Container adjustment method and apparatus, and device, storage medium and program product

Abstract

Provided in the present disclosure are a container adjustment method and apparatus, and a device, a storage medium and a program product. The method comprises: determining a data collection moment for performing data collection on a target pod , wherein the data collection moment is the current moment or a moment after the current moment; at the data collection moment, making a request to the target pod to acquire current service information of the target pod; and performing scaling-up / down processing on the target pod on the basis of the current service information. Therefore, the timeliness of scaling-up / down performed on a pod is improved.

Description

[0001] Container Adjustment Method, Apparatus, Device, Storage Medium, and Program Product This disclosure claims priority to Chinese patent application number 202311846929.1, filed with the China Patent Office on December 28, 2023, entitled "Container Adjustment Method, Apparatus, and Device," the entire contents of which are incorporated herein by reference. Technical Field This disclosure relates to the field of computers, and more particularly to a container adjustment method, apparatus, device, storage medium, and program product. Background: A container orchestration engine (Kubernetes) cluster can include multiple container groups (pods), each of which can include multiple containers and be used to run services. In related technologies, service metrics for a container group can be collected and stored in a time series database. A server can obtain the service metrics from the time series database and scale the container group based on the service metrics. However, in the above approach, because the business metrics acquired are data collected at a historical time and stored in a time series database relative to the current time of scaling, the latency in acquiring the business metrics is significant, resulting in poor timeliness in scaling container groups. SUMMARY Various aspects of the present disclosure provide a container adjustment method, apparatus, device, storage medium, and program product for improving the timeliness of scaling container groups. In a first aspect, embodiments of the present disclosure provide a container adjustment method, comprising: determining a data collection time for data collection from a target container group, the data collection time being the current time or a time after the current time; requesting current business information of the target container group from the target container group at the data collection time; and scaling the target container group based on the current business information. In one possible implementation, determining the data collection time for data collection from the target container group includes: acquiring historical business information and historical scaling information of the target container group; and determining the data collection time based on the historical business information and historical scaling information. In one possible implementation, determining the data collection time based on the historical business information and the historical expansion and contraction information includes: estimating estimated business information of the target container group in a future time period based on the historical business information, where the estimated business information includes an indicator value of at least one business indicator; determining estimated expansion and contraction information of the target container group in the future time period based on the historical expansion and contraction information, where the estimated expansion and contraction information includes a probability of the target container group expanding or contracting at each time point in the future time period; and determining the data collection time based on the estimated business information and the estimated expansion and contraction information.In one possible implementation, determining the data collection time based on the estimated service information and the estimated scaling information includes: determining M first time points based on the estimated service information, where a service indicator value in the estimated service information at the first time point is greater than or equal to a corresponding preset threshold, where M is an integer; determining N second time points based on the estimated scaling information, where a probability of the target container group scaling at the second time point is greater than or equal to a preset probability, where N is an integer; and determining the data collection time based on the M first time points and the N second time points. In one possible implementation, determining the data collection time based on the M first time points and the N second time points includes: determining the earliest time point among the M first time points and the N first time points as the target time point; determining a preset lead time; and determining the data collection time based on the lead time and the target time point, where the data collection time point is before the target time point, and the time difference between the data collection time point and the target time point is equal to the lead time. In one possible implementation, determining a data collection time for data collection for a target container group includes: determining a data collection cycle corresponding to the target container group; determining a third time at which the target container group last collected data; and determining the data collection time based on the data collection cycle and the third time, where the duration between the third time and the data collection time is the duration corresponding to the data collection cycle. In one possible implementation, requesting the target container group to obtain current business information for the target container group at the data collection time includes: determining at least one business indicator and the data collection time corresponding to each business indicator; sending a data collection request to the target container group at the data collection time, the data collection request including the at least one business indicator and the data collection time corresponding to each business indicator; and receiving the current business information sent by the target container group, the current business information including the indicator value of each business indicator within the corresponding data collection time. In a possible implementation, the method is applied to a server, wherein a collection component is provided in the server, and the collection component is directly connected to the target container group; and sending a data collection request to the target container group at the data collection moment includes: sending the data collection request to the target container group through the collection component at the data collection moment.In one possible implementation, receiving the current business information sent by the target container group includes: receiving, via the collection component, multiple data streams sent by the target container group, the data streams including indicator values ​​of each business indicator collected at at least one collection time; wherein, for any data stream, the time difference between the time when the target container group sends the data stream and the time when the target container group collects the indicator value in the data stream is less than or equal to a preset duration. In another possible implementation, scaling the target container group based on the current business information includes: obtaining resource usage information of the target container group; and scaling the target container group based on the current business information and the resource usage information. In a possible implementation, the current business information includes multiple indicator values ​​corresponding to each business indicator of at least one business indicator, and the resource usage information includes resource occupancy information corresponding to each resource type of multiple resource types; scaling the target container group based on the current business information and the resource usage information includes: for any business indicator, determining the indicator type of the business indicator based on the multiple indicator values ​​corresponding to the business indicator in the current business information, where the indicator type is an abnormal type or a normal type; for any resource type, determining the resource usage type corresponding to each resource type based on the multiple resource occupancy rates corresponding to each resource type in the resource usage information, where the resource usage type is an abnormal type or a normal type; if there is an abnormal business indicator in the current business information and / or there is an abnormal resource type in the resource usage information, scaling the target container group. In a second aspect, embodiments of the present disclosure provide a container adjustment device, comprising: a determination module, an acquisition module, and a processing module. The determination module is configured to determine a data collection time for collecting data from a target container group, where the data collection time is the current time or a time after the current time. The acquisition module is configured to request current service information of the target container group from the target container group at the data collection time. The processing module is configured to scale the target container group based on the current service information. In one possible implementation, the determination module is specifically configured to: acquire historical service information and historical scaling information of the target container group; and determine the data collection time based on the historical service information and historical scaling information.In one possible implementation, the determination module is specifically configured to: estimate estimated business information of the target container group in a future time period based on the historical business information, the estimated business information including the value of at least one business indicator; determine estimated scaling information of the target container group in the future time period based on the historical scaling information, the estimated scaling information including the probability of scaling the target container group at each time point in the future time period; and determine the data collection time based on the estimated business information and the estimated scaling information. In one possible implementation, the determination module is specifically configured to: determine, based on the estimated business information, M first time points at which the value of a business indicator in the estimated business information at the first time point is greater than or equal to a corresponding preset threshold, where M is an integer; determine, based on the estimated scaling information, N second time points at which the probability of scaling the target container group at the second time point is greater than or equal to a preset probability, where N is an integer; and determine the data collection time based on the M first time points and the N second time points. In one possible implementation, the determination module is specifically configured to: determine the earliest of the M first moments and the N first moments as the target moment; determine a preset lead time; and determine the data collection moment based on the lead time and the target moment, wherein the data collection moment is before the target moment, and the time difference between the data collection moment and the target moment is equal to the lead time. In another possible implementation, the determination module is specifically configured to: determine a data collection cycle corresponding to the target container group; determine a third moment of the last data collection performed by the target container group; and determine the data collection moment based on the data collection cycle and the third moment, wherein the duration between the third moment and the data collection moment is the duration corresponding to the data collection cycle. In one possible implementation, the acquisition module is specifically configured to: determine at least one business indicator and the data collection duration corresponding to each business indicator; at the data collection time, send a data collection request to the target container group, the data collection request including the at least one business indicator and the data collection duration corresponding to each business indicator; and receive current business information from the target container group, the current business information including the indicator value of each business indicator within the corresponding data collection duration. In one possible implementation, the acquisition module is implemented in a server, the server being provided with a collection component directly connected to the target container group; and at the data collection time, send the data collection request to the target container group via the collection component.In one possible implementation, the acquisition module is specifically configured to receive, via the collection component, multiple data streams sequentially sent by the target container group, where the data streams include the indicator value of each business indicator collected at at least one collection time. For any data stream, the time difference between the time when the target container group sends the data stream and the time when the target container group collects the indicator value in the data stream is less than or equal to a preset duration. In another possible implementation, the processing module is specifically configured to obtain resource usage information of the target container group; and scale the target container group up or down based on the current business information and the resource usage information. In one possible implementation, the current business information includes multiple indicator values ​​corresponding to each of at least one business indicator, and the resource usage information includes resource occupancy information corresponding to each of multiple resource types. The processing module is specifically configured to: determine, for any business indicator, the indicator type of the business indicator based on the multiple indicator values ​​corresponding to the business indicator in the current business information, where the indicator type is either abnormal or normal; determine, for any resource type, the resource usage type corresponding to each resource type based on the multiple resource occupancy rates corresponding to each resource type in the resource usage information, where the resource usage type is either abnormal or normal; and if an abnormal business indicator exists in the current business information and / or an abnormal resource type exists in the resource usage information, perform scaling on the target container group. In a third aspect, an embodiment of the present disclosure provides a server comprising: a memory and a processor; the memory storing computer-executable instructions; and the processor executing the computer-executable instructions stored in the memory, such that the processor performs any of the methods described in the first aspect. In a fourth aspect, embodiments of the present disclosure provide a computer-readable storage medium storing computer-executable instructions. When executed by a processor, the computer-executable instructions are used to implement the method described in any one of the first aspects. In a fifth aspect, embodiments of the present disclosure provide a computer program product, including a computer program. When executed by a processor, the computer program implements the method described in any one of the first aspects. Embodiments of the present disclosure provide a container adjustment method, apparatus, device, storage medium, and program product. A server can determine a data collection time for data collection on a target container group and, at the data collection time, request current service information of the target container group from the target container group. The server can then scale the target container group based on the current service information.Because the server can determine the data collection time and promptly obtain the current business information of the target container group at that time, compared to related art, multiple indicator values ​​of at least one business indicator in the current business information are collected currently rather than at historical times. This reduces the latency in obtaining business indicators, thereby comprehensively improving the timeliness of scaling the container group. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings described herein are provided to provide a further understanding of the present disclosure and constitute a part of this disclosure. The exemplary embodiments of this disclosure and their descriptions are intended to explain this disclosure and do not constitute undue limitations of this disclosure. In the drawings: Figure 1 is a schematic diagram of a scenario provided by an exemplary embodiment of the present disclosure; Figure 2 is a schematic diagram of a container adjustment process in related art; Figure 3 is a flow chart of a container adjustment method provided by an exemplary embodiment of the present disclosure; Figure 4 is a flow chart of another container adjustment method provided by an exemplary embodiment of the present disclosure; Figure 5 is a process diagram of a container adjustment method provided by an exemplary embodiment of the present disclosure; Figure 6 is a schematic diagram of the structure of a container adjustment device provided by an exemplary embodiment of the present disclosure; and Figure 7 is a schematic diagram of the structure of a server provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, storage, and display) involved in this disclosure are all authorized by the user or fully authorized by all parties. The collection, use, and processing of the relevant data must comply with relevant laws, regulations, and standards, and corresponding operation portals are provided for users to choose to authorize or reject. To further clarify the objectives, technical solutions, and advantages of this disclosure, the technical solutions of this disclosure will be clearly and completely described below in conjunction with specific embodiments of this disclosure and the corresponding drawings. Obviously, the described embodiments are only some of the embodiments of this disclosure, and are not exhaustive. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure. Figure 1 is a schematic diagram of a scenario provided by an exemplary embodiment of this disclosure. Referring to Figure 1, a container orchestration engine (Kubernetes) cluster may include multiple container groups, and each container group may include multiple containers. For example, a Kubernetes cluster can include container group 1 and container group 2. Container group 1 can include container 1-1 and container 1-2, and container group 2 can include container 2-1 and container 2-2. Services can run in container groups. For example, service 1 can run in both container group 1 and container group 2.The server can collect current business information from the target container group and scale the container group based on the current business information. The target container group may include container group 1 and container group 2. For example, if the business indicator included in the current business information is the number of network connections, when the number of network connections for both container group 1 and container group 2 exceeds a connection threshold, the container group can be expanded to add container group 3, which can include containers 3-1 and 3-2. In related technologies, business indicators for container groups can be collected and stored in a time series database. The server can obtain the business indicators from the time series database and scale the container group based on the business indicators. However, in this approach, since the business indicators obtained are data collected at a historical time and stored in the time series database relative to the current time of scaling, the latency in obtaining the business indicators is significant, resulting in poor timeliness in scaling the container group. In an embodiment of the present disclosure, a server can determine a data collection time for a target container group and collect current business information from the target container group based on the data collection time. This can then be used to scale the target container group based on the current business information. Because the server can determine the data collection time and collect current business information in a timely manner, the latency of collecting business information is reduced, thereby improving the timeliness of scaling the container group. The following describes the container adjustment process in the related art with reference to Figure 2. Figure 2 is a schematic diagram of the container adjustment process in the related art. As shown in Figure 2, a Kubernetes cluster includes a Horizontal Pod Autoscaling (HPA) component, a Kubernetes service component, a Prometheus collection component, a node management component, multiple container groups, a Prometheus monitoring component, a Prometheus service component, and a time series database. The Kubernetes service component and the Prometheus collection component can be installed on node 1. Node 2 can be equipped with a node management component, which can include a monitoring tool. The monitoring tool, which can be CAdvisor, can be used to obtain resource usage information for a node and the container group on the node. A node can be a virtual machine or server, for example. Resource usage information can include processor utilization, memory utilization, and so on. Steps 1, 2, and 3 are the process of collecting container data, while steps 4, 5, and 6 are the process of the HPA component obtaining container data. The data collection and acquisition processes are relatively independent and do not interfere with each other. The container data can include resource usage information and multiple indicator values ​​of at least one business indicator.In step 1, the Prometheus monitoring component can periodically collect data from container group 1 and container group 2 to obtain multiple indicator values ​​corresponding to at least one business indicator. The Prometheus service component can then send these multiple indicator values ​​to the Prometheus service component. In step 2, the Prometheus service component can store the multiple indicator values ​​corresponding to the at least one business indicator in a time series database. In step 3, the monitoring tool in the node management component can periodically collect data from container group 1 and container group 2 to obtain and store resource usage information for these two groups. In step 4, the HPA component can send a request to obtain business indicators to the Prometheus collection component through the aggregation interface, and can send a request to obtain resource information to the Kubernetes service component. In step ⑤, the Kubernetes service component can send a request to obtain resource usage information to the node management component, and the node management component can send pre-collected historical resource usage information to the Kubernetes service component. Then, the node management component can send the historical resource usage information to the HPA component through the aggregation interface, so that the HPA component obtains the historical resource usage information; in step ⑥, the Prometheus collection component can send a request to obtain business indicators to the Prometheus service component. The Prometheus service component can determine multiple historical indicator values ​​corresponding to at least one pre-collected business indicator in the time series database, and send the multiple historical indicator values ​​to the Prometheus collection component. Then, the Prometheus collection component can send the multiple historical indicator values ​​to the HPA component through the aggregation interface, so that the HPA component obtains the multiple historical indicator values.

[0002] The HPA component can scale container groups based on historical resource usage information and multiple historical indicator values. In related technologies, because the process of collecting container data and the HPA component obtaining container data are performed asynchronously, the business indicators obtained by the HPA component are data collected in advance at historical moments and stored in a time series database, not current business indicators. This means that the business indicators obtained by the HPA component experience a significant delay, ranging from 30 seconds to over 1 minute. This results in poor timeliness for the server to scale container groups based on container data through the HPA component, and the inability to flexibly add custom business indicators. The technical solutions presented in this disclosure are described in detail below through specific embodiments. It should be noted that the following embodiments may exist independently or in combination, and identical or similar content will not be repeated in different embodiments. Figure 3 is a flow chart of a container adjustment method provided by an exemplary embodiment of this disclosure. Referring to Figure 3, the method may include:

[0003] S301. Determine a data collection time for data collection on a target container group. The execution entity of the disclosed embodiment may be a server or a container adjustment device installed in the server. The container adjustment device may be implemented via software or a combination of software and hardware. The container adjustment device may be a processor in the server. For ease of understanding, the following description uses a server as an example. The target container group refers to the container group in the Kubernetes cluster for which data collection is required. Optionally, there may be at least one target container group. The data collection time may be the current time or a time after the current time. In an optional embodiment, the data collection time for data collection on the target container group may be determined by: obtaining historical service information and historical scaling information of the target container group; and determining the data collection time based on the historical service information and historical scaling information. Service information refers to information generated when the target container group runs services. For example, the service information may be the number of network connections. The historical service information may include the indicator value of at least one service indicator at multiple first historical moments within a first historical period. The first historical period may be preset. For example, if the target container group includes container group 1 and container group 2, the service indicator is the number of network connections, the current time is 15:30, and the first historical time is 10 minutes before the current time, then the historical service information may include multiple historical network connection numbers of container group 1 in the past 10 minutes, as shown in Table 1: Table 1 The historical expansion and contraction information may include multiple second historical moments and the number of historical container groups at each second historical moment. For example, the historical expansion and contraction information may be as shown in Table 2: Table 2 For example, if the current time is 3:30 PM, the server can obtain the historical business information of the target container group as shown in Table 1 and the historical scaling information as shown in Table 2. Based on this historical business and scaling information, the server can predict the data collection time. Assuming the data collection time is determined to be 3:55 PM, the above method can be used to determine the data collection time and dynamically adjust the collection interval. When the value of a business indicator fluctuates significantly, the time interval between data collection times can be shorter (i.e., the collection interval is shorter and the frequency of indicator value collection is higher). This improves the accuracy of scaling based on multiple business indicator values. When the fluctuation of a business indicator is smaller, the time interval between data collection times can be longer (i.e., the collection interval is longer and the frequency of indicator value collection is lower), which can reduce server energy consumption. For example, if the number of network connections fluctuates significantly, the time interval between data collection time 1 and data collection time 2 for collecting the number of network connections can be set to 1 minute. If the number of network connections fluctuates smoothly, the time interval between data collection time 1 and data collection time 2 for collecting the number of network connections can be set to 30 minutes. This reduces the frequency of collecting business metrics and thus reduces server energy consumption. In another alternative embodiment, the data collection time for collecting data for the target container group can be determined as follows: determining the data collection period corresponding to the target container group; determining the third time at which data was last collected for the target container group; and determining the data collection time based on the data collection period and the third time. Optionally, the data collection period can be preset. For example, the data collection period can be 30 seconds. The duration between the third time and the data collection time can be the duration corresponding to the data collection period. For example, if the target container groups are container group 1 and container group 2, the server can determine the data collection periods corresponding to container group 1 and container group 2. Assume that the data collection periods for container group 1 and container group 2 are both 30 seconds. Assuming that the server can determine that the third time of the last data collection for container group 1 and container group 2 is 15:00:00, the server can determine the data collection time as 15:00:30 based on the data collection cycle and the third time.

[0004] S302. At the data collection time, a request is made to the target container group to obtain the current business information of the target container group. The current business information may include the value of at least one business indicator. The server may be provided with a collection component. At the data collection time, the server may send a data collection request to the target container group through the collection component to request the current business information of the target container group. The collection component may be a collection component for custom collection indicators. The collection component may be directly connected to the target container group. The collection component may communicate with the target container group via the Hypertext Transfer Protocol (HTTP). The server may collect custom business indicators through the collection component. For example, if the business indicator is the number of network connections, if the data collection time is 15:00:30, and the target container groups are container group 1 and container group 2, then based on the data collection time, collection request 1 can be sent to container group 1 to request the current number of network connections 1 of container group 1; collection request 2 can be sent to container group 2 to request the current number of network connections 2 of container group 2. Assume that the current number of network connections 1 is 2200 and the current number of network connections 2 is 2100.

[0005] S303. Scale the target container group based on the current business information. Since the current business information includes the value of at least one business indicator, after obtaining the current business information, the server can determine a preset threshold corresponding to the at least one business indicator. If the value of a business indicator in the current business information is greater than or equal to the corresponding preset threshold, the target container group can be scaled up. If the value of at least one business indicator in the current business information is less than the corresponding preset threshold, the target container group can be scaled down. In another optional embodiment, the server can further determine a scaling policy and scale the target container group based on the current business information and the scaling policy. Optionally, the scaling policy can be: if a business indicator in the current business information is greater than or equal to the corresponding preset threshold, then n container groups are expanded within a unit time period; if a business indicator in the current business information is less than the corresponding preset threshold and remains so for a preset time period, then n container groups are reduced within the unit time period, where n is a positive integer. Optionally, the preset threshold, preset time period, and unit time period can be manually preset. For example, if the target container group includes container group 1 and container group 2, and if the current number of network connections 1 of container group 1 is 2200 and the current number of network connections 2 of container group 2 is 2100, and the network connection threshold is 2000, since both the current number of network connections 1 and the current number of network connections 2 are greater than the network connection threshold of 2000, then one container group can be expanded within a unit time (e.g., 1 minute), i.e., container group 3 can be added. If the current number of network connections 1 of container group 1 is 1000 and the current number of network connections 2 of container group 2 is 800, both less than the network connection threshold of 2000, and both have remained below the network connection threshold of 2000 for 5 minutes, then container group 2 can be deleted within a unit time (e.g., 1 minute). In this embodiment of the present disclosure, the server can determine a data collection time for collecting data from the target container group and, at the data collection time, request the target container group to obtain current service information of the target container group. Based on the current service information, the server can then perform capacity expansion or contraction on the target container group. Because the server can determine the data collection time and promptly collect current business information, the latency in collecting business information is reduced, thereby improving the timeliness of scaling container groups. The container adjustment method described above will be further described below, based on the embodiment shown in FIG3 and in conjunction with FIG4 . FIG4 is a flow chart of another container adjustment method provided by an exemplary embodiment of the present disclosure. Referring to FIG4 , the method may include:

[0006] S401. Obtain historical business information and historical expansion and contraction information of the target container group. It should be noted that the execution process of step S401 can refer to step S301 and will not be repeated here.

[0007] S402. Estimate the estimated business information of the target container group in the future time period based on the historical business information. The estimated business information may include the indicator value of at least one business indicator. For example, if one of the business indicators is the number of network connections, the estimated business information may include the indicator value of the number of network connections. Optionally, the server may process the historical business information using a preset model to estimate the estimated business information of the target container group in the future time period based on the historical business information. For example, the preset model may be a neural network model. For example, if the target container group includes container group 1 and container group 2, and the historical business information is as shown in Table 1, if the estimated business information includes one business indicator, namely the number of network connections, the server may use the preset model to estimate the number of network connections of container group 1 and container group 2 at each future time in the future time period based on the historical business information. Assuming that the future time period is 15:55-16:00, the estimated business information may be as shown in Table 3: Table 3

[0008] S403: Determine estimated expansion / contraction information of the target container group in a future time period based on the historical expansion / contraction information. The estimated expansion / contraction information may include the probability of the target container group expanding / contracting at each moment in the future time period. Optionally, the server may process the historical expansion / contraction information using a preset model to determine the estimated expansion / contraction information of the target container group in the future time period based on the historical expansion / contraction information. For example, if the historical expansion / contraction information is as shown in Table 2, the server may determine the probability of the target container group expanding / contracting at each moment in the future based on the historical expansion / contraction information using a preset model. Assume that the probability is as shown in Table 4:

[0009] S404: Determine a data collection time based on the estimated service information and the estimated capacity expansion information. In an optional embodiment, the data collection time can be determined based on the estimated service information and the estimated capacity expansion information in the following manner: determine M first time points based on the estimated service information; determine N second time points based on the estimated capacity expansion information; and determine the data collection time based on the M first time points and the N second time points. The estimated service information contains a service indicator whose value at the first time point is greater than or equal to a corresponding preset threshold, where M is an integer. For example, if the service indicator is the number of network connections, the corresponding preset threshold is 2000. If the estimated service information is as shown in Table 3, then at the future time 15:56, the number of network connections of container group 2 (2000) is equal to the corresponding preset threshold of 2000, and at the future time 15:57, the number of network connections of container group 1 and container group 2 is greater than the corresponding preset threshold of 2000, then the future times 15:56 and 15:57 can be determined as the two first time points. The probability of scaling the target container group at the second time is greater than or equal to a preset probability, where N is an integer. For example, if the estimated scaling information is as shown in Table 4, and the preset probability is 90%, then since the probability of scaling at 16:00 in the future is 91% greater than the preset probability of 90%, 16:00 in the future can be determined as one of the two second times. In an optional embodiment, the data collection time can be determined based on M first times and N second times in the following manner: the earliest of the M first times and N second times is determined as the target time; a preset lead time is determined; and the data collection time is determined based on the lead time and the target time. The data collection time can be before the target time, with the time difference between the data collection time and the target time equal to the lead time. The lead time can be manually preset. For example, if there are two first times at 15:56 and 15:57 in the future, and one second time at 16:00 in the future, then the earliest of these three times, 15:56, can be determined as the target time. If the advance time is 1 minute, the data collection time can be determined to be 15:55 based on the advance time and the target time 15:56. In the present disclosure, through steps S401 to S405, the data collection time can be determined based on historical business information and historical expansion and contraction information to dynamically adjust the collection interval.When a business indicator's value fluctuates significantly, the collection interval can be shorter and the frequency of value collection can be higher, thereby improving the accuracy of scaling based on multiple business indicator values. When a business indicator's value fluctuates less, the collection interval can be longer and the frequency of value collection can be lower, reducing server energy consumption. Because the collection interval can be dynamically adjusted by determining the data collection time, compared to a fixed collection cycle, this reduces server energy consumption and improves the accuracy of scaling the target container group.

[0010] S405. Determine at least one business indicator and the data collection duration corresponding to each business indicator. Optionally, for any business indicator, a corresponding data collection duration may be provided. The data collection duration corresponding to each business indicator may be preset. For example, if there are two business indicators, namely the number of network connections and the bandwidth rate, the server may determine a data collection duration 1 corresponding to the number of network connections and a data collection duration 2 corresponding to the bandwidth rate. Assume that data collection duration 1 is 1 minute and data collection duration 2 is 30 seconds.

[0011] S406. At the data collection time, the collection component sends a data collection request to the target container group. The data collection request may include at least one business indicator and the data collection duration corresponding to each business indicator. For example, if the target container group includes container group 1 and container group 2, and the data collection time is 3:55 PM, the server may use the collection component to send data collection request 1 to container group 1 and data collection request 2 to container group 2. If a business indicator is the number of network connections and the corresponding data collection duration is 1 minute, both data collection request 1 and data collection request 2 may include the business indicator number of network connections and a data collection duration of 1 minute.

[0012] 5407. Receive current business information sent by the target container group. The current business information may include the indicator value of each business indicator within the corresponding data collection period. For example, if the business indicator is the number of network connections, if the data collection period is 1 minute, and if the target container group includes container group 1 and container group 2, then the current business information may include the indicator values ​​of the number of network connections within the data collection period of 1 minute and corresponding to multiple collection times, as shown in Table 5: Table 5 Optionally, the current business information sent by the target container group can be received in the following manner: A collection component receives multiple data streams sent by the target container group, where the data streams include the indicator value of each business indicator collected at at least one collection time. For any data stream, the time difference between the time when the target container group sends the data stream and the time when the target container group collects the indicator value in the data stream is less than or equal to a preset duration. The time difference between the time when the target container group sends the data stream and the time when the target container group collects the indicator value in the data stream is less than or equal to the preset duration. For example, the preset duration can be 5 seconds. After the target container group collects the last indicator value, it can immediately send the data stream. For example, if the current service information is as shown in Table 5, the collection component can receive data stream 1 sent by container group 1. Data stream 1 can include the index values ​​2000, 2100, and 2100 of the number of network connections collected at three collection times, namely, 15:55:15, 15:55:30, and 15:55:45, respectively. The collection component can receive data stream 2 sent by container group 2. Data stream 2 can include the index values ​​2160, 2180, and 2200 of the number of network connections collected at three collection times, namely, 15:55:15, 15:55:30, and 15:55:45, respectively. If the time when container group 1 collects the last index value in the data stream is 15:55:45, the time when container group 1 sends data stream 1 can be 15:55:46, and the time difference between the two times can be less than 5 seconds.

[0013] S408. Obtain resource usage information for the target container group. The resource usage information may include resource occupancy rates for each of the multiple resource types. The server may periodically collect resource usage information for the target container group using a general collection component to obtain the resource usage information for the target container group. For example, if the target container group includes container group 1 and container group 2, the resource usage information obtained may be as shown in Table 6: Table 6

[0014] S409: Scaling the target container group based on the current business information and resource usage information. In an optional embodiment, scaling the target container group based on the current business information and resource usage information can be performed as follows: For any business indicator, the indicator type of the business indicator is determined based on multiple indicator values ​​corresponding to the business indicator in the current business information. For any resource type, the resource usage type corresponding to each resource type is determined based on multiple resource occupancy information corresponding to each resource type in the resource usage information. If the current business information contains an abnormal business indicator and / or the resource usage information contains an abnormal resource type, scaling the target container group is performed. The indicator type can be abnormal or normal. If the indicator value of a business indicator is greater than or equal to a value outside a corresponding preset range, it indicates that the indicator value of the business indicator is too low or too high, and the indicator type of the business indicator can be determined as abnormal. If the indicator value of a business indicator is within the corresponding preset range, it indicates that the indicator value of the business indicator is within the appropriate range, and the indicator type of the business indicator can be determined as normal. The resource usage type can be either abnormal or normal. When the resource utilization rate of a resource type is outside the corresponding preset range, it indicates that the resource utilization rate of the resource type is too low or too high, and the resource utilization type of the resource type can be determined as abnormal. When the resource utilization rate of a resource type is within the corresponding preset range, it indicates that the resource utilization rate of the resource type is within the appropriate range, and the resource utilization type of the resource type can be determined as normal. Since any business indicator has multiple indicator values ​​corresponding to the business indicator, a statistical value corresponding to the business indicator can be determined based on the multiple indicator values ​​corresponding to the business indicator, and further, the indicator type corresponding to the business indicator can be determined based on the statistical value. Optionally, the statistical value can be the average, maximum, or minimum value of the multiple indicator values ​​corresponding to the business indicator. For example, if the current business information is as shown in Table 5, there is one business indicator, namely the number of network connections. If the corresponding preset range is [1800, 2000], and if the statistical value is the mean, it can be determined that the mean of the number of network connections in container group 1 is 2066, which is outside the corresponding preset range. Therefore, the indicator type of the number of network connections in container group 1 can be determined to be abnormal. The mean of the number of network connections in container group 2 can be determined to be 2180, which is outside the corresponding preset range. Therefore, the indicator type of the number of network connections in container group 2 can be determined to be abnormal.Since any resource type should have multiple resource utilization rates, a statistical value corresponding to the resource type can be determined based on the multiple resource utilization rates corresponding to the resource type, and further, the resource usage type corresponding to the resource type can be determined based on the statistical value. Optionally, the statistical value can be the average, maximum, or minimum value of the multiple resource utilization rates corresponding to the resource type. For example, if the current business information is as shown in Table 6, and there are two resource types, processor resources and memory resources, then for processor resources, if the corresponding preset range is [85%, 90%], and the statistical value is the average, assuming that the average processor utilization rate in container group 1 is 90.3%, which is outside the corresponding preset range, then the resource utilization type of the processor resources in container group 1 can be determined to be abnormal. If the average processor utilization rate in container group 2 is 90.6%, which is outside the corresponding preset range, then the resource utilization type of the processor resources in container group 2 can be determined to be abnormal. Similarly, for memory resources, if the corresponding preset range is [80%, 85%], assuming that the average memory usage in container group 1 is 81.7%, which is within the corresponding preset range, the resource usage type of the memory resources in container group 1 can be determined to be normal. assuming that the average memory usage in container group 2 is 80.7%, which is within the corresponding preset range, the resource usage type of the memory resources in container group 2 can be determined to be normal. Optionally, if the current business information contains an abnormal business indicator, the server can determine whether the statistical value corresponding to the business indicator is greater than the maximum value of the corresponding preset range. If so, the target container group can be expanded; if not, the target container group can be reduced. And / or, if the resource usage information contains an abnormal resource type, the server can determine whether the statistical value corresponding to the resource type is greater than the maximum value of the corresponding preset range. If so, the target container group can be expanded; if not, the target container group can be reduced.For example, if in the current business information, the average value of the number of network connections in container group 1 is 2066, and the average value of the number of network connections in container group 2 is 2180, both of which are greater than the maximum value 2000 in the corresponding preset range [1800, 2000], then it can be determined to expand the target container group, that is, to add container 3; if the resource type with the abnormal type in the resource usage information is the processor type, since the average value of the processor occupancy in container group 1 is 90.3%, and the average value of the processor occupancy in container group 2 is 90.6%, both of which are greater than the maximum value 90% in the corresponding preset range [85%, 90%], then it can be determined to expand the target container group, that is, to add container 3. In the embodiment of the present disclosure, the server can obtain historical business information and historical expansion and contraction information of the target container group, and can estimate the estimated business information of the target container group in the future time period based on the historical business information; the estimated expansion and contraction information of the target container group in the future time period can be determined based on the historical expansion and contraction information, and then based on the estimated business information and estimated expansion and contraction information, Determine the data collection time. The server can determine at least one business indicator and the data collection duration corresponding to each business indicator, and send a data collection request to the target container group at the data collection time. The server can receive current business information sent by the target container group and obtain resource usage information of the target container group. Based on the current business information and resource usage information, the server can then scale the target container group up or down. Because the server can determine the data collection time and collect current business information in a timely manner, the latency of collecting business information is reduced, thereby improving the timeliness of scaling the container group. The following, in conjunction with Figure 5, describes the container adjustment method in detail, based on any of the above embodiments. Figure 5 is a schematic diagram of the container adjustment method provided by an exemplary embodiment of the present disclosure. As shown in Figure 5, the Kubernetes cluster includes an HPA component, a Kubernetes service component, a Prometheus collection component, a node management component, multiple container groups, and a collection component. The collection component can directly communicate with the container group via HTTP. In step ①, the server can determine the data collection time and use the HPA component to send a data collection request to the collection component through the aggregation interface at the data collection time; the server can also send a resource information acquisition request to the Kubernetes service component through the HPA component.In step 2, after receiving the data collection request, the collection component can send data collection requests to container group 1 and container group 2 to obtain the current business information sent by container group 1 and container group 2, respectively. This current business information may include the value of each business indicator within the corresponding data collection period. After obtaining the current business information sent by container group 1 and container group 2, the collection component can send this information to the HPA component. In step 3, the monitoring tool in the node management component can periodically collect data from container group 1 and container group 2 to obtain and store resource usage information for container group 1 and container group 2. In step 4, the Kubernetes service component can send a resource usage information request to the node management component. The node management component can send the resource usage information to the Kubernetes service component, which can then send the resource usage information to the HPA component via the aggregation interface. After obtaining the current business and resource usage information, the HPA component can scale the target container group based on this information. In the technical solution of this disclosure, the asynchronous acquisition of business metrics is transformed into a synchronous query process. Specifically, the HPA component sends a data collection request to the collection component, allowing the collection component to directly obtain the current business information sent by the target container group. This is equivalent to the HPA component being able to synchronously query current business information, ensuring that the latency for collecting custom business metrics is less than 5 seconds. Furthermore, by directly connecting the collection component to the container group to obtain business metric data (compatible with the Kubernetes metric data specification), custom metrics can be added flexibly. In this embodiment of the disclosure, the data collection time can be determined and the current business information of the target container group can be obtained promptly at that time. Compared to related technologies, multiple indicator values ​​of at least one business metric in the current business information are collected currently rather than at historical times, reducing the latency of obtaining business metrics. Furthermore, by directly connecting the collection component to the container group to collect the business metrics of the target container group, the collection process eliminates the need for other monitoring components and service components, shortening the collection link and reducing network latency, thereby improving the timeliness of scaling the container group. Figure 6 is a schematic diagram of the structure of a container adjustment device provided by an exemplary embodiment of the disclosure.Referring to Figure 6 , the container adjustment device 10 may include a determination module 11, an acquisition module 12, and a processing module 13. The determination module 11 is configured to determine a data collection time for collecting data from a target container group, where the data collection time is the current time or a time after the current time. The acquisition module 12 is configured to request the target container group to obtain its current service information at the data collection time. The processing module 13 is configured to perform capacity expansion / contraction processing on the target container group based on the current service information. The container adjustment device provided in the embodiments of the present disclosure can implement the technical solutions shown in the above-mentioned method embodiments. The implementation principles and beneficial effects are similar and will not be further described here. In one possible implementation, the determination module 11 is specifically configured to: obtain historical service information and historical capacity expansion / contraction information of the target container group; and determine the data collection time based on the historical service information and historical capacity expansion / contraction information. In one possible implementation, the determination module 11 is specifically configured to: estimate estimated business information of the target container group in a future time period based on the historical business information, where the estimated business information includes the value of at least one business indicator; determine estimated scaling information of the target container group in the future time period based on the historical scaling information, where the estimated scaling information includes the probability of scaling the target container group at each time point in the future time period; and determine the data collection time based on the estimated business information and the estimated scaling information. In one possible implementation, the determination module 11 is specifically configured to: determine, based on the estimated business information, M first time points at which the value of a business indicator in the estimated business information at the first time point is greater than or equal to a corresponding preset threshold, where M is an integer; determine, based on the estimated scaling information, N second time points at which the probability of scaling the target container group at the second time point is greater than or equal to a preset probability, where N is an integer; and determine the data collection time based on the M first time points and the N second time points. In one possible implementation, the determination module 11 is specifically configured to: determine the earliest moment among the M first moments and the N first moments as the target moment; determine a preset advance time; and determine the data collection moment based on the advance time and the target moment, where the data collection moment is before the target moment, and a time difference between the data collection moment and the target moment is equal to the advance time.In one possible embodiment, the determination module 11 is specifically configured to: determine a data collection cycle corresponding to the target container group; determine a third time at which the target container group last collected data; and determine the data collection time based on the data collection cycle and the third time, where the duration between the third time and the data collection time is the duration corresponding to the data collection cycle. In one possible embodiment, the acquisition module 12 is specifically configured to: determine at least one business indicator and the data collection time corresponding to each business indicator; at the data collection time, send a data collection request to the target container group, the data collection request including the at least one business indicator and the data collection time corresponding to each business indicator; and receive current business information sent by the target container group, the current business information including the indicator value of each business indicator within the corresponding data collection time. In one possible embodiment, the acquisition module 12 is applied to a server, wherein the server is provided with a collection component directly connected to the target container group. The acquisition module 12 is specifically configured to: at the data collection time, send the data collection request to the target container group via the collection component. In one possible implementation, the acquisition module 12 is specifically configured to receive, via the collection component, multiple data streams sent by the target container group, where the data streams include the indicator value of each business indicator collected at at least one collection time. For any data stream, the time difference between the time when the target container group sends the data stream and the time when the target container group collects the indicator value in the data stream is less than or equal to a preset duration. In one possible implementation, the processing module 13 is specifically configured to obtain resource usage information of the target container group; and scale the target container group based on the current business information and the resource usage information.In one possible implementation, the current business information includes multiple indicator values ​​corresponding to each of at least one business indicator, and the resource usage information includes resource occupancy information corresponding to each of multiple resource types. The processing module 13 is specifically configured to: determine, for any business indicator, the indicator type of the business indicator based on the multiple indicator values ​​corresponding to the business indicator in the current business information, where the indicator type is either abnormal or normal; determine, for any resource type, the resource usage type corresponding to each resource type based on the multiple resource occupancy rates corresponding to each resource type in the resource usage information, where the resource usage type is either abnormal or normal; and if an abnormal business indicator exists in the current business information and / or an abnormal resource type exists in the resource usage information, perform capacity expansion or contraction processing on the target container group. The container adjustment device provided in the embodiments of the present disclosure can implement the technical solutions shown in the above-mentioned method embodiments. The implementation principles and beneficial effects thereof are similar and are not further described here. The exemplary embodiments of the present disclosure provide a schematic structural diagram of a server. See FIG7 . The server 20 may include a processor 21 and a memory 22. Illustratively, the processor 21 and the memory 22 are interconnected via a bus 23. The memory 22 stores computer-executable instructions; the processor 21 executes the computer-executable instructions stored in the memory 22, causing the processor 21 to perform the method described in the above method embodiment. Accordingly, embodiments of the present disclosure provide a computer-readable storage medium storing computer-executable instructions. When executed by a processor, the computer-executable instructions are used to implement the method described in the above method embodiment. Accordingly, embodiments of the present disclosure may also provide a computer program product comprising a computer program. When executed by a processor, the computer program can implement the method described in the above method embodiment. Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code. The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention.It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram. These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram. These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing the computer or other programmable device to execute a series of operational steps to produce a computer-implemented process. The instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more flowcharts and / or one or more blocks in the block diagrams. In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory. Memory may include non-volatile memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash memory. Memory is an example of a computer-readable medium. Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can implement information storage using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.It should also be noted that the terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, product, or apparatus comprising a list of elements may include not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, product, or apparatus. Without further limitation, the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, product, or apparatus comprising the recited elements. The foregoing description is merely an example of the present disclosure and is not intended to limit the present disclosure. Those skilled in the art will readily appreciate that various modifications and variations of the present disclosure are possible. Any modifications, equivalent substitutions, improvements, and the like made within the spirit and principles of the present disclosure are intended to be encompassed by the claims of the present disclosure.

Claims

1. A container adjustment method, wherein, Including: Determine a data collection time for collecting data of a target container group, where the data collection time is the current time or a time after the current time; At the data collection time, request the target container group to obtain the current service information of the target container group; and perform scaling processing on the target container group according to the current service information.

2. The method according to claim 1, wherein, Determining a data collection time for collecting data of a target container group includes: obtaining historical service information and historical scaling information of the target container group; and determining the data collection time according to the historical service information and the historical scaling information.

3. The method according to claim 2, wherein, Determining the data collection time according to the historical service information and the historical scaling information includes: estimating the estimated service information of the target container group in a future period according to the historical service information, where the estimated service information includes the index values of at least one service index; determining the estimated scaling information of the target container group in the future period according to the historical scaling information, where the estimated scaling information includes the probability of scaling the target container group at each moment in the future period; and determining the data collection time according to the estimated service information and the estimated scaling information.

4. The method according to claim 3, wherein, Determining the data collection time according to the estimated service information and the estimated scaling information includes: determining M first moments according to the estimated service information, where there are index values of service indexes at the first moments in the estimated service information that are greater than or equal to the corresponding preset thresholds, and M is an integer; determining N second moments according to the estimated scaling information, where the probability of scaling the target container group at the second moments is greater than or equal to a preset probability, and N is an integer; and determining the data collection time according to the M first moments and the N second moments.

5. The method according to claim 4, wherein Determining the data collection time according to the M first moments and the N second moments includes: determining the earliest moment among the M first moments and the N first moments as the target moment; determining a preset advance duration; and determining the data collection time according to the advance duration and the target moment, where the data collection time is before the target moment, and the time difference between the data collection time and the target moment is equal to the advance duration.

6. The method according to claim 1, wherein Determining a data collection time for collecting data of a target container group includes: determining a data collection period corresponding to the target container group; determining a third moment when the target container group was last data collected; and determining the data collection time according to the data collection period and the third moment, where the duration between the third moment and the data collection time is the duration corresponding to the data collection period.

7. The method according to any one of claims 1-6, wherein At the data collection moment, request the target container group to obtain the current service information of the target container group, including: determining at least one service metric and the data collection duration corresponding to each service metric; at the data collection moment, send a data collection request to the target container group, where the data collection request includes the at least one service metric and the data collection duration corresponding to each service metric; receive the current service information sent by the target container group, and the current service information includes the metric values of each service metric within the corresponding data collection duration.

8. The method according to claim 7, wherein, Applied to a server, a collection component is set in the server, and the collection component is directly connected to the target container group; At the data collection moment, sending a data collection request to the target container group includes: at the data collection moment, sending the data collection request to the target container group through the collection component.

9. The method according to claim 8, wherein Receiving the current service information sent by the target container group includes: through the collection component, receiving multiple data streams sent by the target container group, and the data streams include the metric values of each service metric collected at at least one collection moment; where, for any one data stream, the time difference between the moment when the target container group sends the data stream and the moment when the target container group collects the metric values in the data stream is less than or equal to a preset duration.

10. The method according to any one of claims 1-9, wherein According to the current service information, perform scaling processing on the target container group, including: obtaining the resource usage information of the target container group; performing scaling processing on the target container group according to the current service information and the resource usage information.

11. The method according to claim 10, wherein, The current service information includes multiple metric values corresponding to each service metric among at least one service metric, and the resource usage information includes the resource occupancy information corresponding to each resource type among multiple resource types; performing scaling processing on the target container group according to the current service information and the resource usage information includes: for any one service metric, determining the metric type of the service metric according to the multiple metric values corresponding to the service metric in the current service information, and the metric type is an abnormal type or a normal type; for any one resource type, determining the resource usage type corresponding to each resource type according to the multiple resource occupancy rates corresponding to each resource type in the resource usage information, and the resource usage type is an abnormal type or a normal type; if there is an abnormal type of service metric in the current service information, and / or, there is an abnormal type of resource type in the resource usage information, then perform scaling processing on the target container group.

12. A container adjustment device, wherein, Including: A determination module, an acquisition module, and a processing module, where the determination module is used to determine the data collection moment for data collection on the target container group, and the data collection The moment is the current moment or a moment after the current moment; the obtaining module is configured to request, at the data collection moment, the current service information of the target container group from the target container group; the processing module is configured to perform scaling processing on the target container group according to the current service information.

13. A server, wherein, Including: At least one processor; And a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the server to execute the method according to any one of claims 1-11.

14. A computer-readable storage medium, wherein, Computer-executable instructions are stored in the computer-readable storage medium, and when the processor executes the computer-executable instructions, the method according to any one of claims 1-11 is implemented.

15. A computer program product, comprising a computer program, wherein, When the computer program is executed by the processor, the method according to any one of claims 1-11 is implemented.

Images