Container resource adjustment method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of cloud computing, and particularly discloses a container resource adjustment method and device, an electronic device and a storage medium. The method comprises the following steps: in response to abnormal load dimension data monitored, determining corresponding abnormal load dimension and configuration information of an abnormal container; determining a target service to which the abnormal container belongs based on the configuration information of the abnormal container; obtaining a load reference value of the target service under the abnormal load dimension; and adjusting the container resource according to the load reference value and a preset elastic scaling strategy. The application reduces the difficulty of judging the container resource adjustment based on the recording rule of the load dimension data, improves the accuracy and efficiency of the container resource adjustment, and realizes service elastic scaling. The recording rule of the load dimension data provides conditions for introducing more load dimensions, and based on the finer granularity of the load dimension, the flexibility of the container resource management and the efficiency of the container load data analysis can be improved.

Description

Technical Field

[0001] This application relates to the field of cloud computing technology, and in particular to a container resource adjustment method, apparatus, electronic device, and storage medium. Background Technology

[0002] Containers, as a lightweight, portable, and self-contained software packaging technology, enable applications to run in the same way almost anywhere. In cloud technology, service elastic scaling is a key feature provided by container cloud platforms. These platforms analyze service (business) performance based on container load data and automatically increase or decrease service resources to ensure that services have sufficient resources to handle application loads and guarantee their normal, healthy operation. Elastic scaling effectively controls costs and manages resources, preventing both sudden business surges leading to insufficient service resources and unresponsive situations, and low service request volumes resulting in wasted resources. However, in related technologies, adjusting container resources for business needs often requires manual intervention and deployment, which cannot guarantee the efficiency of container resource adjustments. Summary of the Invention

[0003] To address the low efficiency issues in existing technologies for adjusting container resources, this application provides a container resource adjustment method, apparatus, electronic device, and storage medium: According to a first aspect of this application, a container resource adjustment method is provided, the method comprising: In response to the detected abnormal load dimension data, the corresponding abnormal load dimension and the configuration information of the abnormal container are determined; among them, a load dimension data is used to record the real-time load information of a container under a preset load dimension. The target service to which the abnormal container belongs is determined based on the configuration information of the abnormal container; Obtain the load reference value indicating the target service under the abnormal load dimension; Adjust container resources based on the load reference value and the preset elastic scaling strategy.

[0004] According to a second aspect of this application, a container resource adjustment apparatus is provided, the apparatus comprising: Response module: Used to respond to the detected abnormal load dimension data, determine the corresponding abnormal load dimension and the configuration information of the abnormal container; among them, a load dimension data is used to record the real-time load information of a container under a preset load dimension. Target service determination module: used to determine the target service to which the abnormal container belongs based on the configuration information of the abnormal container; Load reference value acquisition module: used to acquire the load reference value indicating the target service under the abnormal load dimension; Adjustment module: Used to adjust container resources based on the load reference value and preset elastic scaling strategy.

[0005] According to a third aspect of this application, an electronic device is provided, the electronic device including a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the container resource adjustment method as described in the first aspect.

[0006] According to a fourth aspect of this application, a computer-readable storage medium is provided, wherein at least one instruction or at least one program is stored therein, the at least one instruction or the at least one program being loaded and executed by a processor to implement the container resource adjustment method as described in the first aspect.

[0007] According to a fifth aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the container resource adjustment method as described in the first aspect.

[0008] The container resource adjustment method, apparatus, electronic device, and storage medium provided in this application have the following technical advantages: This application identifies the corresponding abnormal load dimension and the configuration information of the abnormal container by monitoring abnormal load dimension data. Then, based on the configuration information of the abnormal container, it determines the target service to which the abnormal container belongs. Furthermore, it obtains the load reference value indicating the target service under the abnormal load dimension, and thus determines whether to adjust container resources based on a preset elastic scaling strategy and the load reference value indicating the same service under the abnormal load dimension. One load dimension data is used to record the real-time load information of a container under a preset load dimension. Abnormal load dimension data can accurately locate the abnormal load dimension and the configuration information of the abnormal container. Simultaneously, the load reference value indicating the target service under the abnormal load dimension is taken into consideration when adjusting container resources. The recording rules of this application based on load dimension data reduce the difficulty of judging container resource adjustments and improve the accuracy and efficiency of container resource adjustments. The recording rules of load dimension data provide conditions for the introduction of more load dimensions. Based on finer-grained load dimensions, the flexibility of container resource management and the efficiency of container load data analysis can be improved. Attached Figure Description

[0009] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of an application environment provided in an embodiment of this application; Figure 2 This is a flowchart illustrating a container resource adjustment method provided in an embodiment of this application; Figure 3 This is a schematic diagram of a process for creating load dimension data provided in an embodiment of this application; Figure 4 This is a schematic diagram of a process for generating load profile data provided in an embodiment of this application; Figure 5 This is a schematic diagram of a process for obtaining a load reference value indicating a target service under the dimension of abnormal load, provided in an embodiment of this application; Figure 6 This is also a flowchart illustrating a process for obtaining a load reference value indicating the target service under the abnormal load dimension, as provided in the embodiments of this application. Figure 7 This is a schematic diagram of a process for collecting load dimension data provided in an embodiment of this application; Figure 8 This is also a schematic diagram of a process for generating load profile data provided in the embodiments of this application; Figure 9 This is a schematic diagram of a process for elastically expanding services provided in an embodiment of this application; Figure 10 This is a schematic diagram of a process for elastically scaling down services provided in an embodiment of this application; Figure 11 This is a block diagram of a container resource adjustment device provided in an embodiment of this application; Figure 12 This is a schematic diagram of a blockchain system provided in an embodiment of this application; Figure 13 This is a schematic diagram of a block structure provided in an embodiment of this application; Figure 14 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0012] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, products or devices.

[0013] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0014] Service (business) elastic scaling: Automatically calculates the amount of resources required for the service based on the service load and adjusts the service resources accordingly.

[0015] Please see Figure 1 , Figure 1 This is a schematic diagram of an application environment provided in an embodiment of this application. This application environment may include a client 10 and a server 20. The client 10 and the server 20 can be directly or indirectly connected via wired or wireless communication. The client can provide a user interface for staff, who can send business requirement information to the server based on the user interface. The server provides corresponding container resources for the relevant business based on the business requirement information. The server determines the corresponding abnormal load dimension and the configuration information of the abnormal container by listening to abnormal load dimension data; then, based on the configuration information of the abnormal container, it determines the target business to which the abnormal container belongs; furthermore, it obtains the load reference value indicating the target business under the abnormal load dimension, and thus determines whether to adjust container resources based on a preset elastic scaling strategy and the load reference value indicating the same business under the abnormal load dimension. It should be noted that... Figure 1 This is just one example.

[0016] Clients can include physical devices such as smartphones, desktop computers, tablets, laptops, augmented reality (AR) / virtual reality (VR) devices, digital assistants, smart speakers, and smart wearable devices. They can also include software running on these physical devices, such as computer programs. The operating systems that support the client can include Android, iOS (a mobile operating system developed by Apple), Linux, and Microsoft Windows.

[0017] The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The server may include network communication units, processors, and memory, etc.

[0018] Cloud computing refers to the delivery and usage model of IT infrastructure, meaning obtaining necessary resources through the network in an on-demand and easily scalable manner. In a broader sense, cloud computing refers to the delivery and usage model of services, meaning obtaining necessary services through the network in an on-demand and easily scalable manner. These services can be IT and software related, internet-related, or other services. Cloud computing is a product of the development and integration of traditional computer and network technologies such as grid computing, distributed computing, parallel computing, utility computing, network storage technologies, virtualization, and load balancing.

[0019] In practical applications, for medium to large-scale container cloud platforms, the number of hosted services (businesses) can reach tens of thousands or more. Services hosted on container cloud platforms can include, but are not limited to, instant messaging, live streaming, navigation, and gaming services. The container cloud platform (specifically, the server-side component within the platform) can utilize the container resource adjustment scheme provided in this application to achieve elastic scaling of services. The elastic scaling strategy for container cloud platforms can be based on dimensions such as Central Processing Unit (CPU) and memory. By collecting and analyzing load data such as CPU and memory usage of containers, the container cloud platform can determine the service's operational status, calculate the resources required for the service, and automatically initiate resource adjustment operations.

[0020] The following describes a specific embodiment of a container resource adjustment method according to this application. Figure 2 This is a flowchart illustrating a container resource adjustment method provided in an embodiment of this application. This application provides the operational steps of the method described in the embodiment or flowchart, but based on conventional or non-inventive labor, it may include more or fewer operational steps. The order of steps listed in the embodiment is merely one possible execution order among many steps and does not represent the only execution order. In actual system or product execution, the method can be executed sequentially according to the embodiment or the accompanying drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment). Specifically, as shown in the flowchart... Figure 2 As shown, the method may include: S201: In response to the detected abnormal load dimension data, determine the corresponding abnormal load dimension and the configuration information of the abnormal container; wherein, a load dimension data is used to record the real-time load information of a container under a preset load dimension. In this embodiment, the server responds to detected abnormal load dimension data by determining the corresponding abnormal load dimension and the configuration information of the abnormal container. A load dimension data record records the real-time load information of a container under a preset load dimension. The real-time load information may include the real-time load value, relevant time information, and the configuration information of the relevant container. For example, load dimension data A records the real-time load information A of container A under the memory dimension. When the real-time memory load value in real-time load information A does not fall within the normal memory load value range, load dimension data A is abnormal load dimension data. The configuration information of the relevant container in real-time load information A can indicate the service to which container A belongs, that is, which service container A provides resources to.

[0021] In one exemplary implementation, such as Figure 3 As shown, the server can create load dimension data based on the following steps: S301: Perform load monitoring on each container in the container cluster under the preset load dimension; S302: Every preset time interval, create corresponding load dimension data based on the first preset data format and the load monitoring data at the current time; wherein, the first preset data format indicates a field for recording load dimensions, a field for recording creation time, a field for recording load values, and a field for recording container configuration information.

[0022] The first preset data format can be used to define the real-time load of containers. The fields in this format, used to record load metric, creation time (timestamp), load value, and container configuration information, can be considered as four variables. The container configuration information can also be viewed as extended information (extend_info), which may include the container name, the service to which the container belongs, and container quotas (such as the number of specification cards for a single container). This first preset data format can be set by staff through a client application.

[0023] The server can monitor the load of each container in the container cluster under preset load dimensions. These preset load dimensions are not limited to CPU and memory dimensions; they can also include Graphics Processing Unit (GPU) dimensions, disk dimensions, network input / output (IO) dimensions, service request volume dimensions, service response time dimensions, and more. Preset load dimensions can be flexibly adjusted according to business needs, reducing human intervention and promoting automated operation and maintenance. New load dimensions can be introduced, and existing load dimensions can be deleted. For example, container B can be monitored under all the load dimensions mentioned above as needed, while container C can be monitored under GPU and service response time dimensions as required.

[0024] The server can create corresponding load dimension data at preset intervals, based on the first preset data format and the load monitoring data at the current time. Every preset interval (e.g., 1 minute), the server encapsulates the load monitoring data (data monitoring container D under the CPU dimension) at the current time (e.g., 13:15 PM) according to the requirements of the first preset data format. The resulting load dimension data contains fields recording the CPU dimension, the current time 13:15 PM (i.e., the creation time of the load dimension data), the real-time memory load value, and the configuration information of container D. It should be noted that when monitoring the same container under at least two preset load dimensions, the load dimension data created at each preset interval corresponds one-to-one with those at least two preset load dimensions; that is, each preset load dimension has a corresponding load dimension data.

[0025] Setting the first preset data format ensures that the load dimension data records real-time container load information under a preset load dimension. Recording rules based on load dimension data improve the computational efficiency of data extraction and integration during container resource adjustments, thereby enhancing the timeliness of container resource adjustments and providing better container resource support for related businesses. Simultaneously, the first preset data format also supports flexible adjustments to the preset load dimension.

[0026] In practical applications, see Figure 7The server side may include a cluster of resource servers providing computing resources for related businesses. These resource servers contain containers created according to business needs, which can be used to build container clusters. The server side may also include node agent servers, deploying one node agent server for each resource server to manage the containers on the resource server. The node agent server can perform real-time monitoring of the containers on the corresponding resource server based on at least one preset load dimension. The server side may also include a load statistics server, which is responsible for receiving raw real-time container load data. The node agent servers can report one piece of raw load monitoring data to the load statistics server at preset intervals (e.g., 1 minute). The load statistics server is also responsible for persistently storing the raw data, after data format processing, in a time-series database. The load statistics server can encapsulate the raw load monitoring data according to the requirements of the first preset data format mentioned above. The resulting load dimension data includes fields recording the load dimension, the reporting time, the real-time load value, and the container configuration information, and is then persistently stored in the time-series database. Data in time-series databases can serve as a data source for subsequent service elastic expansion and container historical profiling analysis.

[0027] Further, see Figure 9 The load statistics server is also responsible for performing anomaly checks on the received load monitoring data or processed load dimension data according to the configured anomaly detection policy to determine whether the load of the relevant containers is at a normal level. If it is not at a normal level, the relevant load dimension data can be sent to an anomaly message queue. The server may include a control server, which listens to this anomaly message queue to obtain abnormal load dimension data. For example, considering the actual impact of high load on relevant business operations, if the load of relevant containers is high, the load statistics server will send the relevant load dimension data to the anomaly message queue. The control server listens to this anomaly message queue to obtain abnormal load dimension data.

[0028] In one exemplary implementation, such as Figure 4 As shown, the server can generate load profile data based on the following steps: S401: Extract the target load dimension data that falls within the current time window from the locally stored load dimension dataset; S402: Generate corresponding load profile data according to the second preset data format and the target load dimension data; wherein, the second preset data format indicates fields for recording load dimensions, fields for recording the time window to which the load belongs, fields for recording load statistics, and fields for recording container configuration information.

[0029] The second preset data format can be used to define the historical profile of a container. The fields in this format, used to record load metrics, time windows, load statistics, and container configuration information, can be considered as four variables. Specifically, the field recording load statistics can be the field recording the average load value (avg_value), maximum load value (max_value), and peak load value (peak_value). The peak load value can indicate the average of multiple load peaks. The container configuration information can also be considered as extended information (extend_info), which can include the container name, the service to which the container belongs, and container quotas (such as the specification card quantity for a single container). The second preset data format can be set by staff through the client. It should be noted that the load metric settings in the first preset data format and the second preset data format must be consistent.

[0030] Load dimension data represents the real-time load of a container at a specific point in time. Locally stored load dimension datasets can serve as a data source for historical container profiling analysis. The server can extract load dimension data from the locally stored dataset that falls within a time window, indicating the same container, to generate corresponding load profile data according to the requirements of a second preset data format. For example, target load dimension data falling within the current time window (00:00-24:00 on [Date]) can be extracted from the locally stored dataset. These target load dimension data indicate the same container. These target load dimension data may indicate different load dimensions (such as CPU dimension, disk dimension), and can be further analyzed according to specific load dimensions to obtain load profile data corresponding to each specific load dimension (such as load profile data corresponding to CPU dimension, load profile data corresponding to disk dimension). During analysis, the fields used to record load statistics, as indicated by the second preset data format, can be combined to calculate the average load, maximum load, and peak load average of containers under that specific dimension within the current time window (one day). The peak load average can be the average of the 10 highest real-time load values ​​of the day.

[0031] The server can determine when to extract load dimension data from the locally stored load dimension dataset based on the time range indicated by the time window. For example, if the time window indicates a day (00:00-24:00), the extraction time can be selected at midnight (00:00) each day, extracting load dimension data from the previous day (00:00-24:00). If the time window indicates an hourly time, the extraction time can be selected at every hour (e.g., 13:00), extracting load dimension data from the previous hour (12:00-13:00).

[0032] The second preset data format ensures that the load profile data records historical container profile information under a preset load dimension. Derived from the first preset data format, the second preset data format adjusts the creation time setting to the corresponding time window and the (real-time) load value setting to load statistics, improving the computational convenience and efficiency of generating load profile data based on load dimension data. The generated load profile data can serve as a reference, improving the effectiveness of subsequent container resource adjustments based on preset elastic scaling strategies.

[0033] In practical applications, see Figure 8 Based on the relevant descriptions in steps S301-S303 above, the server may also include a profile generation server. This server is responsible for analyzing the container's load dimension data and generating load profile data for the container, providing a reference for service elastic scaling. The profile generation server can periodically extract all load dimension data for the relevant container from the aforementioned time-series database each day; this load dimension data can form a sequence. The profile generation server can analyze each specific load dimension to obtain load profile data corresponding to each specific load dimension, and persistently store the load profile data in the database.

[0034] S202: Determine the target service to which the abnormal container belongs based on the configuration information of the abnormal container; In this embodiment, the server determines the target service to which the abnormal container belongs based on its configuration information. That is, the abnormal container provides resources to the target service, and the target service is the recipient of the resources provided by the abnormal container. Once the target service is determined, it can be used as a benchmark to obtain a load reference value, thereby determining whether container resource adjustments are needed. The target service can be an instant messaging service, a live streaming service, a navigation service, a gaming service, etc. For example, the target service can be an instant messaging product, a live streaming product, a navigation product, or a gaming product. The target service can also be a functional module within a related product, such as the voice call function within an instant messaging product.

[0035] S203: Obtain the load reference value indicating the target service under the abnormal load dimension; In this embodiment, the server obtains a load reference value indicating the target service under the abnormal load dimension. There may be more than one abnormal container providing resources for the target service, and triggering container resource adjustments based on the real-time load information of a single abnormal container can be inaccurate. Triggering container resource adjustments using a load reference value defined by both the abnormal load dimension and the target service parameters ensures the effectiveness of the container resource adjustments.

[0036] In one exemplary implementation, such as Figure 5 As shown, obtaining the load reference value indicating the target service under the abnormal load dimension includes: S501: Determine a reference container that belongs to the same target service as the abnormal container; S502: When the abnormal load value in the abnormal load dimension data is greater than a first preset value, obtain the reference load dimension data of the reference container that indicates the abnormal load dimension; S503: Based on the abnormal load value and the load value in the reference load dimension data, obtain the load reference value indicating the target service under the abnormal load dimension.

[0037] The server can determine reference containers belonging to the same target service as the abnormal container based on the container configuration information for that service. In response to a received container resource request for the target service, the server typically configures a corresponding container for the target service by combining the requirement information carried in the request with available resources. The configured containers may be distributed across different machines, and the number of containers on each machine may vary.

[0038] The abnormal load value in the abnormal load dimension data serves as the real-time load value of the abnormal container. If the abnormal load value exceeds a first preset value, it indicates that the abnormal container's load in the abnormal load dimension is not at a normal level, or that the abnormal container is experiencing a high load in this dimension. The first preset value here can indicate the upper limit of the normal load level for the container under this abnormal load dimension. The server can obtain the reference load dimension data indicating the abnormal load dimension for the reference container, i.e., obtain the real-time load information of the reference container under the abnormal load dimension. For example, if there are reference containers 1-3, then the server can obtain the load dimension data 1 indicating the abnormal load dimension for container 1, the load dimension data 2 indicating the abnormal load dimension for container 2, and the load dimension data 3 indicating the abnormal load dimension for container 3.

[0039] The server can obtain a load reference value indicating the target service under the abnormal load dimension based on the abnormal load value and the load value in the reference load dimension data. It can extract the load values ​​from the reference load dimension data (e.g., the real-time load values ​​of container 1, container 2, and container 3) and use these load values ​​together with the abnormal load value as the load reference value. Alternatively, it can extract the load values ​​from the reference load dimension data, calculate the average of these load values, and use the average value together with the abnormal load value as the load reference value. Or, it can extract the load values ​​from the reference load dimension data, calculate the average of these load values ​​and the abnormal load value, and use the average value as the load reference value.

[0040] Since a load dimension data point records the real-time load information of a container under a preset load dimension, it's possible to accurately locate and obtain reference load dimension data that matches the abnormal load dimension. When the load reference value is defined by two parameters—the abnormal load dimension and the target business—the reference container is determined based on the target business. Then, the reference load dimension data is obtained by comparing the abnormal load value with the first preset data, thus yielding the load reference value used as the basis for capacity expansion assessment. This provides a timely load reference value for subsequent capacity expansion assessments, ensuring elastic service expansion and avoiding service access latency and resource overload.

[0041] In practical applications, based on the relevant descriptions in steps S301-S303 above, the control server can parse the target service indicated by the container configuration information based on the abnormal load dimension data, and then pull the load dimension data of other containers belonging to the same target service under the abnormal load dimension.

[0042] In one exemplary implementation, such as Figure 6 As shown, obtaining the load reference value indicating the target service under the abnormal load dimension includes: S601: Determine a reference container that belongs to the same target service as the abnormal container; S602: When the abnormal load value in the abnormal load dimension data is less than the second preset value, obtain the first load profile data of the abnormal container indicating the abnormal load dimension, and obtain the second load profile data of the reference container indicating the abnormal load dimension. S603: Based on the first load profile data and the second load profile data, obtain the load reference value indicating the target service under the abnormal load dimension.

[0043] The server can determine reference containers belonging to the same target service as the abnormal container based on the container configuration information for that service. In response to a received container resource request for the target service, the server typically configures a corresponding container for the target service by combining the requirement information carried in the request with available resources. The configured containers may be distributed across different machines, and the number of containers on each machine may vary.

[0044] The abnormal load value in the abnormal load dimension data serves as the real-time load value of the abnormal container. If the abnormal load value is less than a second preset value, it indicates that the abnormal container's load in the abnormal load dimension is not at a normal level, or that the abnormal container is experiencing a low load situation in the abnormal load dimension. Of course, the second preset value here can indicate the lower limit of the normal load level for the container's load under this abnormal load dimension. The server can obtain the first load profile data indicating the abnormal load dimension for the abnormal container, and the second load profile data indicating the abnormal load dimension for the reference container, that is, obtain the historical profile information of all containers belonging to the same target business under the abnormal load dimension.

[0045] The server can obtain a load reference value indicating the target business under the abnormal load dimension based on the first load profile data and the second load profile data. For example, it can extract the load statistics from the first load profile data and the second load profile data and use the load statistics as the load reference value.

[0046] Since a load profile records the historical profile information of a container under a preset load dimension, it is possible to accurately locate and obtain the first and second load profiles that match the abnormal load dimension. When the load reference value is limited by two parameters: the abnormal load dimension and the target business, the reference container is determined based on the target business. Then, the first and second load profiles are obtained by comparing the abnormal load value with the second preset data, thus obtaining the load reference value as the basis for scaling down assessment. This provides a load reference value with historical experience for subsequent scaling up assessment, ensuring elastic service scaling, releasing resources for high-load services, and promoting resource balance and efficient resource utilization among services.

[0047] In practical applications, based on the relevant records in steps S301-S303 and S401-S403, the control server can parse the abnormal load dimension data to obtain the target service indicated by the container configuration information, and then pull the load profile data of all containers belonging to the target service (equivalent to obtaining the container list of the target service) under the abnormal load dimension.

[0048] S204: Adjust container resources according to the load reference value and the preset elastic scaling strategy.

[0049] In this embodiment, the preset elastic scaling strategy includes preset expansion conditions and preset reduction conditions. When the load reference value meets the preset expansion conditions, the server initiates an expansion process. When the load reference value meets the preset reduction conditions, the server initiates a reduction process.

[0050] The preset expansion conditions indicate a third preset value for a specific load dimension. When the load reference value exceeds this third preset value, expansion is required. The expansion process is initiated when the load reference value meets the preset expansion conditions. This process involves increasing the provided container resources for the relevant services, ensuring their normal operation. Specifically, increasing the provided container resources can mean introducing new containers (i.e., increasing the number of containers) or increasing the quota of existing containers. For example, the load reference values ​​include real-time load values ​​for the CPU dimension: 1 (corresponding to container 1): 80%, 2 (corresponding to container 2): 75%, and 3 (corresponding to container 3): 90%. These percentages represent the CPU utilization of the corresponding containers. The third preset value for the CPU dimension is 70%. If real-time load values ​​1, 2, and 3 all exceed the third preset value, then the load reference value meets the preset expansion conditions. The differences between real-time load value 1 and the third preset value, the differences between real-time load value 2 and the third preset value, and the differences between real-time load value 3 and the third preset value can be calculated separately. These three differences can then be used to determine the container resources that need to be added. For example, if containers 1-3 have 10 CPU resource cards, the required container resources would be 5 more of these CPU resource cards. The expansion process could be: adding a new container with 5 resource cards; adding 5 more CPU resource cards to the existing CPU resource cards in container 1; adding 2 more CPU resource cards to the existing CPU resource cards in container 2; and adding 3 more CPU resource cards to the existing CPU resource cards in container 3.

[0051] The preset scaling-down condition indicates a fourth preset value for a specific load dimension. When the load reference value falls below this fourth preset value, scaling down is required. The scaling-down process is initiated when the load reference value meets the preset scaling-down condition. This process reduces the container resources provided to related services, thereby reducing their operating costs. Specifically, reducing the provided container resources can involve deleting existing containers (i.e., reducing the number of containers) or reducing the quota of existing containers. For example, the load reference value includes the following indicators for the GPU dimension: Average Load 1 (corresponding to the load profile data of container 1): 30%, Average Load 2 (corresponding to the load profile data of container 2): 35%, and Average Load 3 (corresponding to the load profile data of container 3): 25%. These percentages represent the GPU utilization of the corresponding containers. The fourth preset value for the GPU dimension is 40%. If Average Load 1, Average Load 2, and Average Load 3 are all lower than the fourth preset value, then the load reference value meets the preset scaling-down condition. The differences between the average load value 1 and the fourth preset value, the differences between the average load value 2 and the fourth preset value, and the differences between the average load value 3 and the fourth preset value can be calculated separately. These three differences can then be used to determine the container resources that need to be reduced. For example, if containers 1-3 have 3 GPU resource cards, the required reduction would be to remove 3 of these GPU resource cards. Therefore, the scaling-down process could be to delete any of containers 1-3; or it could be to reduce the GPU resource card configuration of container 2 to 2 and the GPU resource card configuration of container 3 to 1.

[0052] In an exemplary implementation, referring to the relevant descriptions in steps S501-S503 above, the process of adjusting container resources based on the load reference value and the preset elastic scaling strategy is as follows: First, a first preset elastic scaling strategy indicating expansion conditions is obtained, and a third preset value is determined based on the first preset elastic scaling strategy indicating expansion conditions; wherein, the third preset value is the maximum value indicating normal operation of the target service under the abnormal load dimension. Then, expansion judgment is performed based on the load reference value and the preset elastic scaling strategy, and the magnitude of the load reference value and the third preset value is determined. Finally, when the judgment result indicates expansion (i.e., the load reference value is greater than the third preset value), the amount of resources required for expansion is calculated based on the load reference value and the normal load value of the target service (which can be the third preset value here, or other values ​​indicating normal operation of the target service under the abnormal load dimension), and container resources are expanded based on the calculation result.

[0053] See Figure 9When a service experiences a sudden surge in requests, and monitoring metrics such as CPU, memory, and service response time show abnormalities, the server analyzes the service's operational status and elastically scales the service. By analyzing the real-time load of containers, it determines whether the service meets the conditions for high-load scaling. If so, it calculates the resources required for the service to return to normal load, initiates the scaling process, and automatically creates an appropriate number of containers or increases container quotas to achieve elastic service scaling. Otherwise, the process ends. Through this process, the container cloud platform can automatically upgrade the underlying resources of the service in abnormal scenarios such as high CPU load, high memory load, and long service response times.

[0054] In an exemplary embodiment, referring to the relevant descriptions in steps S601-S603 above, the process of adjusting container resources based on the load reference value and the preset elastic scaling strategy is as follows: First, a second preset elastic scaling strategy indicating the scaling-down condition is obtained, and a fourth preset value is determined based on the second preset elastic scaling strategy indicating the scaling-down condition; wherein, the fourth preset value is the minimum value indicating the normal operation of the target service under the abnormal load dimension. Then, a scaling-down judgment is performed based on the load reference value and the preset elastic scaling strategy, and the size of the load reference value and the fourth preset value is determined. Finally, when the judgment result indicates scaling-down (i.e., the load reference value is less than the fourth preset value), the amount of resources required for scaling-down is calculated based on the load reference value and the normal load value of the target service (which can be the fourth preset value here, or other values ​​indicating the normal operation of the target service under the abnormal load dimension), and container resources are shrunk based on the calculation result.

[0055] See Figure 10 When service request volume decreases and container load falls below normal levels, the server, by combining historical load data of the containers, determines whether there is excess resource in the service and promptly reclaims service resources to elastically shrink the service. By analyzing historical profile data of all containers for the service, it determines whether the service meets the low-load scaling-up conditions in this dimension. If so, it calculates the amount of idle resources for the service, initiates the scaling-up process, automatically deletes an appropriate number of containers or reduces container quotas to achieve elastic service scaling; otherwise, the process ends. Through the above process, in service idle scenarios such as low CPU load, low memory load, and low service throughput, the container cloud platform automatically reclaims service resources, reducing service operating costs.

[0056] The container resource adjustment solution provided in this application defines an extensible data structure (format) for load dimension data that records real-time load information and load profile data that records historical profile information. Based on this extensible data structure, the resource consumption of containers is managed and analyzed, and service resource adjustment decisions are automatically generated to achieve service elastic scaling. This extensible data structure makes data management and analysis more flexible, allowing businesses to flexibly and quickly add various indicators such as service throughput and service response time to monitor service operation according to their own service operation and maintenance needs, with low addition costs, thereby achieving more effective dynamic adjustment of service resources. At the same time, it reduces the steps of processing redundant data and improves data analysis efficiency. In practical applications, this solution can not only be used in container cloud platforms to achieve scalable service elastic scaling, but also in services deployed on physical machines to achieve scalable resource adjustment.

[0057] As can be seen from the technical solutions provided in the embodiments of this application above, the embodiments of this application determine the corresponding abnormal load dimension and the configuration information of the abnormal container by listening to abnormal load dimension data; then, based on the configuration information of the abnormal container, the target service to which the abnormal container belongs is determined; furthermore, the load reference value indicating the target service under the abnormal load dimension is obtained, thereby determining whether to adjust container resources based on the preset elastic scaling strategy and the load reference value indicating the same service under the abnormal load dimension. One of the load dimension data is used to record the real-time load information of a container under a preset load dimension. The abnormal load dimension data can accurately locate the abnormal load dimension and the configuration information of the abnormal container. At the same time, the load reference value indicating the target service under the abnormal load dimension is taken into consideration when adjusting container resources. The recording rules of the load dimension data in this application reduce the difficulty of judging container resource adjustments and improve the accuracy and efficiency of container resource adjustments. The recording rules of the load dimension data provide conditions for the introduction of more load dimensions. Based on the load dimensions with finer granularity, the flexibility of container resource management and the efficiency of container load data analysis can be improved.

[0058] This application also provides a container resource adjustment device, such as... Figure 11 As shown, the container resource adjustment device 110 includes: Response module 1101: Used to respond to the detected abnormal load dimension data, determine the corresponding abnormal load dimension and the configuration information of the abnormal container; wherein, a load dimension data is used to record the real-time load information of a container under a preset load dimension. Target service determination module 1102: used to determine the target service to which the abnormal container belongs based on the configuration information of the abnormal container; Load reference value acquisition module 1103: used to acquire the load reference value indicating the target service under the abnormal load dimension; Adjustment module 1104: used to adjust container resources according to the load reference value and the preset elastic scaling strategy.

[0059] It should be noted that the apparatus and method embodiments described in the device embodiments are based on the same inventive concept.

[0060] The container cloud platform involved in this application embodiment can be a distributed system formed by connecting clients and multiple nodes (any form of computing device in the network, such as servers and user terminals) through network communication.

[0061] Taking a distributed system as an example, see blockchain system. Figure 12 , Figure 12 This is an optional structural diagram of the distributed system 100 provided in this application embodiment applied to a blockchain system. It consists of multiple nodes (any form of computing device in the network, such as servers or user terminals) and clients, forming a peer-to-peer (P2P) network. The P2P protocol is an application layer protocol running on top of the Transmission Control Protocol (TCP). In the distributed system, any machine, such as a server or terminal, can join and become a node. A node includes a hardware layer, a middleware layer, an operating system layer, and an application layer.

[0062] See Figure 12 The functions of each node in the blockchain system shown include: 1) Routing: A basic function of nodes used to support communication between nodes.

[0063] In addition to routing capabilities, nodes can also have the following functions: 2) Applications are deployed in the blockchain to implement specific business needs. They record data related to the implementation of functions to form record data, carry digital signatures in the record data to indicate the source of the task data, and send the record data to other nodes in the blockchain system. When other nodes successfully verify the source and integrity of the record data, they add the record data to a temporary block.

[0064] For example, the business logic implemented by the application includes: 2.1) A wallet is used to provide the function of conducting electronic currency transactions, including initiating transactions (i.e., sending the transaction record of the current transaction to other nodes in the blockchain system; after other nodes successfully verify the transaction, they store the transaction record data in the temporary block of the blockchain as a response to acknowledge the validity of the transaction; of course, the wallet also supports querying the remaining electronic currency in the electronic currency address. 2.2) Shared ledger, used to provide functions such as storage, query and modification of ledger data. It sends the record data of the operation on the ledger data to other nodes in the blockchain system. After the other nodes verify the validity, as a response to acknowledge the validity of the ledger data, they store the record data in a temporary block. They can also send confirmation to the node that initiated the operation.

[0065] 2.3) Smart contracts are computerized protocols that can execute the terms of a contract. They are implemented through code deployed on a shared ledger that executes when certain conditions are met. Based on actual business needs, the code is used to complete automated transactions, such as querying the logistics status of goods purchased by a buyer and transferring the buyer's electronic money to the merchant's address after the buyer signs for the goods. Of course, smart contracts are not limited to executing contracts for transactions; they can also execute contracts for processing received information.

[0066] 3) A blockchain consists of a series of blocks that are sequentially generated. Once a new block is added to the blockchain, it will not be removed. The blocks contain the data submitted by the nodes in the blockchain system.

[0067] See Figure 13 , Figure 13 This is an optional schematic diagram of the block structure provided in this application embodiment. Each block includes the hash value of the transaction records stored in this block (the hash value of this block) and the hash value of the previous block. The blocks are connected through their hash values ​​to form a blockchain. Additionally, the block may include information such as a timestamp when it was generated. A blockchain is essentially a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains relevant information used to verify the validity of the information (anti-counterfeiting) and to generate the next block.

[0068] This application provides an electronic device including a processor and a memory. The memory stores at least one instruction or at least one program segment, which is loaded and executed by the processor to implement the container resource adjustment method provided in the above method embodiments.

[0069] Furthermore, Figure 14A schematic diagram of the hardware structure of an electronic device for implementing the container resource adjustment method provided in the embodiments of this application is shown. The electronic device may participate in or include the container resource adjustment apparatus provided in the embodiments of this application. Figure 14 As shown, the electronic device 140 may include one or more processors 1402 (shown as 1402a, 1402b, ..., 1402n in the figure) (processor 1402 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 1404 for storing data, and a transmission device 1406 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 14 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device described above. For example, the electronic device 140 may also include... Figure 14 The more or fewer components shown, or having the same Figure 14 The different configurations shown.

[0070] It should be noted that the aforementioned one or more processors 1402 and / or other data processing circuitry are generally referred to herein as "data processing circuitry". This data processing circuitry may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be wholly or partially integrated into any other element within the electronic device 140 (or mobile device). As involved in the embodiments of this application, the data processing circuitry serves as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0071] The memory 1404 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the container resource adjustment method described in this embodiment. The processor 1402 executes various functional applications and data processing by running the software programs and modules stored in the memory 144, thereby implementing the aforementioned container resource adjustment method. The memory 1404 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1404 may further include memory remotely located relative to the processor 1402, and these remote memories can be connected to the electronic device 140 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0072] The transmission device 1406 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the electronic device 140. In one example, the transmission device 1406 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In one embodiment, the transmission device 1406 may be a radio frequency (RF) module for wireless communication with the Internet.

[0073] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows a user to interact with the user interface of the electronic device 140 (or mobile device).

[0074] Embodiments of this application also provide a computer-readable storage medium, which can be disposed in an electronic device to store at least one instruction or at least one program related to implementing a container resource adjustment method in the method embodiment. The at least one instruction or the at least one program is loaded and executed by the processor to implement the container resource adjustment method provided in the above method embodiment.

[0075] Optionally, in this embodiment, the storage medium may be located at at least one of the multiple network servers in a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0076] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0077] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and electronic device embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0078] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0079] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for adjusting container resources, characterized in that, The method includes: In response to detected abnormal load dimension data, determine the corresponding abnormal load dimension and the configuration information of the abnormal container; a load dimension data is used to record the real-time load information of a container under a preset load dimension. The target service to which the abnormal container belongs is determined based on the configuration information of the abnormal container; Obtain the load reference value indicating the target service under the abnormal load dimension: Determine a reference container belonging to the same target service as the abnormal container; when the abnormal load value in the abnormal load dimension data is greater than a first preset value, obtain the reference load dimension data of the reference container indicating the abnormal load dimension, and obtain the load reference value indicating the target service under the abnormal load dimension based on the abnormal load value and the load value in the reference load dimension data; when the abnormal load value in the abnormal load dimension data is less than a second preset value, obtain the first load profile data of the abnormal container indicating the abnormal load dimension, obtain the second load profile data of the reference container indicating the abnormal load dimension, and obtain the load reference value indicating the target service under the abnormal load dimension based on the first load profile data and the second load profile data; the first preset value indicates the upper limit of the load of the container under the abnormal load dimension when the load is at a normal level, and the second preset value indicates the lower limit of the load of the container under the abnormal load dimension when the load is at a normal level. Container resources are adjusted according to the load reference value and the preset elastic scaling strategy; when the load reference value meets the preset expansion conditions, the expansion process is initiated, and when the load reference value meets the preset shrinkage conditions, the shrinkage process is initiated. The method further includes: extracting target load dimension data falling within the current time window from the locally stored load dimension dataset; generating corresponding load profile data according to a second preset data format and the target load dimension data; the second preset data format indicates fields for recording load dimensions, fields for recording the time window, fields for recording load statistics, and fields for recording container configuration information.

2. The method according to claim 1, characterized in that, The method further includes: Load monitoring is performed on each container in the container cluster under the preset load dimension; At preset intervals, corresponding load dimension data is created based on a first preset data format and the load monitoring data at the current time; wherein, the first preset data format indicates fields for recording load dimensions, fields for recording creation time, fields for recording load values, and fields for recording container configuration information.

3. The method according to claim 1, characterized in that, The step of adjusting container resources based on the load reference value and the preset elastic scaling strategy includes: A third preset value is determined based on a first preset elastic scaling strategy that indicates expansion conditions; wherein, the third preset value is the maximum value that indicates the normal operation of the target service under the abnormal load dimension; When the load reference value is greater than the third preset value, the amount of resources required for expansion is calculated based on the load reference value and the third preset value, and the container resources are expanded based on the calculation results.

4. The method according to claim 1, characterized in that, The step of adjusting container resources based on the load reference value and the preset elastic scaling strategy includes: A fourth preset value is determined based on a second preset elastic scaling strategy that indicates scaling down conditions; wherein, the fourth preset value is the minimum value that indicates the normal operation of the target service under the abnormal load dimension; When the load reference value is less than the fourth preset value, the amount of resources required for scaling down is calculated based on the load reference value and the fourth preset value, and the container resources are scaled down based on the calculation results.

5. A container resource adjustment device, characterized in that, The device includes: Response module: Used to respond to the detected abnormal load dimension data, determine the corresponding abnormal load dimension and the configuration information of the abnormal container; a load dimension data is used to record the real-time load information of a container under a preset load dimension; Target service determination module: used to determine the target service to which the abnormal container belongs based on the configuration information of the abnormal container; Load Reference Value Acquisition Module: Used to acquire the load reference value indicating the target service under the abnormal load dimension: determine a reference container belonging to the same target service as the abnormal container; when the abnormal load value in the abnormal load dimension data is greater than a first preset value, acquire the reference load dimension data of the reference container indicating the abnormal load dimension, and obtain the load reference value indicating the target service under the abnormal load dimension based on the abnormal load value and the load value in the reference load dimension data; when the abnormal load value in the abnormal load dimension data is less than a second preset value, acquire the first load profile data of the abnormal container indicating the abnormal load dimension, acquire the second load profile data of the reference container indicating the abnormal load dimension, and obtain the load reference value indicating the target service under the abnormal load dimension based on the first load profile data and the second load profile data; the first preset value indicates the upper limit of the load of the container under the abnormal load dimension when the load is at a normal level, and the second preset value indicates the lower limit of the load of the container under the abnormal load dimension when the load is at a normal level. Adjustment module: used to adjust container resources according to the load reference value and the preset elastic scaling strategy. When the load reference value meets the preset expansion conditions, the expansion process is initiated; when the load reference value meets the preset shrinkage conditions, the shrinkage process is initiated. The device is further configured to: extract target load dimension data falling within the current time window from a locally stored load dimension dataset; generate corresponding load profile data according to a second preset data format and the target load dimension data; the second preset data format indicates fields for recording load dimensions, fields for recording the time window to which the data belongs, fields for recording load statistics, and fields for recording container configuration information.

6. The apparatus according to claim 5, characterized in that, The device is also used for: Load monitoring is performed on each container in the container cluster under the preset load dimension; At preset intervals, corresponding load dimension data is created based on a first preset data format and the load monitoring data at the current time; wherein, the first preset data format indicates fields for recording load dimensions, fields for recording creation time, fields for recording load values, and fields for recording container configuration information.

7. The apparatus according to claim 5, characterized in that, The adjustment module is also used for: A third preset value is determined based on a first preset elastic scaling strategy that indicates expansion conditions; wherein, the third preset value is the maximum value that indicates the normal operation of the target service under the abnormal load dimension; When the load reference value is greater than the third preset value, the amount of resources required for expansion is calculated based on the load reference value and the third preset value, and the container resources are expanded based on the calculation results.

8. The apparatus according to claim 5, characterized in that, The adjustment module is also used for: A fourth preset value is determined based on a second preset elastic scaling strategy that indicates scaling down conditions; wherein, the fourth preset value is the minimum value that indicates the normal operation of the target service under the abnormal load dimension; When the load reference value is less than the fourth preset value, the amount of resources required for scaling down is calculated based on the load reference value and the fourth preset value, and the container resources are scaled down based on the calculation results.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the container resource adjustment method as described in any one of claims 1-4.

10. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction or at least one program segment, which is loaded and executed by a processor to implement the container resource adjustment method as described in any one of claims 1-4.

11. A computer program product, characterized in that, The computer program product includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the container resource adjustment method as described in any one of claims 1-4.

Images