System automatic capacity expansion and contraction method and device, electronic equipment and storage medium

Abstract

The invention provides an automatic capacity expansion and contraction method and device of a system, electronic equipment and a storage medium, and relates to the technical field of system operation and maintenance. The method comprises the steps of obtaining multi-dimensional index data of a target system, wherein the multi-dimensional index data at least comprises system index data and application index data; performing multi-dimensional feature extraction on the multi-dimensional index data to obtain a time sequence feature, a current state feature and a service related feature; predicting the current capacity water level of the target system based on the time sequence feature, the current state feature and the business related feature; and based on the current capacity water level, determining a target capacity expansion and contraction strategy for the target system. According to the method, intelligent system capacity expansion and contraction management can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of system operation and maintenance, and in particular to a system automatic capacity expansion and contraction method and device, an electronic device, and a storage medium. BACKGROUND

[0002] With the dynamic development of enterprise IT system services, the demand of enterprise IT system services for resources also changes dynamically. In order to adapt to the changing demand of resources of the system, the operation and maintenance personnel often manually adjust the resources of the system according to historical experience, for example, when it is determined that the current load of the system is high, the system needs to be expanded to meet the performance demand of the system; when it is determined that the current load of the system is low, the system can be contracted to avoid waste of resources. However, only according to the experience of experts to decide the expansion and contraction strategy, manual adjustment is inefficient and difficult to adapt to the real-time dynamic changes of the system. SUMMARY

[0003] The present application provides a system automatic capacity expansion and contraction method and device, an electronic device, and a storage medium, which can realize intelligent system capacity expansion and contraction management. The technical solution is as follows: According to one aspect of the present application, a system automatic capacity expansion and contraction method is provided, which comprises: obtaining multi-dimensional index data of a target system, the multi-dimensional index data comprising at least system index data and application index data; performing multi-dimensional feature extraction on the multi-dimensional index data to obtain time series features, current state features, and business-related features; predicting a current capacity level of the target system based on the time series features, the current state features, and the business-related features; determining a target capacity expansion and contraction strategy for the target system based on the current capacity level.

[0004] In combination with the first aspect, in some possible implementation manners, the determining of the target capacity expansion and contraction strategy for the target system based on the current capacity level comprises: in a case where the current capacity level is greater than a first preset threshold, determining that a target expansion strategy is executed for the target system; and in a case where the current capacity level is less than a second preset threshold, determining that a target contraction strategy is executed for the target system.

[0005] In combination with the first aspect, in some possible implementation manners, the determining of the target expansion strategy executed for the target system in the case where the current capacity level is greater than the first preset threshold comprises: in the case where the current capacity level is greater than the first preset threshold, obtaining a current resource pool index; and determining the target expansion strategy executed for the target system based on the current resource pool index.

[0006] With reference to the first aspect, in some possible implementations, the predicting the current capacity level of the target system based on the time series feature, the current state feature and the business-related feature comprises: determining a first weight corresponding to the time series feature, a second weight corresponding to the current state feature and a third weight corresponding to the business-related feature; and predicting the current capacity level of the target system based on the time series feature, the first weight, the current state feature, the second weight, the business-related feature and the third weight.

[0007] With reference to the first aspect, in some possible implementations, the method further comprises: obtaining historical index data corresponding to the target system; and identifying the first weight corresponding to the time series feature, the second weight corresponding to the current state feature and the third weight corresponding to the business-related feature based on the historical index data.

[0008] With reference to the first aspect, in some possible implementations, the method further comprises: obtaining an actual water capacity of the target system after the target scaling strategy is executed; and adjusting the target scaling strategy based on the actual water capacity and a target water capacity index.

[0009] According to another aspect of the present application, a system automatic scaling device is provided, which comprises: a first obtaining module configured to obtain multi-dimensional index data of a target system, the multi-dimensional index data comprising at least system index data and application index data; a feature extraction module configured to perform multi-dimensional feature extraction on the multi-dimensional index data to obtain a time series feature, a current state feature and a business-related feature; a load prediction module configured to predict a current capacity level of the target system based on the time series feature, the current state feature and the business-related feature; a scaling module configured to determine a target scaling strategy for the target system based on the current capacity level.

[0010] According to an aspect of the present application, an electronic device is provided, which comprises a processor and a memory storing a program, the program comprising instructions which, when executed by the processor, cause the processor to perform the system automatic scaling method as described above.

[0011] According to another aspect of the present application, a non-transitory computer-readable storage medium storing computer instructions is provided, the computer instructions being used to cause the computer to perform the system automatic scaling method as described above.

[0012] According to another aspect of the present application, a computer program product is provided, which includes computer instructions stored in a computer readable storage medium. A processor of an electronic device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the system automatic scaling method described above.

[0013] The technical scheme provided by the embodiments of the present application has at least the following beneficial effects: By collecting multi-dimensional data indicators and performing multi-dimensional feature extraction on the multi-dimensional data indicators, the current load state (current water capacity) of the target system is predicted, and then the target scaling strategy of the target system is adjusted according to the current water capacity. Intelligent system scaling management can be realized; moreover, multi-dimensional data combination can more comprehensively reflect the system load state, provide a scientific basis for subsequent scaling strategy formulation, improve the precision of system scaling decision, and ensure the stable operation of the target system. BRIEF DESCRIPTION OF DRAWINGS

[0014] In the following description of the exemplary embodiments in conjunction with the accompanying drawings, more details, features and advantages of the present application are disclosed, in which: Figure 1 A flowchart of a system automatic scaling method according to an exemplary embodiment of the present application is shown; Figure 2 is a structural schematic diagram of a system automatic scaling device provided by an embodiment of the present application; Figure 3 A structural block diagram of an exemplary electronic device that can be used to implement the embodiments of the present application is shown. DETAILED DESCRIPTION

[0015] Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms, and should not be interpreted as being limited to the embodiments set forth herein, but rather these embodiments are provided to more thoroughly and completely understand the present application. It should be understood that the drawings and embodiments of the present application are for exemplary purposes only, and are not intended to limit the scope of protection of the present application.

[0016] It should be understood that each step described in the method embodiments of the present application can be executed in different order and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the execution of the steps shown. The scope of the present application is not limited in this respect.

[0017] As used herein, the term "includes" and its variants are to be read to be analogous to "comprises," or "comprising." The term "based on" is to be interpreted as "based, at least in part, on." The term "one embodiment" means "at least one embodiment." The term "another embodiment" means "at least one additional embodiment." The term "some embodiments" means "at least some embodiments." Related terms have analogous meanings. It is noted that the terms "first," "second," and the like, as used herein do not have an ordinal meaning but are used to distinguish one element from another. It is further noted that the terms "a" or "an," as used herein, mean "one or more" or "at least one." The names of the messages or information exchanged between the various apparatuses in the embodiments of the present application are used only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0018] The schemes of the present application are described below with reference to the accompanying drawings, and the technical schemes provided by the embodiments of the present application are described in detail through specific embodiments and application scenarios.

[0019] Reference is made to Figure 1 which shows a flowchart of a system automatic scaling method according to an exemplary embodiment of the present application. An exemplary description is given by taking the case where the method is applied to an electronic device. As shown in Figure 1 , the method comprises: Step 101, obtaining multi-dimensional index data of the target system, the multi-dimensional index data at least comprising system index data and application index data.

[0020] In order to improve the accuracy of automatic scaling decisions, it is necessary to obtain a more comprehensive load state of the current system. In one possible implementation, before making the scaling decision, multi-dimensional index data of the target system is obtained, so as to comprehensively analyze the load state of the target system from multiple dimensions and reduce the possibility of misjudgment.

[0021] The multi-dimensional index data at least comprises system index data and application index data. The system index data can include CPU usage, memory occupation, disk I / O, network throughput, process number, connection number, and other system resource consumption information of the application process, and the information is written into Kafka storage for stream consumption and reading. The application index data includes application traffic pv, task processing time, warning log, error log, running state, and other application system information, and the information is written into Kafka storage for stream consumption and reading.

[0022] After the system index data and application index data are collected, the data needs to be cleaned to remove abnormal values, missing values and other noise data, and the heterogeneous data needs to be standardized. For example, the missing values are filled with 0.

[0023] In step 102, multi-dimensional feature extraction is performed on the multi-dimensional index data to obtain time series features, current state features and business-related features.

[0024] Considering that a large amount of multi-dimensional index data is collected, in order to improve the subsequent prediction efficiency, in one possible implementation, multi-dimensional feature extraction is performed on the multi-dimensional index data to obtain multi-dimensional features such as time series features, current state features and business-related features.

[0025] The time series features can be CPU utilization in the past five minutes (CPU utilization can be calculated by averaging the CPU usage collected in the past five minutes) and same-period resource consumption with a period of one day (resource consumption increases or decreases compared with the same period yesterday); the current state features can be resource usage such as CPU / memory, network throughput, system remaining resource amount, etc.; and the business-related features can be core business indicators, system entry traffic, response time, system processing success rate, etc.

[0026] In step 103, the current capacity level of the target system is predicted based on the time series features, the current state features and the business-related features.

[0027] The capacity level refers to the degree, range and bearing capacity of the resources used by a system or service. It is usually expressed as a percentage. In capacity management, the capacity level is a very important indicator for measuring the health status, resource utilization and future capacity demand of a system or service. The higher the capacity level, the closer the business is to or has reached the resource bottleneck, and expansion or additional IT resource investment is needed; the lower the capacity level, the lower the resource utilization, and there is a large optimization space, so some measures such as capacity reduction and configuration reduction can be taken.

[0028] In one possible implementation, the principal component analysis dimension reduction algorithm is applied to project the multiple time series features, current state features and business-related features onto the current capacity level, so as to represent the current load status of the target system by the current capacity level.

[0029] In view of the difference in the degree of influence of different features on the load, in order to improve the accuracy of predicting the load condition, in a possible implementation, historical index data corresponding to the target system is acquired, the influence degree of different features on the load change is identified according to the historical index data, and the feature weight corresponding to each feature is obtained, that is, the first weight corresponding to the time series feature, the second weight corresponding to the current state feature, and the third weight corresponding to the business-related feature are obtained.

[0030] Specifically, the machine learning models such as random forest, gradient boosting tree, simulated annealing, etc. can be used to train the feature weight corresponding to each feature based on the historical index data. Among them, the historical time series feature is obtained by feature extraction on the historical index data, the historical time series feature is input into the random forest algorithm, and the feature importance prediction of the historical time series feature on the load change is predicted.

[0031] In a possible implementation, step 103 can include step 103A and step 103B.

[0032] Step 103A, determining the first weight corresponding to the time series feature, the second weight corresponding to the current state feature, and the third weight corresponding to the business-related feature.

[0033] Step 103B, predicting the current capacity level of the target system based on the time series feature, the first weight, the current state feature, the second weight, the business-related feature, and the third weight.

[0034] After determining the first weight corresponding to the time series feature, the second weight corresponding to the current state, and the third weight corresponding to the business-related feature, the principal component analysis dimension reduction algorithm can be applied to predict the current capacity level of the target system according to the time series feature, the first weight, the current state feature, the second weight, the business-related feature, and the third weight.

[0035] Step 104, determining the target scaling strategy for the target system based on the current capacity level.

[0036] After predicting the current capacity level of the current target system, the target scaling strategy of the target system can be dynamically decided according to the current capacity level.

[0037] Specifically, step 104 can include step 104A and step 104B.

[0038] Step 104A, in the case where the current capacity level is greater than a first preset threshold, determining to execute the target scaling strategy on the target system.

[0039] If the current capacity water level is too high, it indicates that the system reaches a resource bottleneck, and the resource investment of the system can be increased. In a possible implementation, a first preset threshold is set, and if the current capacity water level is greater than the first preset threshold, it indicates that the system reaches a resource bottleneck, and a target expansion strategy needs to be performed on the target system to increase the resource investment of the target system.

[0040] For example, the first preset threshold can be 70%. That is, when it is determined that the current capacity water level is greater than 70%, the target expansion strategy is performed.

[0041] When determining the target expansion strategy performed on the target system, the resource state in the resource pool also needs to be considered. In a possible implementation, when it is determined that the current capacity water level is greater than the first preset threshold, the current resource pool index and the application resource usage upper limit of each application running in the target system are also obtained, and the target expansion strategy performed on the target system is determined based on the current resource pool index and the application resource usage upper limit.

[0042] The current resource pool index includes the used resource amount, the remaining resource amount, and the per-package distribution prediction value in the resource pool. The application resource usage upper limit is predefined, and the CPU / memory / disk quota package of application A is defined as 2c / 8G / 200G.

[0043] For example, if the target system determines that a 5c package needs to be expanded according to the application resource usage upper limit of the multiple applications running thereon, the remaining cpu in the resource pool is 100c, and is distributed on 20 machines, each of which has 5c, which just meets the requirement of the 5c package, and a machine can be allocated to expand the target system.

[0044] Step 104B, in a case where the current capacity water level is less than the second preset threshold, a target shrinkage strategy is determined to be performed on the target system.

[0045] If the current capacity water level is too low, it indicates that the system resource utilization rate is not high, and in order to avoid resource waste, a shrinkage strategy can be performed on the target system. In a possible implementation, a second preset threshold is set, and if the current capacity water level is less than the second preset threshold, it indicates that the system resource rate is too low, and a target shrinkage strategy needs to be performed on the target system to reduce the resource investment of the target system.

[0046] For example, the second preset threshold can be 30%. That is, when it is determined that the current capacity water level is less than 30%, the target shrinkage strategy is performed.

[0047] Optionally, in order to adapt to the dynamic change of business requirements, the target expansion strategy of the target system can also be considered to be executed in the business peak period, and the target contraction strategy of the target system can be executed in the business low peak period, so as to save resources and ensure the stability of the system.

[0048] In a possible implementation, when the target expansion and contraction strategy is determined, a reinforcement learning algorithm can be used, the state space is defined as the current capacity level of the current system, the action space and the reward value are predicted by the reinforcement learning algorithm based on the state space, the action space is the specific target expansion and contraction strategy, and the target expansion and contraction strategy can indicate a specific adjustment of the resource quota; the reward value can be the adjusted target water quantity index.

[0049] After the target expansion and contraction strategy is executed, the actual water capacity of the target system after the target expansion and contraction strategy is executed can be obtained, and then the target expansion and contraction strategy is adjusted based on the actual water capacity and the target water capacity index in combination with the reinforcement learning algorithm. Specifically, the target expansion and contraction strategy can be adjusted when the actual water capacity and the target water capacity index are greatly different.

[0050] In summary, the embodiments of the present application provide a system automatic expansion and contraction method: by collecting multi-dimensional data indexes and performing multi-dimensional feature extraction on the multi-dimensional data indexes, the current load state (current water capacity) of the target system is predicted, and then the target expansion and contraction strategy of the target system is adjusted according to the current water capacity. Intelligent system expansion and contraction management can be realized; and multi-dimensional data combination can more comprehensively reflect the system load state, provide a scientific basis for subsequent expansion and contraction strategy making, improve the precision of system expansion and contraction decision, and ensure the stable operation of the target system.

[0051] The detailed modules and system workflow of the system automatic expansion and contraction method provided in the present application are as follows: (1) Define the upper limit of application resource usage: a) used to calculate the current load and declare the upper limit of capacity, for example: the CPU / memory / disk quota package of application A is defined as: 2c / 8G / 200G; b) perform stress testing on the application, and record the thread number, file handle number, access volume, response time and other information when it reaches the above package, if the response time fluctuates obviously, return to redefine the quota package.

[0052] (2) Data collection and preprocessing: a) Collect system indicators: including CPU usage, memory usage, disk I / O, network throughput, process number, connection number, and other system resource consumption information, write the information to Kafka storage, and perform stream consumption reading; b) Collect application indicators: application traffic pv, task processing time, warning logs, error logs, running status, and other application system information, write the information to Kafka storage, and perform stream consumption reading; c) Collect resource pool indicators: used resource amount, remaining resource amount, and per-package allocation prediction value; d) Data cleaning: remove noise data such as outliers and missing values, and standardize heterogeneous data. For example, fill 0 for missing values.

[0053] (3) Multi-dimensional feature data extraction: a) Design multiple feature combinations, including time series features, current state features, and business-related features. i. Time series features: average CPU utilization in the past five minutes, and same-period resource consumption with a cycle of days; ii. Current state features: CPU / memory usage, network throughput, and system remaining resource amount; iii. Business-related features: core business indicators, system entry traffic, response time, and system processing success rate; b) Apply principal component analysis (PCA) dimension reduction algorithm to map high-dimensional data to low-dimensional space, for example, the positive correlation between access volume and CPU usage, memory usage, and the negative correlation with response time, which are projected as capacity water level load.

[0054] (4) Feature weight allocation: a) Analyze historical data to identify the impact of different features on load changes; b) Use machine learning models such as random forest, gradient boosting tree, and simulated annealing algorithm to train feature weights. i. Capacity overload range assertion: random forest is used to determine whether it is single-instance capacity overload or cluster capacity overload when the overall time consumption increases; ii. Capacity upper limit change assertion: gradient boosting tree is used to determine the impact of each application iteration version on performance changes and the influence on the capacity upper limit under the same resource amount, and to update the current system capacity upper limit in real time; iii. Most economical resource amount: simulated annealing is used to identify the optimal resource amount required for expansion / shrinkage of the current load and system capacity upper limit, to ensure the efficiency of resource usage.

[0055] (5) Dynamic feedback adjustment decision: a) Define state space, action space, and reward function, where the state includes the current system load indicators and resource usage, and the action includes adjusting resource quota; b) Provide expansion and contraction decision results, and call the deployment system execution layer interface; c) Adapt to business changes to achieve the goals of saving resources and ensuring system stability.

[0056] Please refer to Figure 2It is a structural schematic diagram of a system automatic expansion and contraction device provided by an embodiment of the present application. As shown in the example Figure 2 The device 200 includes.

[0057] The first acquisition module 201 is configured to acquire multi-dimension index data of a target system, wherein the multi-dimension index data at least includes system index data and application index data. The feature extraction module 202 is configured to perform multi-dimension feature extraction on the multi-dimension index data to obtain time sequence features, current state features and business related features. The load prediction module 203 is configured to predict a current capacity level of the target system based on the time sequence features, the current state features and the business related features. The expansion and contraction module 204 is configured to determine a target expansion and contraction strategy for the target system based on the current capacity level.

[0058] Optionally, the expansion and contraction module 204 is further configured to: determine to execute a target expansion strategy for the target system if the current capacity level is greater than a first preset threshold; determine to execute a target contraction strategy for the target system if the current capacity level is less than a second preset threshold.

[0059] Optionally, the expansion and contraction module 204 is further configured to: acquire current resource pool indexes if the current capacity level is greater than the first preset threshold; determine the target expansion strategy for the target system based on the current resource pool indexes.

[0060] Optionally, the load prediction module 203 is further configured to: determine a first weight corresponding to the time sequence features, a second weight corresponding to the current state features and a third weight corresponding to the business related features; predict the current capacity level of the target system based on the time sequence features, the first weight, the current state features, the second weight, the business related features and the third weight.

[0061] Optionally, the device further includes: a second acquisition module configured to acquire historical index data corresponding to the target system; a weight identification module configured to identify the first weight corresponding to the time sequence features, the second weight corresponding to the current state features and the third weight corresponding to the business related features based on the historical index data.

[0062] Optionally, the apparatus further comprises: a third acquisition module configured to acquire an actual water capacity of the target system after the target scale-out / in policy is executed; a policy adjustment module configured to adjust the target scale-out / in policy based on the actual water capacity and the target water capacity index.

[0063] To sum up, the embodiment of the present application provides a system automatic scale-out / in method: by collecting multi-dimensional data indexes and performing multi-dimensional feature extraction on the multi-dimensional data indexes to predict the current load state (current water capacity) of the target system, and then adjusting the target scale-out / in policy of the target system according to the current water capacity. Intelligent system scale-out / in management can be realized; moreover, multi-dimensional data combination can more comprehensively reflect the system load state, provide a scientific basis for subsequent scale-out / in policy making, improve the precision of system scale-out / in decision-making, and guarantee the stable operation of the target system.

[0064] The exemplary embodiments of the present application also provide an electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication. The memory stores a computer program capable of being executed by the at least one processor, and the computer program, when executed by the at least one processor, is configured to cause the electronic device to perform the system automatic scale-out / in method according to the embodiments of the present application.

[0065] The exemplary embodiments of the present application also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is configured to cause the computer to perform the system automatic scale-out / in method according to the embodiments of the present application.

[0066] The exemplary embodiments of the present application also provide a computer program product comprising a computer program, wherein the computer program, when executed by a processor of a computer, is configured to cause the computer to perform the network attack detection method according to the embodiments of the present application.

[0067] Reference Figure 3The present invention describes a structural block diagram of an electronic device 300 that can serve as a server or client of this application, which is an example of a hardware device that can be applied to various aspects of this application. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the application described and / or claimed herein.

[0068] like Figure 3 As shown, the electronic device 300 includes a computing unit 301, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 302 or a computer program loaded from a storage unit 308 into a random access memory (RAM) 303. The RAM 303 may also store various programs and data required for the operation of the electronic device 300. The computing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0069] Multiple components in electronic device 300 are connected to I / O interface 305, including: input unit 306, output unit 307, storage unit 308, and communication unit 309. Input unit 306 can be any type of device capable of inputting information to electronic device 300. Input unit 306 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of electronic device. Output unit 307 can be any type of device capable of presenting information and may include, but is not limited to, a display, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 308 may include, but is not limited to, disk and optical disk. Communication unit 309 allows electronic device 300 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0070] The computing unit 301 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 301 performs various methods and processes described above. For example, in some embodiments, Figure 1 The illustrated methods can be implemented as a computer software program tangibly embodied in a machine readable medium, such as the storage unit 308. In some embodiments, portions or all of the computer program can be loaded and / or installed onto the electronic device 300 via the ROM 302 and / or the communication unit 309. In some embodiments, the computing unit 301 can be configured, by way of firmware or software modules stored in the storage unit 308 or memory within the computing unit 301, to carry out Figure 1 The illustrated methods.

[0071] Program code for carrying out the methods of the present application can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or the block diagrams. The program code can be embodied in whole or in part within a machine readable medium, which can be any medium for storing or transmitting the program code. The program code can be transmitted in the form of signals over a transmission medium via a number of means known to one of skill in the art, including a wireless channel, optical, electromagnetic, or any suitable combination thereof. The program code segments can be downloaded to and / or stored within the storage unit 308 or memory within the computing unit 301.

[0072] In the context of the present application, a machine readable medium can be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine readable medium can be a machine readable signal medium or a machine readable storage medium. The machine readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine readable storage medium will include one or more lines of a program of electrical connection, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0073] As used in this application, the terms "machine-readable medium" and "computer- readable medium" refer to any computer program product, apparatus and / or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0074] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0075] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0076] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

Claims

1. A system automatic scaling method, characterized in that, The method comprises: acquiring multi-dimensional index data of a target system, the multi-dimensional index data comprising at least system index data and application index data; performing multi-dimensional feature extraction on the multi-dimensional index data to obtain time series features, current state features, and business-related features; predicting a current capacity level of the target system based on the time series features, the current state features, and the business-related features; determining a target scaling strategy for the target system based on the current capacity level.

2. The method of claim 1, wherein, The determining of the target scaling strategy for the target system based on the current capacity level comprises: determining that a target scaling-up strategy is to be performed on the target system if the current capacity level is greater than a first preset threshold value; determining that a target scaling-down strategy is to be performed on the target system if the current capacity level is less than a second preset threshold value.

3. The method of claim 2, wherein, The determining of the target scaling-up strategy to be performed on the target system if the current capacity level is greater than the first preset threshold value comprises: acquiring current resource pool indicators if the current capacity level is greater than the first preset threshold value; determining the target scaling-up strategy to be performed on the target system based on the current resource pool indicators.

4. The method according to any one of claims 1 to 3, characterized in that, The predicting of the current capacity level of the target system based on the time series features, the current state features, and the business-related features comprises: determining a first weight corresponding to the time series features, a second weight corresponding to the current state features, and a third weight corresponding to the business-related features; predicting the current capacity level of the target system based on the time series features, the first weight, the current state features, the second weight, the business-related features, and the third weight.

5. The method of claim 4, wherein, The method further comprises: acquiring historical index data corresponding to the target system; identifying the first weight corresponding to the time series features, the second weight corresponding to the current state features, and the third weight corresponding to the business-related features based on the historical index data.

6. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: acquiring an actual water capacity of the target system after the target scaling strategy is performed; adjusting the target scaling strategy based on the actual water capacity and a target water capacity indicator.

7. A system automatic expansion and contraction device, characterized in that, The apparatus comprises: a first acquiring module configured to acquire multi-dimensional index data of a target system, the multi-dimensional index data comprising at least system index data and application index data; a feature extraction module configured to perform multi-dimensional feature extraction on the multi-dimensional index data to obtain time series features, current state features, and business-related features; a load prediction module configured to predict a current capacity level of the target system based on the time series features, the current state features, and the business-related features; a scaling module configured to determine a target scaling strategy for the target system based on the current capacity level.

8. The apparatus of claim 7, wherein, The scaling module is further configured to: determine that a target scaling-up strategy is to be performed on the target system if the current capacity level is greater than a first preset threshold value; determine that a target scaling-down strategy is to be performed on the target system if the current capacity level is less than a second preset threshold value. In a case where the current capacity water level is less than a second preset threshold, it is determined to execute a target shrinkage strategy on the target system. 9.An electronic device, comprising: a processor; and a memory storing a program, wherein the program includes instructions that, when executed by the processor, cause the processor to perform the system automatic scaling method according to any one of claims 1-6.

10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to cause the computer to perform the system automatic scaling method according to any one of claims 1-6.

Images