Request control method, electronic device, storage medium and computer program product

Abstract

The present application provides a request control method, an electronic device, a storage medium and a computer program product, and relates to the field of computer technology. The method includes: counting resource utilization at preset time intervals; if the resource utilization is not less than a first preset utilization threshold, executing a defense strategy; the defense strategy indicates that the service node no longer receives requests and only continues to process the requests being processed; based on the changes in resource utilization counted during the execution of the defense strategy, a request receiving strategy for the service node is generated; the request receiving strategy indicates the number of requests received by the service node within a request receiving cycle; and the received requests are controlled according to the request receiving strategy. Starting from the dimension of the service node, the present application controls the number of requests received based on the actual situation of its own request processing, so as to maximize the utilization of the resources of the service node and improve the stability of the service node in processing requests.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a request control method, electronic device, storage medium and computer program product. Background Art

[0002] In a microservice architecture, when the system faces a large number of user accesses and the load is too high, a load balancing strategy is usually used to evenly distribute requests to different service nodes of the same service for processing. However, since different requests actually use different resources when processing, and the load balancing strategy can only perceive the load of the service node before distributing the request, it cannot perceive the situation when the service node actually processes the request, which causes the service node to crash during the request processing process, affecting the stability of the service node in processing requests. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a request control method, an electronic device, a storage medium and a computer program product, which are used to control the number of received requests based on the actual situation of the service node's own processing requests, so as to maximize the utilization of the service node resources and improve the stability of the service node in processing requests.

[0004] In a first aspect, an embodiment of the present application provides a request control method, which is applied to a service node, and the method includes: counting resource utilization at preset time intervals; if the resource utilization is not less than a first preset utilization threshold, executing a defense strategy; the defense strategy indicates that the service node no longer receives requests and only continues to process the requests being processed; based on the changes in resource utilization counted during the execution of the defense strategy, a request receiving strategy for the service node is generated; the request receiving strategy indicates the number of requests received by the service node within a request receiving cycle; and the received requests are controlled according to the request receiving strategy.

[0005] The embodiment of the present application counts resource utilization at preset time intervals to understand the real-time status of the service node, and executes a defense strategy when the resource utilization is greater than a first preset utilization threshold, so that the service node no longer receives new requests and only processes the requests being processed to avoid the entry of new requests causing pressure on the service node, thereby improving the stability of the service node in processing requests. Furthermore, the request receiving strategy of the service node generated according to the changes in resource utilization counted during the execution of the defense strategy controls the received requests, so that the resources of the service node are fully utilized and the phenomenon of crash is avoided, thereby improving the stability of the service node, thereby improving the stability of the service node in processing requests.

[0006] In some embodiments, a request receiving strategy for a service node is generated based on changes in resource utilization counted during the execution of a defense strategy, including: if the latest counted resource utilization is less than a second preset utilization threshold, continue to receive requests; use the time when requests are started to be received as the start time; in the process of receiving requests, if the latest counted resource utilization is not less than a third preset utilization threshold, stop receiving requests; use the time when requests are stopped as the end time; the second preset utilization threshold is less than the third preset utilization threshold; the third preset utilization threshold is less than or equal to the first preset utilization threshold; the start time to the end time constitute a first request receiving cycle; and the request receiving strategy for a service node is generated based on the time length of the first request receiving cycle and the resource utilization during the request processing process.

[0007] The embodiment of the present application determines when to continue receiving requests and when to stop receiving requests based on the changes in resource utilization counted during the execution of the defense strategy, so that the service node will not crash and its resources will not be idle, thereby improving the stability of the service node in processing requests on the basis of maximizing the use of service node resources.

[0008] In some embodiments, a request receiving strategy for a service node is generated based on the time length of a first request receiving cycle and the resource utilization during request processing, including: if the latest statistical resource utilization is not less than a first preset utilization threshold before being less than a second preset utilization threshold, then obtaining a first time when the resource utilization first rises to the first preset utilization threshold, and a second time when the resource utilization last drops below the first preset utilization threshold before being less than the second preset utilization threshold; if the time difference between the first time and the second time is greater than or equal to the time length of the first request receiving cycle, then generating a request receiving strategy for the service node based on the number of requests received in the first request receiving cycle.

[0009] The embodiment of the present application determines whether the service node is in a high-load state for a long time by judging the time difference between the first time and the second time and the time length of the first request receiving cycle, and generates a request receiving strategy for the service node based on the number of requests received in the first request receiving cycle. The received requests are controlled by the request receiving strategy to improve the stability of the service node, thereby improving the stability of the service node in processing requests.

[0010] In some embodiments, a request receiving strategy for a service node is generated based on the number of requests received in a first request receiving cycle, including: dividing the time length of the first request receiving cycle into multiple intervals according to a preset interval size; determining the median time point of each interval; and determining the number of requests received by the service node at each time point in the interval corresponding to the current request receiving cycle based on the number of requests received corresponding to the median time point in the first request cycle and the preset interval size.

[0011] In an embodiment of the present application, the time difference between the first time and the second time is greater than or equal to the time length of the first request receiving cycle, indicating that the service node is in a high-load state for a long time. At this time, by reducing the number of requests received in the current request receiving cycle, the load pressure of the service node is reduced and the stability of the service node is improved, thereby improving the stability of the service node in processing requests.

[0012] In some embodiments, the method further includes: if the time difference between the first time and the second time is less than the time length of the first request receiving cycle, obtaining the third time when the resource utilization last dropped below a fourth preset utilization threshold before being less than the second preset utilization threshold; the fourth preset utilization threshold is greater than the second preset utilization threshold and less than the third preset utilization threshold; and generating a request receiving strategy for the service node based on the first time, the second time, the third time and the number of requests received within the first request receiving cycle.

[0013] In an embodiment of the present application, the time difference between the first time and the second time is less than the time length of the first request receiving cycle, and a request receiving strategy of the service node is generated based on the first time, the second time, the third time and the number of requests received within the first request receiving cycle, so that the generated request receiving strategy is generated based on the actual situation of the service node processing the request, so that the received requests can be more effectively controlled and the stability of the service node can be improved, thereby improving the stability of the service node processing requests.

[0014] In some embodiments, a request receiving strategy for a service node is generated based on the number of requests received within a first time, a second time, a third time, and a first request receiving cycle, including: taking the center time point of the start time and the third time as the first midpoint; taking the center time point of the first time and the second time as the second midpoint; if the absolute difference between the second midpoint and the first midpoint is less than a preset threshold; then determining a control point based on the second midpoint, the start time, the end time, and the third time; determining the number of first requests received by the control point within a preset time length in the first request receiving cycle with the control point as the center; multiplying the first request number by a preset ratio to obtain the number of requests received within a preset time length in the current request receiving cycle with the control point as the center.

[0015] In an embodiment of the present application, when the time difference between the first time and the second time is less than the time length of the first request receiving cycle, the control point is calculated, and within the current request receiving cycle, only the number of requests received in the time area near the control point is reduced, thereby improving the stability of the service node in processing requests on the basis of maximizing the use of service node resources.

[0016] In some embodiments, the method also includes: if the latest statistical resource utilization is not greater than or equal to the first preset utilization threshold before being less than the second preset utilization threshold, then multiplying the number of requests received in the first request receiving cycle by a preset multiple to obtain the number of requests received in the current request receiving cycle.

[0017] In an embodiment of the present application, after receiving a request in the first request receiving cycle, if the resource utilization of the service node is never greater than or equal to the first preset utilization threshold during the process of processing the request, it indicates that the resource utilization of the service node has always been idle. Therefore, the number of received requests can be increased in the current request receiving cycle to maximize the utilization of the service node's resources and improve the stability of the service node in processing requests.

[0018] In some embodiments, executing a defense strategy includes: returning a preset prompt message to the upstream caller so that the upstream caller can re-initiate a request later based on the preset prompt message; and if there are pending requests that have not been processed in the memory of the service node, deleting the pending requests.

[0019] The embodiment of the present application executes a defense strategy to avoid unprocessed requests or receiving new requests that put pressure on the service node, thereby protecting the service node and improving the stability of the service node, thereby improving the stability of the service node in processing requests.

[0020] In some embodiments, the received requests are controlled according to the request receiving policy, including: if the number of requests issued by the upstream caller is greater than the number of requests in the request receiving policy, then generating a request receiving policy for the next request receiving cycle based on the resource utilization during the request processing, until the latest statistical resource utilization is less than the fifth preset utilization threshold, and the number of requests issued by the upstream caller in the next request receiving cycle is less than or equal to the number of requests in the request receiving policy for the corresponding request receiving cycle, and receiving all requests issued by the upstream caller.

[0021] In an embodiment of the present application, when the number of requests sent by the upstream caller is greater than the number of requests in the request receiving strategy, the request receiving strategy is iteratively generated until the service node resource utilization is less than the fifth preset utilization threshold, and the number of requests sent by the upstream caller is less than the number of requests in the request receiving strategy, so as to maximize the utilization of the service node resources and maintain the stability of the service node, thereby improving the stability of the service node in processing requests.

[0022] In the second aspect, an embodiment of the present application provides a request control device, which includes: a statistical module, which is used to count resource utilization at preset time intervals; an execution module, which is used to execute a defense strategy if the resource utilization is not less than a first preset utilization threshold; the defense strategy indicates that the service node no longer receives requests and only continues to process the requests being processed; a generation module, which is used to generate a request receiving strategy for the service node based on the changes in resource utilization counted during the execution of the defense strategy; the request receiving strategy indicates the number of requests received by the service node within a request receiving cycle; and a control module, which is used to control the received requests according to the request receiving strategy.

[0023] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, a storage medium and a bus, wherein the processor and the memory communicate with each other through the bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method steps of the first aspect.

[0024] In a fourth aspect, an embodiment of the present application provides a non-transitory computer-readable storage medium, comprising: the computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the method steps of the first aspect.

[0025] In a fifth aspect, an embodiment of the present application provides a computer program product, including: a computer program, which executes the method steps of the first aspect when executed by a processor.

[0026] The beneficial effects of this application are as follows:

[0027] The present application counts resource utilization at preset time intervals to understand the real-time status of the service node, and executes a defense strategy when the resource utilization is greater than a first preset utilization threshold, so that the service node no longer receives new requests and only processes the requests being processed to avoid the pressure on the service node caused by the entry of new requests, thereby improving the stability of the service node in processing requests. Furthermore, the request receiving strategy of the service node generated according to the changes in resource utilization counted during the execution of the defense strategy controls the received requests, so that the resources of the service node are fully utilized and the phenomenon of crash is avoided, thereby improving the stability of the service node, thereby improving the stability of the service node in processing requests.

[0028] Other features and advantages of the present application will be described in the subsequent description, and in part will become apparent from the description, or will be understood by implementing the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 A flowchart of a request control method provided in an embodiment of the present application;

[0031] Figure 2 A schematic diagram of the structure of a request control device provided in an embodiment of the present application;

[0032] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0034] It should be noted that all technical and scientific terms used in this document have the same meanings as those generally understood by technicians in the technical field to which this application belongs; the terms used in this document are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0035] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0036] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0037] Under the microservice architecture, the system faces a large number of user accesses. When the load is too high, a load balancing strategy is usually used to evenly distribute the requests to different service nodes of the same service for processing, so as to improve the stability of the system.

[0038] However, in actual scenarios, different requests use different resources when they are actually processed. For example, when a service node processes request A, it only needs to perform a judgment action. At this time, it only needs to occupy the resources of the service node itself, and does not need to obtain external third-party data. In addition, since the logic is simple, it can be processed quickly, so it will not occupy the resources of the service node for too long. When the service node processes request B, it needs to access the database to read the required data. At this time, in addition to occupying the resources of the service node itself, it also needs to consider the resource situation of the database. If the database has no idle connections at this time, request B needs to wait for processing. The waiting process will continue to occupy the resources of the service node, so that the resources of the service node cannot be released.

[0039] When controlling the distribution of requests, the load balancing strategy can only perceive the resource utilization of the service node before distributing the request, but cannot perceive the situation when the service node actually processes the request. That is, the load balancing strategy can only control the number of distributed requests, but cannot control the type of distributed requests.

[0040] If the load balancing strategy distributes requests to service nodes based on the perceived resource utilization of the service nodes, but these requests are time-consuming and resource-consuming, the service nodes may crash, affecting the stability of the service nodes in processing requests.

[0041] Based on this, an embodiment of the present application provides a request control method, which is used to control the number of requests received from the service node based on the actual situation of its own request processing, so as to maximize the utilization of service node resources and improve the stability of the service node in processing requests.

[0042] Figure 1A flowchart of a request control method provided in an embodiment of the present application, which is applied to a service node. It is understandable that the service node can be deployed in a terminal device (also referred to as an electronic device) and a server; the terminal device can specifically be a smart phone, a tablet computer, a computer, a personal digital assistant (PDA), etc.; the server can specifically be an application server or a Web server. To facilitate understanding of the technical solution provided in the embodiment of the present application, the application scenario of the request control method provided in the embodiment of the present application is introduced by taking the server as the execution subject as an example.

[0043] like Figure 1 As shown, the method includes:

[0044] Step S101: The server collects statistics on resource utilization at preset time intervals.

[0045] During the execution process, the preset time interval is a preset value, for example, the resource utilization rate may be counted once every 1 second, or once every 30 milliseconds.

[0046] Since the method is applied to service nodes, and different service nodes process requests differently, the preset time intervals set for different service nodes may be different. Optionally, custom settings may be made based on the historical request processing of each service node.

[0047] Resources refer to all resources that may cause the service node to become a bottleneck. Resources include at least internal resources, which refer to at least one of the memory, CPU, disk, network bandwidth, etc. of the physical machine where the service node is located.

[0048] Resources may also include external resources, which refer to at least one of database resources, middleware resources, and interfaces provided by third-party organizations.

[0049] The interface provided by a third-party organization refers to the service interface of a third party that needs to be accessed during the request processing. For example, if a request needs to access the interface provided by the Baidu server, the situation of the interface needs to be considered.

[0050] Therefore, the resource utilization counted by the server at preset time intervals may include only the internal resource utilization of the service node itself, and may also include the internal resource utilization of the service node itself and the external resource utilization that the service node needs to use when processing requests.

[0051] When counting the internal resource utilization, the server runs the pre-edited code at a preset time interval to obtain the internal resource utilization of the service node.

[0052] For example, in Java, the CPU utilization of the JVM can obtain operating system-level information through com.sun.management.OperatingSystemMXBean to calculate the CPU utilization.

[0053] Memory utilization You can obtain memory utilization through java.lang.management.MemoryMXBean and java.lang.management.MemoryUsage.

[0054] For external resource utilization of database resources and middleware resources, the resource utilization of database resources or middleware resources can also be obtained by running pre-edited codes.

[0055] The external resource utilization rate of interfaces and human resources provided by third-party organizations can be determined based on the actual processing of historical requests.

[0056] Step S102: If the resource utilization is not less than the first preset utilization threshold, the server executes a defense strategy; the defense strategy indicates that the service node no longer receives requests and only continues to process the requests being processed.

[0057] The first preset utilization rate threshold is a preset value.

[0058] The resource utilization rate counted at the preset time interval reflects the real-time status of the service node. During the execution process, if the resource utilization rate counted is not less than the first preset utilization rate threshold, the defense strategy is executed, and the defense strategy indicates that the service node no longer receives requests and only continues to process the requests being processed. Therefore, the first preset utilization rate threshold is usually set to any value between 95% and 100%, indicating that the service node has no extra resources to process the requests that have not been processed.

[0059] Exemplarily, in the embodiment of the present application, the first preset utilization threshold is set to 100%.

[0060] Therefore, by executing a defense strategy, we can prevent unprocessed requests from putting pressure on the service nodes and prevent the service nodes from crashing.

[0061] The server executes a defense strategy, including: returning a preset prompt message to the upstream caller so that the upstream caller can re-initiate a request later based on the preset prompt message; and if there are pending requests that have not been processed in the memory of the service node, deleting the pending requests.

[0062] Step S103, the server generates a request receiving strategy for the service node based on the changes in resource utilization counted during the execution of the defense strategy; the request receiving strategy represents the number of requests received by the service node in a request receiving cycle.

[0063] Step S104: the server controls the received request according to the request receiving policy.

[0064] During the execution process, as no new requests come in, over time, the occupied resources of the service node will be released one by one. At this time, the service node has the ability to process new requests. In order to maximize the utilization of the service node resources, a request receiving strategy for the service node is generated based on the changes in resource utilization counted during the execution of the defense strategy, so that the server can control the received requests according to the request receiving strategy.

[0065] By executing the request receiving strategy to control the number of requests received, the resources of the service node can be fully utilized to avoid crashes and improve the stability of the service node, thereby improving the stability of the service node in processing requests.

[0066] In some embodiments, the server generates a request receiving strategy for the service node based on the changes in resource utilization counted during the execution of the defense strategy, including: if the latest counted resource utilization is less than a second preset utilization threshold, continue to receive requests; use the time when the request is started as the start time; in the process of receiving requests, if the latest counted resource utilization is not less than a third preset utilization threshold, stop receiving requests; use the time when the request is stopped as the end time; the second preset utilization threshold is less than the third preset utilization threshold; the third preset utilization threshold is less than or equal to the first preset utilization threshold; the start time to the end time constitute a first request receiving cycle; the request receiving strategy for the service node is generated based on the time length of the first request receiving cycle and the resource utilization during the request processing process.

[0067] In the process of executing the defense strategy, since there are no new requests coming in, the occupied resources of the service node will be released in turn over time. In order to make full use of the resources of the service node, it is necessary to continue to receive requests after the resource utilization rate of the service node is less than a certain threshold. The time when the request is started to be received is recorded and the time when the request is started to be received is used as the start time. .

[0068] Exemplarily, if the latest statistical resource utilization is less than the second preset utilization threshold, the request continues to be received.

[0069] The second preset utilization rate threshold is a preset value, and its specific value can be set according to actual conditions. The second preset utilization rate threshold is less than the first preset utilization rate threshold. For example, the second preset utilization rate threshold can be set to 50%, or 30%, etc.

[0070] Since the service node is also processing requests while receiving them, the resource utilization of the service node will be greater than a certain threshold at a certain point in time after receiving the request. To maintain the stability of the service node in processing requests, the service node stops receiving requests after the resource utilization exceeds a certain threshold. The time when the request is stopped is recorded and the time when the request is stopped is used as the end time. .

[0071] Exemplarily, during the process of receiving a request, if the latest statistical resource utilization is not less than a third preset utilization threshold, the request is stopped from being received.

[0072] Therefore, the third preset utilization threshold is greater than the second preset utilization threshold.

[0073] Since the receiving of requests is stopped when the latest statistical resource utilization is not less than the third preset utilization threshold, and the receiving of requests is also stopped under the condition that the resource utilization is not less than the first preset utilization threshold, the third preset utilization threshold may be equal to the first preset utilization threshold.

[0074] In order to shorten the time that the resource utilization of the service node remains at 100% and reduce the pressure on the service node to a certain extent, the third preset utilization threshold can also be set to be smaller than the first preset utilization threshold. For example, if the first preset utilization threshold is set to 100%, the third preset utilization threshold can be set to 90% or 93%, etc.

[0075] The specific value of the third preset utilization threshold can be set according to actual conditions, as long as it satisfies the relationship between the first preset utilization threshold and the second preset utilization threshold.

[0076] Since the decision of when to continue receiving requests and when to stop receiving requests is based on the changes in resource utilization statistics during the execution of the defense strategy, for ease of understanding, the start time is To end time As the first request receiving cycle. And the time length of subsequent request receiving cycles is defined as .

[0077] Finally, the request receiving strategy of the service node is generated based on the time length of the first request receiving cycle and the resource utilization rate during the request processing.

[0078] The embodiment of the present application determines when to continue receiving requests and when to stop receiving requests based on the changes in resource utilization counted during the execution of the defense strategy, so that the service node will not crash and its resources will not be idle, thereby improving the stability of the service node in processing requests on the basis of maximizing the use of service node resources.

[0079] In some embodiments, the server generates a request receiving strategy for the service node based on the time length of the first request receiving cycle and the resource utilization during the request processing, including: if the latest statistical resource utilization is not less than the first preset utilization threshold before being less than the second preset utilization threshold, the server obtains the first time when the resource utilization first rises to the first preset utilization threshold, and the second time when the resource utilization last drops below the first preset utilization threshold before being less than the second preset utilization threshold; if the time difference between the first time and the second time is greater than or equal to the time length of the first request receiving cycle, the server generates a request receiving strategy for the service node based on the number of requests received in the first request receiving cycle.

[0080] In the specific implementation process, the resource utilization of the service node changes dynamically during the process of processing requests. For example, the logic of request C is simple, and the request C can be processed quickly. At this time, the resources occupied by the request C can be released quickly, thereby reducing the resource utilization of the service node. Then the request D is processed. The logic of the request D is complex and needs to access the database. The processing time is long, which has occupied the service node resources all the time, so that the service node resources cannot be released and have been maintained at a high level.

[0081] Therefore, before the latest statistical resource utilization is less than the second preset utilization threshold, there may be multiple instances where the resource utilization is not less than the first preset utilization threshold, and multiple instances where the resource utilization is less than the first preset utilization threshold.

[0082] In order to know the time period when the resource utilization of the service node is not less than the first preset utilization threshold, the server obtains the first time when the resource utilization first rises to the first preset utilization threshold, and the second time when the resource utilization last drops below the first preset utilization threshold before it falls below the second preset utilization threshold. The first time is recorded as , and the second time is recorded as .

[0083] Whether the service node is in a high-load state for a long time is determined by judging the time difference between the first time and the second time and the time length of the first request receiving cycle.

[0084] If the time difference between the first time and the second time is greater than or equal to the time length of the first request receiving cycle, that is , the server generates a request receiving strategy for the service node based on the number of requests received in the first request receiving cycle.

[0085] The server generates a request receiving strategy for the service node based on the number of requests received in the first request receiving cycle, including:

[0086] The number of requests received in the first request receiving cycle is multiplied by the preset ratio to obtain the number of requests received by the service node in the current request receiving cycle. The preset ratio is one-third, two-thirds, etc. For example, if the number of requests received in the first request receiving cycle is , then the number of requests received in the current request receiving cycle is .

[0087] or,

[0088] The time length of the first request receiving cycle is divided into multiple intervals according to a preset interval size; the median time point of each interval is determined; based on the number of requests received corresponding to the median time point in the first request cycle and the preset interval size, the number of requests received by the service node at each time point in the interval corresponding to the current request receiving cycle is determined.

[0089] For example, assuming Seconds, according to one small interval every three seconds, it can be divided into 6 small intervals, the number of requests received in the second second of each small interval is taken as the number of requests received in the small interval, and the number of requests is divided by the interval size to determine the number of requests received by the service node at each time point in the interval corresponding to the current request receiving cycle.

[0090] For example, the first small interval receives 300 requests in the first second, 600 requests in the second second, and 900 requests in the third second. The number of requests received in the second second is divided by 3, that is, 600 / 3=200. 200 requests are received at each time point in the corresponding interval of the current request receiving cycle.

[0091] In an embodiment of the present application, the time difference between the first time and the second time is greater than or equal to the time length of the first request receiving cycle, indicating that the service node is in a high-load state for a long time. At this time, by reducing the number of requests received in the current request receiving cycle, the load pressure of the service node is reduced and the stability of the service node is improved, thereby improving the stability of the service node in processing requests.

[0092] In some embodiments, the method also includes: if the time difference between the first time and the second time is less than the time length of the first request receiving cycle, the server obtains the third time when the resource utilization last dropped below a fourth preset utilization threshold before it was less than the second preset utilization threshold; the fourth preset utilization threshold is greater than the second preset utilization threshold and less than the third preset utilization threshold; the server generates a request receiving strategy for the service node based on the first time, the second time, the third time and the number of requests received in the first request receiving cycle.

[0093] In the specific implementation process, considering that when the service node processes a request, the resource utilization rate only reaches a peak within a certain time range, and the time length of this time range is short, therefore, the time difference between the first time and the second time may also be less than the time length of the first request receiving cycle, that is, , at this time, you only need to reduce the number of requests within this time range to reduce the impact on the service node.

[0094] At this time, the server obtains the third time when the resource utilization rate last dropped below the fourth preset utilization rate threshold before being less than the second preset utilization rate threshold.

[0095] The fourth preset utilization threshold is greater than the second preset utilization threshold, and the fourth preset utilization threshold is less than the third preset utilization threshold. For example, the fourth preset utilization threshold can be set to 70%, or 75%, etc. The third time is recorded as .

[0096] Server based on first time , Second Time , Third Time The request receiving strategy of the service node is generated based on the number of requests received in the first request receiving cycle.

[0097] Optionally, the server generates a request receiving strategy for the service node based on the first time, the second time, the third time and the number of requests received in the first request receiving cycle, including: taking the central time point of the start time and the third time as the first midpoint ; The central time point between the first time and the second time is taken as the second midpoint ; If the absolute difference between the second midpoint and the first midpoint is less than the preset threshold, it indicates that the peak center point and the center point in the request processing cycle corresponding to the first request receiving cycle can be regarded as the same point.

[0098] The request processing cycle refers to the time from the start time of the previous request receiving cycle to the start time of the next request receiving cycle. Since the time consumed by the service node to process different requests is different, the time length of different request processing cycles may also be different.

[0099] Taking the center time point of the start time and the third time as the first midpoint, it reflects the average time from the start of receiving the request to the last time the resource utilization drops to the fourth preset utilization threshold, and reflects the time required for the system to recover to a relatively stable state when the load increases, thereby evaluating the system's recovery capability and the rapidity of adjusting resource allocation.

[0100] The center time point of the first time and the second time is taken as the second midpoint, which reflects the average time point when the resource utilization first rises to the first preset utilization threshold and the last time it drops to the first preset utilization threshold, reflecting the duration of the resource under high load state and the performance of the system under extreme load conditions.

[0101] At this time, the control point is determined based on the second midpoint, the start time, the end time and the third time; with the control point as the center, the number of first requests received by the control point within the preset time length in the first request receiving cycle is determined; the number of first requests is multiplied by the preset ratio to obtain the number of requests received within the preset time length in the current request receiving cycle with the control point as the center.

[0102] The preset threshold is a preset value, which can be set to 0.5, or 0.6, etc. The value of the preset threshold can be set according to actual conditions.

[0103] The preset time length is a pre-set value, which can be a time length of six seconds with the control point as the center, three seconds before and three seconds after the control point, or a time length of seven seconds with the control point as the center, four seconds before and three seconds after the control point. The preset time length can be set according to actual conditions.

[0104] The preset ratio is a preset value, which can be set to two-thirds, three-quarters, etc.

[0105] For example, when When , it indicates that the difference between the peak center point and the center point in the request processing cycle corresponding to the first request receiving cycle is no more than 0.5S. The peak center point and the center point can be equated to one point. At this time, the control point of the current request receiving cycle is calculated by the following formula .

[0106]

[0107] Right now

[0108]

[0109] After the control point is calculated, two-thirds of the number of requests for a total of six seconds before and after the control point in the first request receiving cycle is used as the number of requests for a total of six seconds before and after the control point in the current request receiving cycle.

[0110] It should be noted that, since the request receiving policy of the current request receiving cycle of the service node is generated based on the time length of the first request receiving cycle and the resource utilization rate during request processing, the current request receiving cycle at this time is regarded as the second request receiving cycle.

[0111] In an embodiment of the present application, when the time difference between the first time and the second time is less than the time length of the first request receiving cycle, the control point is calculated, and within the current request receiving cycle, only the number of requests received in the time area near the control point is reduced, so as to improve the stability of the service node in processing requests.

[0112] In some embodiments, the method also includes: if the latest statistical resource utilization is not greater than or equal to the first preset utilization threshold before being less than the second preset utilization threshold, then multiplying the number of requests received in the first request receiving cycle by a preset multiple to obtain the number of requests received in the current request receiving cycle.

[0113] During the specific implementation process, if the latest statistical resource utilization is not greater than or equal to the first preset utilization threshold before being less than the second preset utilization threshold, it indicates that the resource utilization of the service node has not been fully occupied during the request processing process, indicating that the resources of the service node are not fully utilized.

[0114] In order to maximize the utilization of the service node's resources, the number of requests received in the first request receiving cycle is multiplied by a preset multiple to obtain the number of requests received in the current request receiving cycle, that is, the number of requests processed by the service node is increased.

[0115] The preset multiple is a pre-set value, which can be set according to actual conditions.

[0116] The preset multiple includes a first preset multiple and a second preset multiple. The first preset multiple is smaller than the second preset multiple.

[0117] If the latest resource utilization statistics are not greater than or equal to the first preset utilization threshold before being less than the second preset utilization threshold, but is greater than the fourth preset utilization threshold, then the number of requests received in the first request receiving cycle is multiplied by the first preset multiple to obtain the number of requests received in the current request receiving cycle.

[0118] For example, the total number of requests received in the first request receiving cycle is magnified by 1.2 times as the number of requests received in the current request receiving cycle, such as: A total of 1000 requests are received, and 1200 requests are received in the current request receiving cycle. Alternatively, the number of requests received per second in the first request receiving cycle is multiplied by 1.2 as the number of requests received per second in the current request receiving cycle.

[0119] If the latest resource utilization statistics are not greater than or equal to the first preset utilization threshold before being less than the second preset utilization threshold, and are not greater than the fourth preset utilization threshold, then the number of requests received in the first request receiving cycle is multiplied by the second preset multiple to obtain the number of requests received in the current request receiving cycle.

[0120] For example, the total number of requests received in the first request receiving cycle is magnified by 1.5 times as the number of requests received in the current request receiving cycle, such as: A total of 1000 requests are received, and 1500 requests are received in the current request receiving cycle. Alternatively, the number of requests received per second in the first request receiving cycle is multiplied by 1.5 as the number of requests received per second in the current request receiving cycle.

[0121] In an embodiment of the present application, after receiving a request in the first request receiving cycle, if the resource utilization of the service node is never greater than or equal to the first preset utilization threshold during the process of processing the request, it indicates that the resource utilization of the service node has always been idle. Therefore, the number of received requests can be increased in the current request receiving cycle to maximize the utilization of the service node's resources and improve the stability of the service node in processing requests.

[0122] In some embodiments, the server controls the received requests according to the request receiving policy, including: if the number of requests issued by the upstream caller is greater than the number of requests in the request receiving policy, then generating a request receiving policy for the next request receiving cycle based on the resource utilization during the request processing, until the latest statistical resource utilization is less than the fifth preset utilization threshold, and the number of requests issued by the upstream caller in the next request receiving cycle is less than or equal to the number of requests in the request receiving policy for the corresponding request receiving cycle, and receiving all requests issued by the upstream caller.

[0123] In a specific implementation, the fifth preset utilization rate threshold may be the same as the second preset utilization rate threshold, or may be less than the second preset utilization rate threshold. The specific value of the fifth preset utilization rate threshold may be set according to actual conditions. For example, the fifth preset utilization rate threshold may be set to 25%, or 30%, etc.

[0124] When the service node is controlling the received requests according to the request receiving strategy, if the number of requests sent by the upstream caller is greater than the number of requests in the request receiving strategy, it indicates that the current resource utilization of the service node cannot normally process all the requests sent by the upstream caller. At this time, it is necessary to control the number of requests sent by the upstream caller, that is, only receive the number of requests that the service node can process, and refuse to receive any excess requests.

[0125] In order to always maximize the utilization of the resources of the service node, if the number of requests sent by the upstream caller is greater than the number of requests in the request receiving policy, the request receiving policy for the next request receiving cycle is generated based on the resource utilization during the request processing until the latest statistical resource utilization is less than the fifth preset utilization threshold, and the number of requests sent by the upstream caller in the next request receiving cycle is less than or equal to the number of requests in the request receiving policy for the corresponding request receiving cycle, then the request receiving policy will no longer be generated, all requests sent by the upstream caller will be received, and the received requests will be processed.

[0126] Among them, the request receiving strategy for generating the next request receiving cycle based on the resource utilization rate during the request processing process can be found in the above embodiment and will not be repeated here.

[0127] However, the above process generates a request receiving policy corresponding to the second request receiving cycle based on the request receiving situation of the first request receiving cycle and the resource utilization of the service node for processing requests. Before the latest statistical resource utilization is less than the fifth preset utilization threshold, and the number of requests sent by the upstream caller in the next request receiving cycle is less than or equal to the number of requests in the request receiving policy of the corresponding request receiving cycle, multiple request receiving cycles may be required, and each request receiving cycle corresponds to a request receiving policy. In the process of sequentially generating request receiving policies, the following process is also included:

[0128] If the absolute difference between the second midpoint and the first midpoint is less than the preset threshold, it indicates that the peak center point and the center point in the request processing cycle corresponding to the current request receiving cycle have not shifted. At this time, the control point of the current request receiving cycle is used as the control point of the next request receiving cycle.

[0129] It should be noted that, if during the processing of the request received in the current request receiving period, if the latest statistical resource utilization is not greater than the first preset utilization threshold before being less than the second preset utilization threshold, but is greater than the fourth preset utilization threshold, the server obtains the fourth time when the resource utilization first drops to the fourth preset utilization threshold , and take the center time point of the fourth time and the third time as the second midpoint.

[0130] If the absolute difference between the second midpoint and the first midpoint is not less than the preset threshold, it indicates that the peak center point and the center point in the request processing cycle corresponding to the current request receiving cycle are offset. At this time, it is necessary to calculate the control point moving distance, and obtain the control point of the next request receiving cycle through the control point moving distance calculation.

[0131] For example, when , that is, the peak center point in the current request processing cycle is on the left / right side of the center point, and the peak center point shifts to the left / right.

[0132] When the peak center point in the request processing cycle corresponding to the current request receiving cycle shifts to the left / right, the control point of the request receiving strategy of the next request receiving cycle should be moved to the right / left.

[0133] The correlation coefficient a is calculated by the following formula.

[0134]

[0135] Right now

[0136]

[0137] in, represents the correlation coefficient, Indicates the second midpoint of the current request receiving cycle, Indicates the second midpoint of the last request receiving cycle, Indicates the displacement length between the control point of the current request reception cycle and the control point of the previous request reception cycle.

[0138] Positive, indicating the control point of the current request reception cycle Move rightward based on the control point of the previous request reception cycle. Negative, indicating the control point of the current request receiving cycle Move left based on the control point of the previous request reception cycle.

[0139] After obtaining the correlation coefficient a, calculate the distance the control point needs to move next time , the calculation formula is:

[0140]

[0141] Right now

[0142]

[0143] That is, the control point of the current request receiving cycle move (Right is positive, left is negative) Get the control point of the next request reception cycle , the calculation formula is as follows:

[0144]

[0145] After the control point is calculated, two-thirds of the number of requests for a total of six seconds before and after the control point in the current request receiving cycle is used as the number of requests for a total of six seconds before and after the control point in the next request receiving cycle.

[0146] In an embodiment of the present application, when the number of requests sent by the upstream caller is greater than the number of requests in the request receiving strategy, the request receiving strategy is iteratively generated until the service node resource utilization is less than the fifth preset utilization threshold, and the number of requests sent by the upstream caller is less than the number of requests in the request receiving strategy, so as to maximize the utilization of the service node resources and maintain the stability of the service node, thereby improving the stability of the service node in processing requests.

[0147] Figure 2 A schematic diagram of a request control device provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the device includes: a statistical module 201, an execution module 202, a generation module 203 and a control module 204; wherein,

[0148] The statistical module 201 is used to count the resource utilization rate at preset time intervals; the execution module 202 is used to execute the defense strategy if the resource utilization rate is not less than the first preset utilization rate threshold; the defense strategy indicates that the service node no longer receives requests and only continues to process the requests being processed; the generation module 203 is used to generate the request receiving strategy of the service node based on the changes in the resource utilization rate counted during the execution of the defense strategy; the request receiving strategy indicates the number of requests received by the service node within a request receiving cycle; the control module 204 is used to control the received requests according to the request receiving strategy.

[0149] Based on the above embodiment, the generation module 203 is specifically used for: if the latest statistical resource utilization is less than the second preset utilization threshold, continue to receive requests; the time when the request is started to be continued is used as the start time; in the process of receiving requests, if the latest statistical resource utilization is not less than the third preset utilization threshold, stop receiving requests; the time when the request is stopped is used as the end time; the second preset utilization threshold is less than the third preset utilization threshold; the third preset utilization threshold is less than or equal to the first preset utilization threshold; the start time to the end time constitute a first request receiving cycle; based on the time length of the first request receiving cycle and the resource utilization in the process of processing the request, a request receiving strategy for the service node is generated.

[0150] Based on the above embodiment, the generation module 203 is specifically used for: if the latest statistical resource utilization is not less than the first preset utilization threshold before being less than the second preset utilization threshold, then obtaining the first time when the resource utilization first rises to the first preset utilization threshold, and the second time when the resource utilization last drops below the first preset utilization threshold before being less than the second preset utilization threshold; if the time difference between the first time and the second time is greater than or equal to the time length of the first request receiving cycle, then generating a request receiving strategy for the service node based on the number of requests received in the first request receiving cycle.

[0151] Based on the above embodiment, the generation module 203 is specifically used to: divide the time length of the first request receiving cycle into multiple intervals according to a preset interval size; determine the median time point of each interval; based on the number of requests received corresponding to the median time point in the first request cycle and the preset interval size, determine the number of requests received by the service node at each time point in the interval corresponding to the current request receiving cycle.

[0152] Based on the above embodiment, the generation module 203 is specifically used for: if the time difference between the first time and the second time is less than the time length of the first request receiving cycle, then obtaining the third time when the resource utilization rate last dropped below the fourth preset utilization threshold before it was less than the second preset utilization threshold; the fourth preset utilization threshold is greater than the second preset utilization threshold and less than the third preset utilization threshold; and generating a request receiving strategy for the service node based on the first time, the second time, the third time and the number of requests received in the first request receiving cycle.

[0153] Based on the above embodiment, the generation module 203 is specifically used to: take the central time point of the start time and the third time as the first midpoint; the central time point of the first time and the second time as the second midpoint; if the absolute difference between the second midpoint and the first midpoint is less than a preset threshold; then determine the control point based on the second midpoint, the start time, the end time and the third time; with the control point as the center, determine the number of first requests received by the control point within the preset time length in the first request receiving cycle; multiply the first request number by the preset ratio to obtain the number of requests received within the preset time length in the current request receiving cycle with the control point as the center.

[0154] Based on the above embodiment, the generation module 203 is specifically used for: if the latest statistical resource utilization is not greater than or equal to the first preset utilization threshold before being less than the second preset utilization threshold, then the number of requests received in the first request receiving cycle is multiplied by the preset multiple to obtain the number of requests received in the current request receiving cycle.

[0155] Based on the above embodiment, the execution module 202 is specifically used to: return a preset prompt message to the upstream caller so that the upstream caller can re-initiate a request later based on the preset prompt message; and if there are pending requests that have not been processed in the memory of the service node, delete the pending requests.

[0156] Based on the above embodiment, the control module 204 is specifically used for: if the number of requests issued by the upstream caller is greater than the number of requests in the request receiving policy, then generating a request receiving policy for the next request receiving cycle based on the resource utilization during the request processing, until the latest statistical resource utilization is less than the fifth preset utilization threshold, and the number of requests issued by the upstream caller in the next request receiving cycle is less than or equal to the number of requests in the request receiving policy for the corresponding request receiving cycle, and receiving all requests issued by the upstream caller.

[0157] It should be understood that the device corresponds to the above-mentioned request control method embodiment and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be found in the above description. To avoid repetition, the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or fixed in the operating system (OS) of the device.

[0158] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 3 As shown, the electronic device includes a processor 301 (processor), a memory 302 (memory) and a bus 303; wherein the processor 301 and the memory 302 communicate with each other through the bus 303. The processor 301 is used to call the program instructions in the memory 302 to execute the methods provided by the above-mentioned method embodiments.

[0159] The processor 301 may be an integrated circuit chip with signal processing capability. The processor 301 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. It may implement or execute various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0160] The memory 302 can be implemented by any type of volatile or non-volatile storage 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 red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0161] This embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments.

[0162] This embodiment provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable the computer to execute the methods provided by the above-mentioned method embodiments.

[0163] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0164] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0165] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0166] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A request control method, characterized in that: The method is applied to a service node, and the method comprises: Statistics on resource utilization at preset time intervals; If the resource utilization is not less than a first preset utilization threshold, a defense strategy is executed; the defense strategy indicates that the service node no longer receives requests and only continues to process the requests being processed; Generate a request receiving strategy for the service node based on the changes in resource utilization rates counted during the execution of the defense strategy; the request receiving strategy represents the number of requests received by the service node in a request receiving cycle; Controlling received requests according to the request receiving strategy; The generating of the request receiving strategy of the service node based on the change of resource utilization rate counted during the execution of the defense strategy includes: If the latest resource utilization rate is less than the second preset utilization rate threshold, continue to receive requests; and use the time when the request is started to be received as the start time; In the process of receiving requests, if the latest statistical resource utilization is not less than the third preset utilization threshold, stop receiving requests; the time of stopping receiving requests is used as the end time; the second preset utilization threshold is less than the third preset utilization threshold; the third preset utilization threshold is less than or equal to the first preset utilization threshold; the start time to the end time constitute a first request receiving cycle; Generate a request receiving strategy for the service node based on the time length of the first request receiving cycle and the resource utilization rate in the process of processing the request; The generating of the request receiving strategy of the service node based on the time length of the first request receiving cycle and the resource utilization rate in the process of processing the request includes: If the latest statistical resource utilization is not less than the first preset utilization threshold before being less than the second preset utilization threshold, then obtaining a first time when the resource utilization first rises to the first preset utilization threshold, and a second time when the resource utilization last drops below the first preset utilization threshold before being less than the second preset utilization threshold; If the time difference between the first time and the second time is greater than or equal to the time length of the first request receiving cycle, generating the request receiving strategy of the service node based on the number of requests received in the first request receiving cycle multiplied by a first preset ratio; the value range of the first preset ratio is (0,1); If the latest resource utilization counted is not greater than or equal to the first preset utilization threshold before being less than the second preset utilization threshold, the number of requests received in the first request receiving cycle is multiplied by the preset multiple to obtain the number of requests received in the current request receiving cycle; the preset multiple is greater than 1.

2. The method according to claim 1, characterized in that The step of generating the request receiving strategy of the service node based on the number of requests received in the first request receiving cycle multiplied by a first preset ratio includes: Dividing the time length of the first request receiving cycle into multiple intervals according to a preset interval size; determining the median time point for each of said intervals; Based on the number of requests received corresponding to the median time point in the first request cycle and the preset interval size, the number of requests received by the service node at each time point in the interval corresponding to the current request receiving cycle is determined.

3. The method according to claim 1, characterized in that The method further comprises: If the time difference between the first time and the second time is less than the time length of the first request receiving cycle, obtaining a third time when the resource utilization rate last dropped below a fourth preset utilization threshold before it was less than the second preset utilization threshold; the fourth preset utilization threshold is greater than the second preset utilization threshold and less than the third preset utilization threshold; A request receiving strategy of the service node is generated based on the first time, the second time, the third time, and the number of requests received in the first request receiving period.

4. The method according to claim 3, characterized in that The generating the request receiving strategy of the service node based on the first time, the second time, the third time and the number of requests received in the first request receiving cycle includes: The center time point between the start time and the third time is taken as the first midpoint; the center time point between the first time and the second time is taken as the second midpoint; If the absolute difference between the second midpoint and the first midpoint is less than a preset threshold, determining a control point based on the second midpoint, the start time, the end time and the third time; Taking the control point as the center, determining the number of first requests received by the control point within a preset time length in the first request receiving cycle; The first number of requests is multiplied by a second preset ratio to obtain the number of requests received within a preset time length in the current request receiving cycle with the control point as the center; the value range of the second preset ratio is (0,1).

5. The method according to claim 1, characterized in that The execution defense strategy includes: Returning a preset prompt message to the upstream caller, so that the upstream caller can re-initiate a request later based on the preset prompt message; And, if there are pending requests that have not been processed in the memory of the service node, the pending requests are deleted.

6. The method according to any one of claims 1 to 5, characterized in that: The controlling the received request according to the request receiving strategy includes: If the number of requests sent by the upstream caller is greater than the number of requests in the request receiving policy, the request receiving policy for the next request receiving cycle is generated based on the resource utilization during the request processing, until the latest statistical resource utilization is less than the fifth preset utilization threshold, and the number of requests sent by the upstream caller in the next request receiving cycle is less than or equal to the number of requests in the request receiving policy for the corresponding request receiving cycle, and all requests sent by the upstream caller are received.

7. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and when the machine-readable instructions are executed by the processor, the method according to any one of claims 1 to 6 is performed.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is executed.

9. A computer program product, characterized in that include: A computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is executed.

Images