling server, which can solve the problem of complete equipment loss of control caused by the failure of the cloud scheduling server, ensure the continuity of the service, and prevent service interruption from affecting the user experience. Among them, the number of allocatable servers that the cloud working server (or cloud allocation working server) can allocate to the edge server is manually set by the operation and maintenance personnel according to the hardware resource situation of the server and the nature and activity of the edge server, and can also be dynamically changed according to the actual operation situation to adapt to the needs of the edge server and improve the utilization rate of server resources.

[0088] In some embodiments, allocating the cloud-allocated primary server and the cloud-allocated backup server to the edge server further includes:

[0089] If the connection between the edge server and the cloud allocation master server is interrupted, reconnect to the cloud allocation master server after a preset dynamic interval, wherein the value of the preset dynamic interval increases with the increase in the number of connections;

[0090] If the preset dynamic interval is greater than the preset time threshold, the cloud allocation backup server is connected.

[0091] In some embodiments, if the preset dynamic interval is greater than a preset time threshold, connecting to the cloud allocation standby server further includes:

[0092] Sending a master server reconnection request to the cloud allocation master server based on a preset master reconnection interval;

[0093] If the reconnection request of the primary server is successful, the connection with the cloud-allocated backup server is disconnected.

[0094] In some embodiments, if the preset dynamic interval is greater than a preset time threshold, connecting to the cloud allocation standby server further includes:

[0095] If the connection to the cloud-allocated backup server fails, a rescheduling request message is sent to the cloud-allocated backup server.

[0096] It should be noted that if Figure 2 As shown, the edge server determines whether the connection with the cloud allocation main server is interrupted by monitoring the heartbeat between the edge server and the cloud allocation main server. If the edge server reconnects with the cloud allocation main server successfully, the communication with the cloud allocation main server is restored. The initial value of the preset dynamic interval can be 2 seconds, which gradually increases until it stops increasing after 2 minutes. The preset time threshold can be 2 minutes. When the preset dynamic interval is greater than the preset time threshold, it is usually the case that the cloud allocation main server fails. At this time, the cloud allocation main server is no longer reconnected, but the cloud allocation backup server is connected to ensure normal operation.

[0097] It should be noted that when connecting to the cloud-allocated backup server, the judgment is also made based on the connection time interval. That is, if the preset dynamic interval when connecting to the cloud-allocated backup server is greater than the preset time threshold, the connection to the cloud-allocated backup server fails. At this time, a rescheduling request message can be sent to the cloud scheduling server to reallocate the cloud working server to ensure the normal progress of the work.

[0098] It should be noted that the preset master reconnection interval may be 15 minutes. If the master server reconnection request fails, it is determined that the cloud allocation master server has not yet recovered, and the cloud allocation master server continues to be reconnected periodically.

[0099] It should be noted that, if the cloud-allocated main server and the cloud-allocated backup server are unable to communicate at the same time, the edge server will start the process of reallocating the cloud-allocated working server, such as Figure 3 As shown, the specific process of the edge server reacquiring the cloud-allocated primary server and the cloud-allocated backup server is described in detail as follows:

[0100] 1) The edge server detects whether it is currently connected to the Internet;

[0101] 2) If there is no Internet, end this process; if the Internet connection is normal, proceed to the next step;

[0102] 3) The edge server sends a message to the cloud scheduling server to reallocate the cloud work server;

[0103] 4) The cloud scheduling server allocates the cloud allocation primary server and the cloud allocation backup server to the edge server according to the rules;

[0104] 5) The cloud scheduling server returns the information of the cloud-allocated primary server and the cloud-allocated backup server to the edge server;

[0105] 6) The edge server connects to the cloud distribution master server (cloud working server).

[0106] It should be noted that when a cloud-allocated primary server fails, the originally connected edge server will automatically switch to the cloud-allocated backup server. When the cloud-allocated primary server recovers from the failure, the edge server needs to switch back to the original server to avoid long-term connection to the cloud-allocated backup server, which causes uneven server load, affects reliability, and wastes server resources. The edge server can automatically switch from the cloud-allocated backup server back to the cloud-allocated primary server without manual operation, but it takes a long time and the timing interval is large. When it is necessary to switch back to the cloud-allocated primary server as soon as possible, the operation and maintenance personnel can operate the switch. When operated by someone, the specific process of the edge server switching from the cloud-allocated backup server back to the cloud-allocated primary server is described in detail as follows:

[0107] 1) The failed cloud allocation master server returns to normal;

[0108] 2) The cloud scheduling server detects that the heartbeat of the cloud allocation master server has recovered;

[0109] 3) Operation and maintenance personnel operate to start the subsequent switching process;

[0110] 4) The cloud dispatch server sends instructions to the remaining cloud working servers in the cloud allocation server group where the original failed server is located;

[0111] 5) The cloud work server that receives the command sends the command to all currently connected edge servers;

[0112] 6) The edge server receives a failure recovery notification instruction;

[0113] 7) The edge server determines the current connection status. If the current connection is to the primary server, the process ends;

[0114] 8) If the current connection is to the backup server, proceed to the next step;

[0115] 9) The edge server randomly waits for a period of time, for example, randomly waits for 1-60 seconds;

[0116] 10) The edge server attempts to connect to the cloud allocation master server;

[0117] 11) The edge server restores the connection with the cloud-allocated master server.

[0118] In summary, the present application provides a server scheduling method under an edge-cloud collaborative architecture, by obtaining scheduling request information of an edge server, wherein the scheduling request information includes edge address information; based on the edge address information and the principle of smaller area priority, a cloud working server set that can provide services is selected from the entire server set, wherein the cloud working server set includes multiple cloud working server groups; based on the principle of minimum load rate, a cloud allocation working server group is determined from multiple cloud working server groups, wherein the cloud allocation working server group includes multiple cloud allocation working servers; based on the principle of minimum load rate, a cloud allocation main server and a cloud allocation backup server are determined from multiple cloud allocation working servers; the cloud allocation main server and the cloud allocation backup server are allocated to the edge server, which can solve the problems of low reliability, low communication efficiency and low cloud server utilization rate of the current edge-cloud collaborative overall system, and improve the reliability, communication efficiency and cloud server utilization rate of the edge-cloud collaborative overall system.

[0119] Second, as Figure 4 As shown, the embodiment of the present application provides a server scheduling system under an edge-cloud collaborative architecture, including:

[0120] A cloud scheduling server is used to obtain scheduling request information of an edge server, wherein the scheduling request information includes edge address information; based on the edge address information and the principle of smaller area priority, a cloud working server set that can provide services is selected from the entire server set, wherein the cloud working server set includes multiple cloud working server groups; based on the principle of minimum load rate, a cloud allocation working server group is determined from the multiple cloud working server groups, wherein the cloud allocation working server group includes multiple cloud allocation working servers; based on the principle of minimum load rate, a cloud allocation main server and a cloud allocation backup server are determined from the multiple cloud allocation working servers; and the cloud allocation main server and the cloud allocation backup server are allocated to the edge server.

[0121] A cloud working server, connected to the cloud scheduling server, is used to provide cloud connection, communication, computing and data storage functions;

[0122] An edge server connected to the cloud scheduling server and the cloud working server;

[0123] A smart device is connected to the edge server.

[0124] It should be noted that the cloud scheduling server is one type of cloud server, which is mainly responsible for scheduling cloud working servers, can establish connections with cloud working servers, and monitor the status of each server through a heartbeat mechanism, can provide cloud working server allocation functions for edge servers, and when a certain edge server initiates a request, can allocate cloud allocation primary servers and cloud allocation backup servers to edge servers according to rules, and can record server allocation information for calculation of allocation rules. When a cloud working server fails or recovers from a failure, it can schedule the edge server to restore the connection.

[0125] It should be noted that the cloud working server is one of the cloud servers, which is used to provide cloud connection, communication, computing, data storage and other functions, provide cloud support for edge servers, and work with edge servers. When the user is in a mobile state and cannot communicate with the edge server, the cloud working server provides services. For example, when the user is not at home, he can still control the lights at home with a mobile phone app. The actual communication link is from the mobile phone app to the cloud working server to the edge server to the smart lamp.

[0126] It should be noted that the cloud work server group is a logical concept. Several cloud work servers deployed in the same region (work area) can be organized into a group. The servers in the same group can serve as primary and backup for each other, so that when a server fails, the remaining servers can provide hot backup. Generally, two servers can be set as a group, and the servers in the group are numbered as A and B. When the cloud scheduling server assigns work servers to edge servers, a certain group will be selected for assignment. For example, for edge server 1, server A of cloud work server group 1 is assigned as the cloud-allocated primary server, and server B of cloud work server group 1 is assigned as the cloud-allocated backup server; for edge server 2, server B of cloud work server group 1 is assigned as the cloud-allocated primary server, and server A of cloud work server group 1 is assigned as the cloud-allocated backup server; for edge server X, server A of cloud work server group n is assigned as the cloud-allocated primary server, and server B of cloud work server group n is assigned as the cloud-allocated backup server.

[0127] It should be noted that the edge server (intelligent gateway) is deployed on the edge side, that is, close to the smart device. For example, if the smart device is installed in the user's home, the edge server is installed in the home; if the smart device is installed in the factory, the edge server is installed in the factory. The edge server provides a series of functions for the smart device, such as device connection, data communication, device control, data storage, and status monitoring. When the user can directly establish communication with the edge server, it can work independently and provide services without the participation of the cloud work server. For example, when the user is at home, even if there is only a local area network at home and no Internet connection, you can still use the mobile phone App to control the lights at home. The actual communication link is from the mobile phone App to the edge server to the smart lamp. In an environment with the Internet, the edge server will maintain a connection with the cloud work server and work together. When the cloud server fails or recovers from a failure, the edge server can automatically schedule and switch the connection by itself to maintain the availability of the cloud service; the smart device can connect and communicate with the edge server through the Internet of Things to achieve intelligent and automated control and management. The application range of smart devices is very wide, covering various fields such as home, city, industry, agriculture, and medical care. For example, the smart home field can include smart lamps, smart temperature sensors, security equipment, etc.; the smart city field can include traffic management equipment, environmental monitoring equipment, energy monitoring equipment, etc. An edge server can connect to several smart devices. Depending on the IoT protocol and device performance, the number of connections can range from dozens to hundreds to tens of thousands.

[0128] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and / or computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0129] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0131] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A server scheduling method under edge-cloud collaborative architecture, characterized in that: include: Obtaining scheduling request information of an edge server, wherein the scheduling request information includes edge address information; Based on the edge address information and the principle of smaller area priority, a cloud working server set that can provide services is selected from all server sets, wherein the cloud working server set includes multiple cloud working server groups; Determining a cloud allocation work server group from the plurality of cloud work server groups based on a minimum load rate principle, wherein the cloud allocation work server group includes a plurality of cloud allocation work servers; Determining a cloud allocation main server and a cloud allocation standby server from the plurality of cloud allocation working servers based on a minimum load rate principle; The cloud allocation primary server and the cloud allocation backup server are allocated to the edge server.

2. The server scheduling method under the edge-cloud collaborative architecture according to claim 1 is characterized in that: The cloud working server set that can provide services is selected from all server sets based on the edge address information and the principle of smaller area priority, specifically: Determine a target city where the edge server is located based on the edge address information; Filtering all the server sets based on the target city to obtain a cloud city server set; If the cloud city server set is a non-empty set, the cloud city server set is screened based on the online status to obtain the cloud working server set.

3. The server scheduling method under the edge-cloud collaborative architecture according to claim 2 is characterized in that: The screening of all server sets based on the target city to obtain a cloud city server set further includes: If the cloud city server set is an empty set, determining the target province where the target city is located based on the target city; Filtering all server sets based on the target province to obtain a cloud province server set; If the cloud province server set is a non-empty set, the cloud province server set is filtered based on the online status to obtain the cloud working server set.

4. The server scheduling method under the edge-cloud collaborative architecture according to claim 3 is characterized in that: The filtering of all server sets based on the target province to obtain a cloud province server set also includes: If the cloud province server set is an empty set, determining the target country where the target city is located based on the target city and / or the target province; Filtering all server sets based on the target country to obtain a cloud country server set; If the cloud national server set is a non-empty set, the cloud national server set is screened based on the online status to obtain the cloud working server set.

5. The server scheduling method under the edge-cloud collaborative architecture according to claim 1 is characterized in that: The cloud allocation working server group is determined from the plurality of cloud working server groups based on the minimum load rate principle, wherein the cloud allocation working server group includes a plurality of cloud allocation working servers, specifically: Obtain the number of groups currently allocated to edge servers of the cloud working server group; Obtaining the allocatable number of the cloud working server group that can be allocated to the edge server; Obtaining current group load rates of the plurality of cloud server groups based on the group allocated quantity and the group allocatable quantity; The cloud working server group with the smallest current group load rate is selected as the cloud allocated working server group.

6. The server scheduling method under the edge-cloud collaborative architecture according to claim 1 is characterized in that: The determining of the cloud allocation main server and the cloud allocation standby server from the plurality of cloud allocation working servers based on the principle of minimum current load rate is specifically: Obtain the number of allocated servers currently allocated to edge servers by the cloud allocation work server; Obtaining the number of allocable servers that the cloud allocation work server can allocate to edge servers; Obtaining current load rates of the plurality of cloud servers based on the number of allocated server groups and the number of allocatable servers; The cloud allocation working server with the smallest current load rate is selected as the cloud allocation main server, and the other cloud allocation working servers are selected as the cloud allocation backup servers.

7. The server scheduling method under the edge-cloud collaborative architecture according to claim 1 is characterized in that: The allocating the cloud allocation main server and the cloud allocation backup server to the edge server also includes: If the connection between the edge server and the cloud allocation master server is interrupted, reconnect to the cloud allocation master server after a preset dynamic interval, wherein the value of the preset dynamic interval increases with the increase in the number of connections; If the preset dynamic interval is greater than the preset time threshold, the cloud allocation backup server is connected.

8. The server scheduling method under the edge-cloud collaborative architecture according to claim 7 is characterized in that: If the preset dynamic interval is greater than a preset time threshold, connecting to the cloud allocation standby server also includes: Sending a master server reconnection request to the cloud allocation master server based on a preset master reconnection interval; If the reconnection request of the primary server is successful, the connection with the cloud-allocated backup server is disconnected.

9. The server scheduling method under the edge-cloud collaborative architecture according to claim 7, characterized in that: If the preset dynamic interval is greater than a preset time threshold, connecting to the cloud allocation standby server also includes: If the connection to the cloud-allocated backup server fails, a rescheduling request message is sent to the cloud-allocated backup server.

10. A server scheduling system under edge-cloud collaborative architecture, characterized in that: include: A cloud scheduling server is used to obtain scheduling request information of an edge server, wherein the scheduling request information includes edge address information; based on the edge address information and the principle of smaller area priority, a cloud working server set that can provide services is selected from the entire server set, wherein the cloud working server set includes multiple cloud working server groups; based on the principle of minimum load rate, a cloud allocation working server group is determined from the multiple cloud working server groups, wherein the cloud allocation working server group includes multiple cloud allocation working servers; based on the principle of minimum load rate, a cloud allocation main server and a cloud allocation backup server are determined from the multiple cloud allocation working servers; and the cloud allocation main server and the cloud allocation backup server are allocated to the edge server. A cloud working server, connected to the cloud scheduling server, is used to provide cloud connection, communication, computing and data storage functions; An edge server connected to the cloud scheduling server and the cloud working server; A smart device is connected to the edge server.

Images