Automatic horizontal extension system and method based on access delay

Abstract

The invention provides an automatic horizontal extension system and method based on access delay, and relates to the technical field of cloud computing. The system comprises a resource scheduling server which screens out a plurality of resident areas and deploys a service instance cluster in a resource pool of each resident area; the domain name server is used for analyzing the access request of each client to a service instance cluster of a resident area with the nearest spatial distance; the log processing server obtains the access delay of each client according to the log data of each service instance cluster; the monitoring alarm server obtains alarm information when monitoring that the access time delay of at least one client is greater than a time delay threshold value; and the automatic horizontal expansion server obtains an automatic horizontal expansion strategy according to the alarm information and triggers the resource scheduling server to execute the automatic horizontal expansion strategy. According to the system, when it is monitored that the access time delay of the client is larger than the time delay threshold value, the automatic horizontal extension strategy is generated and executed, and the user experience and the service quality are improved.

Description

Technical Field

[0001] The present application relates to the field of cloud computing technology, and in particular to an automatic horizontal expansion system and method based on access latency. Background Art

[0002] To ensure the stable operation of cloud computing services, the current mainstream practice is to monitor service instances running on the server side in real time, focusing on key performance indicators such as CPU usage, memory utilization, disk I / O, and network bandwidth. When the system detects an anomaly in a service instance, it automatically triggers the auto-scaling mechanism, dynamically adjusting the number of service instances to share the load by expanding or replacing the abnormal instance, thereby ensuring the overall availability and stability of the cloud service.

[0003] However, this monitoring and optimization strategy primarily focuses on the server's own operational status and resource utilization, lacking effective awareness of the actual interactions between the client and the server. While cloud computing services appear accessible from the server's perspective, degradation in client access quality due to network latency, geographical distance, or other network link issues often goes undetected. Such issues can typically only be identified retrospectively through user feedback or log analysis, resulting in prolonged client access instability before the issue is located and fixed, impacting user experience and service quality. Summary of the Invention

[0004] The present application provides an automatic horizontal expansion system and method based on access delay. When the access delay of the client is monitored to be greater than the delay threshold, an automatic horizontal expansion strategy is generated and executed, thereby improving user experience and service quality.

[0005] A first aspect of the present application provides an automatic horizontal expansion system based on access latency, comprising: a resource scheduling server, a domain name server, a log processing server, a monitoring and alarm server, and an automatic horizontal expansion server;

[0006] The resource scheduling server selects multiple resident regions from multiple preset geographic regions and deploys a service instance cluster in the resource pool of each resident region;

[0007] The domain name server determines the resident region closest to each client based on its network address and resolves each client's access request to the service instance cluster in the corresponding resident region; each network address matches a geographic region.

[0008] The log processing server obtains the log data of each service instance cluster in real time and obtains the client access information of each service instance cluster based on the log data of each service instance cluster. The client access information of the service instance cluster includes: the access latency of each client accessing the corresponding service instance cluster;

[0009] The monitoring and alarm server monitors the access latency of each client. When the access latency of at least one client is greater than the preset latency threshold, an alarm message is generated based on the client access information of each service instance cluster and sent to the automatic horizontal scaling server.

[0010] The automatic horizontal expansion server obtains the automatic horizontal expansion strategy based on the alarm information and triggers the resource scheduling server to execute the automatic horizontal expansion strategy.

[0011] In one possible design, a monitoring alarm server includes: a candidate cluster determining device and an alarm generating device;

[0012] The candidate cluster determining device selects at least one candidate service instance cluster from a plurality of service instance clusters; wherein the candidate service instance cluster refers to a service instance cluster accessed by a client whose access delay is greater than a delay threshold;

[0013] The alarm generating device obtains alarm information according to the client access information of each candidate service instance cluster.

[0014] In one possible design, the client access information of the service instance cluster further includes: the number of client accesses and the network address of each client accessing the corresponding service instance cluster;

[0015] An automatic horizontal expansion server comprising: an abnormal cluster determining device, an abnormal client determining device and a policy generating device;

[0016] The abnormal cluster determining device selects at least one abnormal service instance cluster from at least one candidate service instance cluster; the number of client accesses to the abnormal service instance cluster is greater than a preset first threshold;

[0017] An abnormal client determining device, screening out at least one abnormal client from a plurality of clients accessing any abnormal service instance cluster; wherein the network address of the abnormal client does not match the resident area corresponding to the accessed abnormal service instance cluster;

[0018] The policy generating device obtains an automatic horizontal expansion policy according to the network address of at least one abnormal client.

[0019] In one possible design, the resource scheduling server includes: a region determination device and a policy execution device;

[0020] A region determination device, which selects at least one elastic region from a plurality of geographical regions according to an automatic horizontal expansion strategy; wherein the network address of each abnormal client is matched with a elastic region;

[0021] A policy execution device deploys a cluster of elastic service instances in the resource pool of each elastic zone;

[0022] The domain name server resolves the access request of each abnormal client to the elastic service instance cluster of the corresponding elastic zone.

[0023] In a possible design, the policy execution device stops running and destroys an elastic service instance cluster whose number of client accesses is less than or equal to a preset second threshold.

[0024] In one possible design, the resource scheduling server further includes: a region determining device;

[0025] The region determining device obtains the client distribution information of each geographical region and selects a plurality of resident regions from the plurality of geographical regions according to the client distribution information of each geographical region.

[0026] In one possible design, a service instance cluster includes: a load balancer, multiple service instances, a traffic mirror, and a log collector;

[0027] Domain name server, which resolves each client's access request to the load balancer of the corresponding resident area;

[0028] For any service instance cluster, the load balancer distributes the client access requests resolved by the domain name server to multiple service instances;

[0029] Traffic mirror, which mirrors the traffic between each service instance and the corresponding client to the log collector;

[0030] The log collector obtains the log data of the corresponding service instance cluster based on the traffic mirrored by the log collector.

[0031] A second aspect of the present application provides an automatic horizontal expansion method based on access latency, which is applied to a resource scheduling server of an automatic horizontal expansion system based on access latency, wherein the system further includes an automatic horizontal expansion server. The method includes:

[0032] Filter multiple resident areas from multiple preset geographical areas;

[0033] Deploy a service instance cluster in the resource pool of each resident region;

[0034] When triggered by the automatic horizontal scaling server, obtain the automatic horizontal scaling policy sent by the automatic horizontal scaling server;

[0035] Execute automatic horizontal scaling policy.

[0036] A third aspect of the present application provides an automatic horizontal expansion method based on access delay, which is applied to a monitoring and alarm server of an automatic horizontal expansion system based on access delay, wherein the system further includes a log processing server and an automatic horizontal expansion server. The method includes:

[0037] Obtain client access information for each service instance cluster generated by the log processing server in real time; client access information for a service instance cluster includes: access latency for each client accessing the corresponding service instance cluster;

[0038] Monitor the access latency of each client;

[0039] When the access delay of at least one client is monitored to be greater than the preset delay threshold, an alarm message is obtained based on the client access information of each service instance cluster;

[0040] Sends warning information to the automatic horizontal scaling server.

[0041] A fourth aspect of the present application provides an automatic horizontal expansion method based on access latency, which is applied to an automatic horizontal expansion server of an automatic horizontal expansion system based on access latency, wherein the system further includes a log processing server and a resource scheduling server. The method includes:

[0042] When the log processing server monitors that the access delay of at least one client is greater than a preset delay threshold, obtaining alarm information sent by the log processing server;

[0043] Based on the alarm information, an automatic horizontal expansion strategy is obtained;

[0044] Trigger the resource scheduling server to execute the automatic horizontal scaling policy.

[0045] A fifth aspect of the present application provides an automatic horizontal expansion method based on access delay, which is applied to a domain name server of an automatic horizontal expansion system based on access delay, and the method includes:

[0046] Upon receiving an access request from a client, determining the resident region closest to the corresponding client space based on the network address of each client; wherein each network address is matched to a geographical region;

[0047] Parse each client's access request to the service instance cluster in the corresponding resident area.

[0048] A sixth aspect of the present application provides an automatic horizontal expansion method based on access delay, which is applied to a log processing server of an automatic horizontal expansion system based on access delay. The method includes:

[0049] Get log data of each service instance cluster in real time;

[0050] Based on the log data of each service instance cluster, obtain the client access information of each service instance cluster;

[0051] Send client access information of each service instance cluster to the monitoring alarm server.

[0052] A seventh aspect of the present application provides a resource scheduling server, the resource scheduling server comprising:

[0053] an area determination device for screening out a plurality of permanent areas from a plurality of preset geographical areas;

[0054] A cluster deployment device, used to deploy a service instance cluster in the resource pool of each resident area;

[0055] A policy acquisition device, configured to acquire an automatic horizontal expansion policy sent by the automatic horizontal expansion server when triggered by the automatic horizontal expansion server;

[0056] A policy execution device is used to execute the automatic horizontal expansion policy.

[0057] An eighth aspect of the present application provides a monitoring alarm server, the monitoring alarm server comprising:

[0058] An access information acquisition device is used to acquire, in real time, client access information of each service instance cluster generated by the log processing server; the client access information of the service instance cluster includes: the access delay of each client accessing the corresponding service instance cluster;

[0059] A delay monitoring device is used to monitor the access delay of each client;

[0060] An alarm generating device is used to obtain alarm information based on the client access information of each service instance cluster when the access delay of at least one client is monitored to be greater than a preset delay threshold;

[0061] The alarm sending device is used to send the alarm information to the automatic horizontal expansion server.

[0062] A ninth aspect of the present application provides an automatic horizontal expansion server, the automatic horizontal expansion server comprising:

[0063] An alarm obtaining device is used to obtain the alarm information sent by the log processing server when the log processing server monitors that the access delay of at least one client is greater than a preset delay threshold;

[0064] A strategy generating device for obtaining an automatic horizontal expansion strategy based on the alarm information;

[0065] The policy triggering device is used to trigger the resource scheduling server to execute the automatic horizontal expansion policy.

[0066] A tenth aspect of the present application provides a domain name server, comprising:

[0067] A region matching device is used to determine, upon receiving an access request from a client, the resident region closest to the corresponding client based on the network address of each client; wherein each network address is matched with a geographical region;

[0068] Domain name resolution device, used to resolve each client's access request to the service instance cluster in the corresponding resident area.

[0069] An eleventh aspect of the present application provides a log processing server, comprising:

[0070] Log acquisition device, used to obtain log data of each service instance cluster in real time;

[0071] An access information parsing device, configured to obtain client access information of each service instance cluster based on the log data of each service instance cluster;

[0072] The access information sending device is used to send the client access information of each service instance cluster to the monitoring alarm server.

[0073] A twelfth aspect of the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0074] Memory stores computer-executable instructions;

[0075] When the processor executes the computer-executable instructions stored in the memory, it is used to implement the automatic horizontal expansion method based on access latency as described in any one of the second to sixth aspects.

[0076] The thirteenth aspect of the present application provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement an automatic horizontal expansion method based on access latency as described in any one of the second to sixth aspects.

[0077] The fourteenth aspect of the present application provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the automatic horizontal expansion method based on access latency as described in any one of the second to sixth aspects.

[0078] The present application provides an automatic horizontal expansion system and method based on access delay, which includes: a resource scheduling server, which screens out multiple resident areas and deploys a service instance cluster in the resource pool of each resident area; a domain name server, which resolves the access request of each client to the service instance cluster of the resident area with the closest spatial distance; a log processing server, which obtains the access delay of each client based on the log data of each service instance cluster; a monitoring and alarm server, which obtains an alarm message when monitoring that the access delay of at least one client is greater than the delay threshold; an automatic horizontal expansion server, which obtains an automatic horizontal expansion policy based on the alarm message and triggers the resource scheduling server to execute the automatic horizontal expansion policy. The following technical effects are achieved: by deploying resource pools in multiple geographical areas, the service instances are deployed nearby, reducing the access delay caused by spatial distance; the log data of each service instance cluster is obtained by the log processing server, the access information of different clients is obtained, and the acquisition time of the client's access delay is reduced; the automatic horizontal expansion policy is generated by the automatic horizontal expansion server, and the resource scheduling server is triggered to execute the automatic horizontal expansion policy, thereby achieving on-demand elastic scaling of the service instances. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the technical solutions in the present application or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0080] Figure 1 A schematic diagram of a scenario of automatic horizontal expansion based on access latency provided in an embodiment of the present application;

[0081] Figure 2 A schematic diagram of the structure of an automatic horizontal expansion system based on access latency provided in an embodiment of the present application;

[0082] Figure 3 A schematic diagram of the structure of a service instance cluster provided in an embodiment of the present application;

[0083] Figure 4 Schematic diagram of the process of the automatic horizontal expansion method based on access delay provided in the embodiment of the present application Figure 1 ;

[0084] Figure 5Schematic diagram of the process of the automatic horizontal expansion method based on access delay provided in the embodiment of the present application Figure 2 ;

[0085] Figure 6 Schematic diagram of the process of the automatic horizontal expansion method based on access delay provided in the embodiment of the present application Figure 3 ;

[0086] Figure 7 A schematic diagram of the structure of the resource scheduling server provided in an embodiment of the present application;

[0087] Figure 8 A schematic diagram of the structure of the monitoring alarm server provided in an embodiment of the present application;

[0088] Figure 9 A schematic diagram of the structure of an automatic horizontal expansion server provided in an embodiment of the present application;

[0089] Figure 10 A schematic diagram of the structure of a domain name server provided in an embodiment of the present application;

[0090] Figure 11 A schematic diagram of the structure of the log processing server provided in an embodiment of the present application;

[0091] Figure 12 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0092] Reference numerals:

[0093] 110-Client; 120-Automatic horizontal expansion system based on access latency;

[0094] 210-Resource Scheduling Server; 220-Domain Name Server; 230-Log Processing Server; 240-Monitoring and Alarm Server; 250-Automatic Horizontal Scaling Server; 260-Service Instance Cluster; 261-Load Balancer; 262-Service Instance; 263-Traffic Mirror; 264-Log Collector;

[0095] 710 - area determination device; 720 - cluster deployment device; 730 - policy acquisition device; 740 - policy execution device;

[0096] 810 - access information acquisition device; 820 - delay monitoring device; 830 - alarm generating device; 840 - alarm sending device;

[0097] 910-alarm acquisition device; 920-strategy generation device; 930-strategy triggering device;

[0098] 1010-region matching device; 1020-domain name resolution device;

[0099] 1110 - log acquisition device; 1120 - access information analysis device; 1130 - access information sending device;

[0100] 1210 - processor; 1220 - memory; 1230 - communication component; 1240 - bus. DETAILED DESCRIPTION

[0101] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0102] In this application, words such as "first" and "second" are used to distinguish between identical or similar items with substantially the same functions and effects. Those skilled in the art will understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit differences. It should be noted that, in this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described in this application as "exemplary" or "for example" should not be interpreted as being more preferred or advantageous than other embodiments or design. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way. In this application, "at least one" refers to one or more, and "more than one" refers to two or more.

[0103] It should be noted that the "at..." in this application can be the instant when a certain situation occurs, or it can be a period of time after a certain situation occurs, and this application does not make specific limitations on this. In addition, the automatic horizontal expansion system and method based on access delay provided in this application is only an example, and the automatic horizontal expansion system and method based on access delay can also include more or less content. The user information (including but not limited to user device information and user personal information, etc.) and data (including but not limited to data for analysis, stored data and displayed data, etc.) involved in one or more embodiments of this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0104] To facilitate a clear description of the technical solution of this application, the following briefly introduces some of the terms and technologies involved in this application:

[0105] Cloud computing services: A service model that provides various computing resources and information, such as servers, storage, databases, and networks, to users via the internet. Cloud computing services allow users to access and use these resources on demand without having to own or maintain physical hardware.

[0106] A service instance is a specific application replica or service unit running on the cloud computing infrastructure. A service instance can be a virtual machine, container, or any other runtime environment, containing all the software and configuration required to run a specific application or perform a specific task. Each service instance can independently process requests and can be dynamically created and released to adapt to changes in load.

[0107] Autoscaling refers to the technology used in cloud computing services to dynamically create and release service instances to cope with changing workloads. If resource utilization of a service instance is excessive (for example, CPU usage or memory usage reaches preset thresholds), potentially leading to performance degradation, the autoscaling mechanism automatically increases the number of service instances (i.e., scales them out) to distribute the load. Conversely, if resource utilization is low, the autoscaling mechanism automatically reduces the number of service instances (i.e., scales them down) to save costs.

[0108] In order to clearly understand the technical solution of the present application, the solution of the prior art is first introduced in detail.

[0109] With the continuous development of internet technology, ensuring the stable operation of cloud computing services and improving the client user experience have become core challenges for cloud computing service providers. To ensure the stable operation of cloud computing services, the current mainstream approach is to monitor service instances running on the server side in real time, focusing on key performance indicators such as central processing unit (CPU) utilization, memory usage, disk input / output (I / O), and network bandwidth. When the system detects an anomaly in a service instance, it automatically triggers the auto-scaling mechanism to dynamically adjust the number of service instances, sharing the load by expanding or replacing the abnormal instance, thereby ensuring the overall availability and stability of the cloud service.

[0110] For example, a common monitoring and optimization strategy involves scaling multiple service instances through elastic scaling to improve the processing power of client access requests and thus reduce access latency. This approach can, to a certain extent, improve the concurrent processing capacity and response speed of cloud computing services.

[0111] However, this monitoring and optimization strategy primarily focuses on the server's own operational status and resource utilization, lacking effective awareness of the actual interaction between the client and the server. While cloud computing services are accessible from the server's perspective, degradation in client access quality due to network latency, geographical distance, or other network link issues often goes undetected. Therefore, in some cases, even if the key performance indicators of a service instance are within normal ranges, long network paths or unstable links can still lead to excessive client access latency, impacting client access quality and user experience.

[0112] In summary, such problems can usually only be identified after the fact through user feedback or log analysis, causing the client to be in an unstable access state for a long time before the problem is located and fixed, thereby affecting user experience and service quality.

[0113] Research has identified a feasible solution to these technical issues, including adding a data tracking or feedback mechanism to the client to collect network latency information for each access request. Cloud computing services can analyze this client-side access information, taking into account different users' geographic locations, access patterns, and usage habits. Based on this analysis, they can dynamically scale multiple service instances through elastic scaling to improve overall service quality.

[0114] However, this approach still has certain limitations. First, because it relies on historical data or periodically aggregated statistical information, its response is delayed, making it difficult to perceive users' current operations in real time and adjust resource allocation in a timely manner. Second, if the user's region has not yet deployed a cloud service resource pool, local capacity expansion cannot be used to improve access latency, limiting the improvement of user experience.

[0115] Based on this, the embodiment of the present application intends to achieve the deployment of service instances nearby by deploying distributed cloud service resource pools in multiple geographical areas, thereby reducing the access delay caused by spatial distance;

[0116] Furthermore, a traffic monitoring system is deployed in each cloud service resource pool to collect access requests from different clients in real time, including key indicators such as the number of requests, response time, and success rate, and use this to determine the access latency of each client;

[0117] Furthermore, through preset policies, clients whose access latency does not comply with the policy are automatically identified, and service instances are created or released in the corresponding resource pool to achieve on-demand elastic scaling.

[0118] The following introduces the application scenarios of the automatic horizontal expansion method based on access latency provided by this application.

[0119] Figure 1 Schematic diagram of the scenario of automatic horizontal expansion based on access latency provided by the embodiment of this application. It should be noted that, Figure 1 What is shown are merely examples of scenarios in which the present application can be applied, to help those skilled in the art understand the technical content of the present application, but does not mean that the present application cannot be used in other devices, systems, environments or scenarios.

[0120] like Figure 1 As shown, the application scenario includes: multiple clients 110 and an automatic horizontal expansion system 120 based on access latency.

[0121] Client 110 refers to a device running on the user side that sends access requests to the cloud and receives responses. It serves as the entry point for communication between the user and the cloud computing service. It is responsible for translating user operations into access requests (e.g., port call requests or page load requests) and presenting the data returned by the cloud to the user. Client 110 can be a computer, mobile phone, tablet, smartwatch, or other IoT device. These devices can have different operating systems, hardware configurations, and user interfaces, and are used to run web browsers, mobile applications (apps), or client agents.

[0122] The access latency-based automatic horizontal scaling system 120 is a distributed system running in the cloud that dynamically adjusts service instances based on client access latency. The access latency-based automatic horizontal scaling system 120 receives access requests from clients 110, monitors and analyzes them in real time, and, based on pre-set policies, triggers an elastic scaling mechanism to improve user experience and service stability.

[0123] The technical solution of the present application is described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0124] Figure 2 This is a schematic diagram of the structure of the automatic horizontal expansion system based on access delay provided in the embodiment of the present application. Figure 1 and Figure 2 As shown, in an embodiment of the present application, the automatic horizontal expansion system 120 based on access delay includes: a resource scheduling server 210, a domain name server 220, a log processing server 230, a monitoring alarm server 240 and an automatic horizontal expansion server 250.

[0125] The resource scheduling server 210 selects a plurality of resident areas from a plurality of preset geographical areas, and deploys a service instance cluster 260 in a resource pool of each resident area.

[0126] Specifically, a geographic region refers to a region where resource pools are physically distributed across different locations. A resident region refers to an area where service demand is concentrated, as assessed based on historical data. It typically has one or a combination of the following characteristics: a high number of potential clients and low average access latency.

[0127] Technical personnel or resource scheduling server 210 evaluate the characteristics of each geographic region and, based on this, select multiple resident regions from the multiple geographic regions. Resource scheduling server 210 then deploys a service instance cluster 260, such as a Kubernetes cluster, within the resource pool of each resident region. The cluster size is determined by factors such as historical load peaks and service level agreement (SLA) requirements, forming a multi-region distributed architecture. It should be noted that deploying or destroying a service instance cluster 260 specifically refers to deploying or destroying the service instances within that cluster.

[0128] The domain name server 220 determines the resident area closest to the corresponding client 110 based on the network address of each client, and resolves each client's access request to the service instance cluster 260 of the corresponding resident area; wherein each network address matches a geographical area.

[0129] Specifically, resource scheduling server 210 registers the network address of the newly deployed service instance with domain name server 220. When client 110 initiates an access request, domain name server 220 traverses the spatial distance between the client's network address and each resident zone, employing a spatial distance-priority resolution mechanism to determine the resident zone closest to each client 110. Each client's access request is then resolved to the service instance cluster 260 in the corresponding resident zone, thereby reducing the access latency impacted by spatial distance.

[0130] The log processing server 230 obtains the log data of each service instance cluster in real time, and obtains the client access information of each service instance cluster based on the log data of each service instance cluster; the client access information of the service instance cluster includes: the access delay of each client accessing the corresponding service instance cluster.

[0131] Specifically, for a particular service instance cluster 260, while it is providing services to a client 110, the log processing server 230 obtains log data from that service instance cluster 260 in real time. This data includes the request five-tuple (which may include the client's network address and port number, the service instance's network address and port number, and the transport layer protocol) and the client request initiation timestamp. The log processing server 230 then calculates the access latency of each client and aggregates the data to generate client access information for each service instance cluster 260.

[0132] The monitoring alarm server 240 monitors the access delay of each client. When it is monitored that the access delay of at least one client is greater than the preset delay threshold, an alarm message is obtained based on the client access information of each service instance cluster, and the alarm message is sent to the automatic horizontal expansion server 250.

[0133] Specifically, the monitoring and alarm server 240 monitors the client access information received from the log processing server 230 in real time. When the monitoring and alarm server 240 detects that the access latency of at least one client exceeds a preset latency threshold, it generates an alarm message (which may include the identifier of the service instance cluster involved, as well as the corresponding client's access information, network address, and access latency), and sends the alarm message to the automatic horizontal scaling server 250, triggering subsequent automatic horizontal scaling policies.

[0134] The automatic horizontal expansion server 250 obtains the automatic horizontal expansion strategy according to the alarm information, and triggers the resource scheduling server 210 to execute the automatic horizontal expansion strategy.

[0135] Specifically, the automatic horizontal expansion server 250 receives the alarm information from the monitoring alarm server 240 and generates a corresponding automatic horizontal expansion strategy based on the alarm information. The specific processing process may include:

[0136] Parse the alarm information and identify the affected service instance cluster 260 and its clients;

[0137] Invoke a preset automatic scaling policy template, or automatically generate a new automatic horizontal scaling policy based on factors such as current load and geographic distribution; exemplary automatic horizontal scaling policies include: expanding the number of service instances in existing resident regions, and deploying new service instance clusters 260 in non-resident regions among multiple geographic regions;

[0138] The resource scheduling server 210 is triggered to execute the automatic horizontal expansion strategy to deploy or adjust the service instance cluster.

[0139] After the resource scheduling server 210 completes the execution, it will register the network address of the newly deployed service instance with the domain name server 220 so that the client 110 can be rerouted to a more optimal path.

[0140] An embodiment of the present application provides an automatic horizontal expansion system based on access delay, which includes: a resource scheduling server, which screens out multiple resident areas and deploys a service instance cluster in the resource pool of each resident area; a domain name server, which resolves the access request of each client to the service instance cluster of the resident area with the closest spatial distance; a log processing server, which obtains the access delay of each client based on the log data of each service instance cluster; a monitoring and alarm server, which obtains an alarm message when monitoring that the access delay of at least one client is greater than the delay threshold; an automatic horizontal expansion server, which obtains an automatic horizontal expansion policy based on the alarm message and triggers the resource scheduling server to execute the automatic horizontal expansion policy. The following technical effects are achieved: by deploying resource pools in multiple geographical areas, the service instances are deployed nearby, reducing the access delay caused by spatial distance; the log data of each service instance cluster is obtained by the log processing server, the access information of different clients is obtained, and the acquisition time of the client's access delay is reduced; the automatic horizontal expansion policy is generated by the automatic horizontal expansion server, and the resource scheduling server is triggered to execute the automatic horizontal expansion policy, thereby achieving on-demand elastic scaling of the service instances.

[0141] In one possible design, the monitoring alarm server 240 includes: a candidate cluster determining device and an alarm generating device;

[0142] The candidate cluster determining device selects at least one candidate service instance cluster from the plurality of service instance clusters 260; wherein the candidate service instance cluster refers to a service instance cluster accessed by a client whose access delay is greater than a delay threshold;

[0143] The alarm generating device obtains alarm information according to the client access information of each candidate service instance cluster.

[0144] Specifically, the specific processing involved in the candidate cluster determination device may include:

[0145] Traverse the client access information of each service instance cluster and obtain the access latency of each client;

[0146] From these clients 110, clients whose access delay is greater than a delay threshold are screened out; wherein the delay threshold can be set according to the user service level, for example, the delay threshold can be 200ms;

[0147] The service instance clusters accessed by the clients with high access latency are screened out from the plurality of service instance clusters 260 , and the screened out service instance clusters are determined as candidate service instance clusters.

[0148] The technical effect of the embodiment of the present application is: at least one candidate service instance cluster is screened out from multiple service instance clusters, and alarm information is obtained based on the client access information of each candidate service instance cluster, thereby avoiding false alarms for irrelevant service instance clusters, reducing the generation time of alarm information, and further reducing the acquisition time of the client's access delay.

[0149] In one possible design, the client access information of the service instance cluster further includes: the number of client accesses and the network address of each client accessing the corresponding service instance cluster;

[0150] The automatic horizontal expansion server 250 includes: an abnormal cluster determination device, an abnormal client determination device, and a policy generation device;

[0151] The abnormal cluster determining device selects at least one abnormal service instance cluster from at least one candidate service instance cluster; the number of client accesses to the abnormal service instance cluster is greater than a preset first threshold;

[0152] An abnormal client determining device, screening out at least one abnormal client from a plurality of clients accessing any abnormal service instance cluster; wherein the network address of the abnormal client does not match the resident area corresponding to the accessed abnormal service instance cluster;

[0153] The policy generating device obtains an automatic horizontal expansion policy according to the network address of at least one abnormal client.

[0154] Specifically, the number of client accesses refers to the total number of client requests accessing the service instance cluster per unit time, such as the number of requests per second (QPS) or the number of concurrent connections, which are used to reflect the load pressure of the service instance cluster.

[0155] The abnormal cluster determination device counts the number of client accesses to each candidate service instance cluster, and screens out candidate service instance clusters whose number of client accesses is greater than a first threshold value from these candidate service instance clusters, and determines these screened candidate service instance clusters as abnormal service instance clusters; wherein the first threshold value can be set according to system performance, for example, QPS=5000.

[0156] For each abnormal service instance cluster, the abnormal client identification device parses the client's network address in the client access information and identifies the client's geographic region by matching the network address with the geographic region. Subsequently, from the multiple clients accessing the abnormal cluster, the device screens out clients whose geographic region does not match the normal region of the abnormal cluster and identifies these clients as abnormal clients.

[0157] The policy generation device performs statistical analysis on the network address distribution of all abnormal clients and generates an automatic horizontal expansion strategy based on this. The automatic horizontal expansion strategy can be to deploy a new service instance cluster in the geographical area where the abnormal clients are located, or to expand the capacity of the existing abnormal service instance cluster to improve its service capabilities.

[0158] Finally, the automatic horizontal scaling strategy is executed by the resource scheduling server to achieve dynamic adjustment of service resources.

[0159] The technical effect of the embodiments of the present application is: determining the abnormal service instance cluster based on the number of client accesses, and determining the abnormal client based on the network address of each client in the abnormal service instance cluster, avoiding blind deployment of new service instance clusters, thereby improving resource utilization and reducing overall deployment costs.

[0160] In one possible design, the resource scheduling server 210 includes: a region determination device and a policy execution device;

[0161] A region determination device, which selects at least one elastic region from a plurality of geographical regions according to an automatic horizontal expansion strategy; wherein the network address of each abnormal client is matched with a elastic region;

[0162] A policy execution device deploys a cluster of elastic service instances in the resource pool of each elastic zone;

[0163] The domain name server 220 resolves the access request of each abnormal client to the elastic service instance cluster of the corresponding elastic zone.

[0164] Specifically, the region determination device parses the abnormal client list in the automatic horizontal scaling policy and matches the abnormal client's geographical region based on its network address. The region determination device then selects the abnormal client's geographical region from multiple geographical regions and identifies the selected geographical region as the elastic region.

[0165] The policy execution device applies for resources in the elastic zone and creates an elastic service instance cluster. Afterwards, the domain name server 220 dynamically updates the Domain Name System (DNS) record to resolve the access request from the abnormal client to the newly created elastic service instance cluster.

[0166] The technical effect of the embodiment of the present application is: by parsing the access request of the abnormal client to the elastic service instance cluster of the corresponding elastic zone, the access delay of the abnormal client caused by the spatial distance is reduced.

[0167] In a possible design, the policy execution device stops running and releases an elastic service instance cluster whose number of client accesses is less than or equal to a preset second threshold.

[0168] Specifically, the policy execution device collects the client access information of the elastic service instance cluster in real time through the log processing server 230, especially the number of client accesses to the elastic service instance cluster.

[0169] When the number of client accesses to a particular elastic service instance cluster is less than or equal to a second threshold value for multiple monitoring cycles, the elastic service instance cluster is marked as a releasable cluster. The second threshold value can be set based on system performance, such as QPS = 100. Domain name server 220 then gradually switches clients accessing the elastic service instance cluster back to the service instance cluster in the resident region they originally accessed, halting and releasing the elastic service instance cluster.

[0170] The technical effect of the embodiment of the present application is: according to the number of client accesses to the elastic service instance cluster, the operation of part of the elastic service instance cluster is stopped and released, thereby releasing idle resources and reducing the overall deployment cost.

[0171] In one possible design, the resource scheduling server 210 further includes: a region determining device;

[0172] The region determining device obtains the client distribution information of each geographical region and selects a plurality of resident regions from the plurality of geographical regions according to the client distribution information of each geographical region.

[0173] Specifically, client distribution information describes the distribution characteristics of clients in a geographic area based on dimensions such as number, activity, and access latency. The client distribution information is positively correlated with client density. This client distribution information can be automatically generated by the region determination device based on raw client logs, or it can be independently generated by a technician and input into the region determination device.

[0174] After obtaining the client distribution information of each geographical area, the area determination device excludes geographical areas where the resource pool is unavailable from the multiple geographical areas, and selects multiple resident areas according to the client distribution information of each geographical area.

[0175] The technical effect of the embodiments of the present application is: deploying a service instance cluster in a geographical area with a high concentration of clients, thereby reducing the access delay of the client caused by cross-region access.

[0176] Figure 3 This is a schematic diagram of the structure of the service instance cluster provided in the embodiment of this application. Figure 2 and Figure 3 As shown, in one possible design, a service instance cluster 260 includes: a load balancer 261, multiple service instances 262, a traffic mirror 263, and a log collector 264;

[0177] The domain name server 220 resolves each client's access request to the load balancer 261 of the corresponding resident area;

[0178] For any service instance cluster 260 , the load balancer 261 distributes the client access request resolved by the domain name server 220 to multiple service instances 262 ;

[0179] Traffic mirror 263, mirrors the traffic between each service instance 262 and the corresponding client to log collector 264;

[0180] The log collector 264 obtains the log data of the corresponding service instance cluster according to the traffic mirrored by the log collector 264 .

[0181] Specifically, load balancer 261 is the entry component of service instance cluster 260. When a client initiates an access request, domain name server 220 returns the corresponding resident load balancer 261 through DNS resolution. Upon receiving the client's access request, load balancer 261 selects an appropriate load balancer 261 based on service instance weights, such as the inverse weight of CPU usage, and distributes the client's access request to the corresponding load balancer 261.

[0182] The traffic mirrorer 263 replicates the traffic between each service instance 262 and the corresponding client in real time through port mirroring or proxy mode, and sends it to the independent network plane of the log collector 264 to avoid interfering with production traffic.

[0183] Log collector 264 parses the mirrored traffic, extracting data such as the request five-tuple and the client request initiation timestamp. Combined with the metadata of service instance 262, it generates log data for service instance cluster 260. Log collector 264 then transmits the log data for service instance cluster 260 to log processing server 230 via a message queue.

[0184] The technical effect of the embodiment of the present application is: covering the access requests of all clients by mirroring the traffic, eliminating the monitoring blind spots, and ensuring the integrity of the client's access delay.

[0185] Figure 4 Schematic diagram of the process of the automatic horizontal expansion method based on access delay provided in the embodiment of the present application Figure 1 .like Figure 4 As shown, an embodiment of the present application further provides an automatic horizontal expansion method based on access delay, which is applied to a resource scheduling server of an automatic horizontal expansion system based on access delay, and the system also includes an automatic horizontal expansion server. The method includes:

[0186] S401: Filter out multiple resident areas from multiple preset geographical areas.

[0187] S402: Deploy a service instance cluster in the resource pool of each resident area.

[0188] S403: When triggered by the automatic horizontal scaling server, obtain the automatic horizontal scaling policy sent by the automatic horizontal scaling server.

[0189] S404: Execute the automatic horizontal expansion strategy.

[0190] In one possible design, S404 includes:

[0191] S4041. According to the automatic horizontal expansion strategy, at least one elastic zone is selected from multiple geographical regions; wherein the network address of each abnormal client is matched with an elastic zone.

[0192] S4042. Deploy an elastic service instance cluster in the resource pool of each elastic zone.

[0193] In one possible design, after S4042, the method further includes:

[0194] S4043. Stop running and destroy the elastic service instance cluster whose number of client accesses is less than or equal to a preset second threshold.

[0195] In one possible design, S401 includes:

[0196] S4011. Obtain client distribution information in each geographical area.

[0197] S4012. Filter out multiple resident areas from multiple geographical areas based on the client distribution information of each geographical area.

[0198] The automatic horizontal expansion method based on access delay provided in the embodiment of the present application is used to achieve the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0199] Figure 5 Schematic diagram of the process of the automatic horizontal expansion method based on access delay provided in the embodiment of the present application Figure 2 .like Figure 5 As shown, an embodiment of the present application further provides an automatic horizontal expansion method based on access delay, which is applied to a monitoring alarm server of an automatic horizontal expansion system based on access delay, wherein the system also includes a log processing server and an automatic horizontal expansion server. The method includes:

[0200] S501. Acquire client access information of each service instance cluster generated by a log processing server in real time; the client access information of the service instance cluster includes: access delay of each client accessing the corresponding service instance cluster.

[0201] S502: Monitor the access delay of each client.

[0202] S503: When it is monitored that the access delay of at least one client is greater than a preset delay threshold, alarm information is obtained according to the client access information of each service instance cluster.

[0203] S504: Send the alarm information to the automatic horizontal expansion server.

[0204] In one possible design, S503 includes:

[0205] S5031. Filter out at least one candidate service instance cluster from multiple service instance clusters; wherein the candidate service instance cluster refers to a service instance cluster accessed by a client whose access latency is greater than a latency threshold.

[0206] S5032: Obtain alarm information based on client access information of each candidate service instance cluster.

[0207] The automatic horizontal expansion method based on access delay provided in the embodiment of the present application is used to achieve the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2The product embodiments shown are similar and will not be described in detail in the present application.

[0208] Figure 6 Schematic diagram of the process of the automatic horizontal expansion method based on access delay provided in the embodiment of the present application Figure 3 .like Figure 6 As shown, an embodiment of the present application further provides an automatic horizontal expansion method based on access delay, which is applied to an automatic horizontal expansion server of an automatic horizontal expansion system based on access delay, wherein the system also includes a log processing server and a resource scheduling server. The method includes:

[0209] S601: When a log processing server monitors that the access delay of at least one client is greater than a preset delay threshold, obtain alarm information sent by the log processing server.

[0210] S602: Obtain an automatic horizontal expansion strategy based on the alarm information.

[0211] S603: Trigger the resource scheduling server to execute the automatic horizontal expansion strategy.

[0212] In one possible design, the client access information of the service instance cluster further includes: the number of client accesses and the network address of each client accessing the corresponding service instance cluster;

[0213] Then S602 includes:

[0214] S6021. Filter out at least one abnormal service instance cluster from at least one candidate service instance cluster; the number of client accesses to the abnormal service instance cluster is greater than a preset first threshold.

[0215] S6022. Filter out at least one abnormal client from multiple clients accessing any abnormal service instance cluster; wherein the network address of the abnormal client does not match the resident area corresponding to the accessed abnormal service instance cluster.

[0216] S6023. Obtain an automatic horizontal expansion strategy according to the network address of at least one abnormal client.

[0217] The automatic horizontal expansion method based on access delay provided in the embodiment of the present application is used to achieve the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0218] The embodiment of the present application further provides an automatic horizontal expansion method based on access delay, which is applied to a domain name server of an automatic horizontal expansion system based on access delay. The method includes:

[0219] S701. Upon receiving an access request from a client, determine the resident area closest to the corresponding client space based on the network address of each client; wherein each network address matches a geographical area.

[0220] S702: Parse each client's access request to the corresponding service instance cluster in the resident area.

[0221] In one possible design, the method further includes:

[0222] S703: When receiving an access request from an abnormal client, determine the elastic zone closest to the corresponding abnormal client according to the network address of each abnormal client.

[0223] S704: Parse the access request of each abnormal client to the elastic service instance cluster of the corresponding elastic zone.

[0224] In one possible design, S702 includes:

[0225] S7021. Parse each client's access request to the load balancer of the corresponding resident area.

[0226] The automatic horizontal expansion method based on access delay provided in the embodiment of the present application is used to achieve the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0227] The embodiment of the present application further provides an automatic horizontal expansion method based on access delay, which is applied to a log processing server of an automatic horizontal expansion system based on access delay. The method includes:

[0228] S801. Obtain log data of each service instance cluster in real time.

[0229] S802: Obtain client access information of each service instance cluster based on the log data of each service instance cluster.

[0230] S803: Send the client access information of each service instance cluster to the monitoring alarm server.

[0231] The automatic horizontal expansion method based on access delay provided in the embodiment of the present application is used to achieve the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0232] Figure 7This is a schematic diagram of the structure of the resource scheduling server provided in the embodiment of the present application. Figure 7 As shown, an embodiment of the present application further provides a resource scheduling server, the resource scheduling server comprising:

[0233] The region determining means 710 is used to select a plurality of permanent regions from a plurality of preset geographical regions;

[0234] Cluster deployment means 720, for deploying a service instance cluster in a resource pool in each resident area;

[0235] The policy acquisition means 730 is used to acquire the automatic horizontal expansion policy sent by the automatic horizontal expansion server when triggered by the automatic horizontal expansion server;

[0236] The policy execution device 740 is used to execute the automatic horizontal expansion policy.

[0237] In one possible design, the region determining means 710 is configured to select at least one elastic region from a plurality of geographical regions according to an automatic horizontal scaling policy; wherein the network address of each abnormal client is matched with a elastic region;

[0238] The policy execution device 740 is used to deploy an elastic service instance cluster in the resource pool of each elastic zone.

[0239] In a possible design, the policy execution device 740 is used to stop running and destroy an elastic service instance cluster whose number of client accesses is less than or equal to a preset second threshold.

[0240] In one possible design, the resource scheduling server further includes:

[0241] Distribution information acquisition means, for acquiring client distribution information in each geographical area;

[0242] The region determining means 710 is configured to select a plurality of resident regions from a plurality of geographical regions according to the client distribution information of each geographical region.

[0243] The resource scheduling server provided in the embodiment of the present application is used to implement the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0244] Figure 8 This is a schematic diagram of the structure of the monitoring alarm server provided in the embodiment of the present application. Figure 8 As shown, the embodiment of the present application further provides a monitoring alarm server, which includes:

[0245] The access information acquisition device 810 is used to obtain the client access information of each service instance cluster generated by the log processing server in real time; the client access information of the service instance cluster includes: the access delay of each client accessing the corresponding service instance cluster;

[0246] The delay monitoring device 820 is used to monitor the access delay of each client;

[0247] The alarm generating device 830 is configured to obtain alarm information based on the client access information of each service instance cluster when the access delay of at least one client is monitored to be greater than a preset delay threshold;

[0248] The alarm sending device 840 is used to send the alarm information to the automatic horizontal expansion server.

[0249] In one possible design, the monitoring alarm server further includes:

[0250] A candidate cluster determining device is used to screen out at least one candidate service instance cluster from multiple service instance clusters; wherein the candidate service instance cluster refers to a service instance cluster accessed by a client whose access delay is greater than a delay threshold;

[0251] The alarm generating device 830 is used to obtain alarm information according to the client access information of each candidate service instance cluster.

[0252] The monitoring alarm server provided in the embodiment of the present application is used to implement the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0253] Figure 9 This is a schematic diagram of the structure of the automatic horizontal expansion server provided in the embodiment of the present application. Figure 9 As shown, an embodiment of the present application further provides an automatic horizontal expansion server, the automatic horizontal expansion server comprising:

[0254] The alarm obtaining means 910 is configured to obtain the alarm information sent by the log processing server when the log processing server monitors that the access delay of at least one client is greater than a preset delay threshold;

[0255] Strategy generating means 920, for obtaining an automatic horizontal expansion strategy based on the alarm information;

[0256] The policy triggering device 930 is used to trigger the resource scheduling server to execute the automatic horizontal expansion policy.

[0257] In one possible design, automatically scaling servers horizontally also includes:

[0258] The abnormal cluster determining device is used to screen out at least one abnormal service instance cluster from at least one candidate service instance cluster; the number of client accesses to the abnormal service instance cluster is greater than a preset first threshold;

[0259] An abnormal client determining device is used to screen out at least one abnormal client from multiple clients accessing any abnormal service instance cluster; wherein the network address of the abnormal client does not match the resident area corresponding to the accessed abnormal service instance cluster;

[0260] The policy generating device 920 is used to obtain an automatic horizontal expansion policy according to the network address of at least one abnormal client.

[0261] The automatic horizontal expansion server provided in the embodiment of the present application is used to implement the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0262] Figure 10 This is a schematic diagram of the structure of the domain name server provided in the embodiment of this application. Figure 10 As shown, the embodiment of the present application further provides a domain name server, which includes:

[0263] The region matching means 1010 is configured to determine, upon receiving an access request from a client, the resident region closest to the corresponding client based on the network address of each client; wherein each network address is matched with a geographical region;

[0264] The domain name resolution device 1020 is used to resolve each client's access request to the service instance cluster of the corresponding resident area.

[0265] In one possible design, the region matching device 1010 is configured to, upon receiving an access request from an abnormal client, determine, based on the network address of each abnormal client, the elastic region that is closest to the corresponding abnormal client in spatial distance;

[0266] The domain name resolution device 1020 is used to resolve the access request of each abnormal client to the elastic service instance cluster of the corresponding elastic zone.

[0267] In one possible design, the domain name resolution device 1020 is used to resolve each client's access request to the load balancer of the corresponding resident area.

[0268] The domain name server provided in the embodiment of the present application is used to implement the following Figure 2The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0269] Figure 11 This is a schematic diagram of the structure of the log processing server provided in the embodiment of the present application. Figure 11 As shown, an embodiment of the present application further provides a log processing server, the log processing server comprising:

[0270] Log acquisition device 1110, used to acquire log data of each service instance cluster in real time;

[0271] The access information parsing device 1120 is used to obtain the client access information of each service instance cluster based on the log data of each service instance cluster;

[0272] The access information sending means 1130 is used to send the client access information of each service instance cluster to the monitoring alarm server.

[0273] The log processing server provided in the embodiment of the present application is used to implement the following Figure 2 The technical solution of the product embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The product embodiments shown are similar and will not be described in detail in the present application.

[0274] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. This embodiment of the present application also provides an electronic device, such as Figure 12 As shown, the electronic device includes: at least one processor 1210 and a memory 1220. The electronic device also includes a communication component 1230. The processor 1210, the memory 1220 and the communication component 1230 are connected via a bus 1240.

[0275] During the specific implementation process, at least one processor 1210 executes the computer-executable instructions stored in the memory 1220, so that the at least one processor 1210 is used to implement the automatic horizontal expansion method based on access latency of the above embodiment.

[0276] The specific implementation process of the processor 1210 can be found in the above-mentioned method embodiment. Its implementation principle and technical effects are similar, and the embodiments of this application will not be repeated here.

[0277] In the above embodiments, it should be understood that the processor 1210 may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), or application-specific integrated circuits (ASICs). A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the application may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0278] The memory 1220 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.

[0279] Bus 1240 can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Bus 1240 can be divided into an address bus, a data bus, a control bus, and so on. For ease of illustration, the bus 1240 in the drawings of this application is not limited to a single bus or a single type of bus.

[0280] The above functions implemented by the electronic device and the main control device have been used to introduce the solutions provided in the embodiments of the present application. It is understandable that, in order to implement the above functions, the electronic device or the main control device includes hardware structures and / or software modules corresponding to the execution of each function. In combination with the units and algorithm steps of the various examples described in the embodiments disclosed in the embodiments of the present application, the embodiments of the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiments of the present application.

[0281] The present application also provides a computer-readable storage medium storing computer-executable instructions. When executed by a processor, the computer-executable instructions implement the aforementioned automatic horizontal scaling method based on access latency. In the specific implementation of the aforementioned automatic horizontal scaling method based on access latency, each module may be implemented as a processor.

[0282] The readable storage medium may be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0283] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Alternatively, the readable storage medium may be an integral part of the processor. The processor and the readable storage medium may be located in an application-specific integrated circuit (ASIC). Alternatively, the processor and the readable storage medium may be present as discrete components in an electronic device or a host control device.

[0284] An embodiment of the present application also provides a computer program product, including a computer program. When the computer program is executed by a processor, it is used to implement the automatic horizontal expansion method based on access latency of the above embodiment.

[0285] The computer program is stored in a readable storage medium. At least one processor can read the computer program from the readable storage medium, and at least one processor can execute the computer program to perform the solution provided in any of the above embodiments.

[0286] Those skilled in the art will appreciate that all or part of the steps in implementing the aforementioned embodiments of the application can be accomplished by hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the aforementioned method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0287] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the scope of protection of the present application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solution of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solution to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. An automatic horizontal expansion system based on access latency, characterized in that: include: Resource scheduling server, domain name server, log processing server, monitoring and alarm server, and automatic horizontal expansion server; The resource scheduling server selects a plurality of resident areas from a plurality of preset geographical areas, and deploys a service instance cluster in a resource pool of each of the resident areas; The domain name server determines, based on the network address of each client, a resident region that is spatially closest to the corresponding client, and resolves each client's access request to a service instance cluster in the corresponding resident region; wherein each network address matches one of the geographical regions; The log processing server obtains log data of each of the service instance clusters in real time, and obtains client access information of each of the service instance clusters based on the log data of each of the service instance clusters; the client access information of the service instance cluster includes: the access delay of each of the clients accessing the corresponding service instance cluster; The monitoring and alarm server monitors the access delay of each of the clients, and when it is monitored that the access delay of at least one of the clients is greater than a preset delay threshold, obtains an alarm message based on the client access information of each of the service instance clusters, and sends the alarm message to the automatic horizontal scaling server; The automatic horizontal expansion server obtains an automatic horizontal expansion strategy according to the alarm information, and triggers the resource scheduling server to execute the automatic horizontal expansion strategy.

2. The system according to claim 1, wherein: The monitoring alarm server comprises: a candidate cluster determining device and an alarm generating device; The candidate cluster determining device selects at least one candidate service instance cluster from the plurality of service instance clusters; wherein the candidate service instance cluster refers to a service instance cluster accessed by a client whose access delay is greater than the delay threshold; The alarm generating device obtains alarm information according to the client access information of each candidate service instance cluster.

3. The system according to claim 2, characterized in that The client access information of the service instance cluster further includes: the number of client accesses and the network address of each client accessing the corresponding service instance cluster; The automatic horizontal expansion server includes: abnormal cluster determination means, abnormal client determination means and policy generation means; The abnormal cluster determining device selects at least one abnormal service instance cluster from at least one candidate service instance cluster; the number of client accesses to the abnormal service instance cluster is greater than a preset first threshold; The abnormal client determining device selects at least one abnormal client from a plurality of clients accessing any of the abnormal service instance clusters; wherein the network address of the abnormal client does not match the resident area corresponding to the accessed abnormal service instance cluster; The policy generating device obtains the automatic horizontal expansion policy according to the network address of the at least one abnormal client.

4. The system according to claim 3, characterized in that The resource scheduling server includes: a region determination device and a policy execution device; The region determination device selects at least one elastic region from the multiple geographical regions according to the automatic horizontal expansion strategy; wherein the network address of each abnormal client matches one of the elastic regions; The policy execution device deploys an elastic service instance cluster in the resource pool of each elastic zone; The domain name server resolves the access request of each abnormal client to the elastic service instance cluster of the corresponding elastic zone.

5. The system according to claim 4, characterized in that The policy execution device stops running and destroys the elastic service instance cluster whose number of client accesses is less than or equal to a preset second threshold.

6. The system according to claim 1, wherein: The resource scheduling server further includes: an area determination device; The region determining device obtains the client distribution information of each of the geographical regions, and screens out the multiple resident regions from the multiple geographical regions based on the client distribution information of each of the geographical regions.

7. The system according to claim 1, wherein: The service instance cluster includes: a load balancer, multiple service instances, a traffic mirror and a log collector; The domain name server resolves the access request of each client to the load balancer of the corresponding resident area; For any of the service instance clusters, the load balancer distributes the client's access request resolved by the domain name server to the multiple service instances; The traffic mirrorer mirrors the traffic between each of the service instances and the corresponding client to the log collector; The log collector obtains log data of the corresponding service instance cluster according to the traffic mirrored by the log collector.

8. An automatic horizontal expansion method based on access delay, characterized in that: The method is applied to a resource scheduling server of an automatic horizontal expansion system based on access delay, and the system further includes an automatic horizontal expansion server. The method includes: Filter multiple resident areas from multiple preset geographical areas; Deploying a service instance cluster in a resource pool in each of the resident areas; When triggered by the automatic horizontal expansion server, obtaining the automatic horizontal expansion policy sent by the automatic horizontal expansion server; The automatic horizontal scaling policy is executed.

9. An automatic horizontal expansion method based on access delay, characterized in that: The method is applied to a monitoring alarm server of an automatic horizontal expansion system based on access delay, and the system further includes a log processing server and an automatic horizontal expansion server. The method includes: Acquire in real time the client access information of each service instance cluster generated by the log processing server; the client access information of the service instance cluster includes: the access delay of each client accessing the corresponding service instance cluster; Monitoring the access delay of each client; When it is monitored that the access delay of at least one of the clients is greater than a preset delay threshold, an alarm message is obtained according to the client access information of each of the service instance clusters; The alarm information is sent to the automatic horizontal expansion server.

10. An automatic horizontal expansion method based on access delay, characterized in that: The method is applied to an automatic horizontal expansion server of an automatic horizontal expansion system based on access delay, and the system further includes a log processing server and a resource scheduling server. The method includes: When the log processing server monitors that the access delay of at least one client is greater than a preset delay threshold, obtaining alarm information sent by the log processing server; Obtaining an automatic horizontal expansion strategy based on the alarm information; The resource scheduling server is triggered to execute the automatic horizontal expansion strategy.

Images