Host load adjustment processing method and device, server and storage medium

Abstract

The invention discloses a host load adjustment processing method and device, a server and a storage medium, and relates to the technical field of servers, and the method comprises the steps: carrying out the data processing of a plurality of read operation data of a host, so as to obtain the processed operation data; inputting the one or more processed operation data into a pre-generated target clustering model for load analysis to obtain a load analysis result; performing load adjustment on the corresponding one or more elements according to the load analysis result to obtain one or more pieces of adjusted operation data; the one or more adjusted operation data is output to the display terminal, load adjustment can be automatically carried out on the operation state of one or more corresponding elements only by obtaining the load analysis result, stable operation of the system is guaranteed, and the energy consumption adjustment of the host is more flexible.

Description

Technical Field

[0001] The present application relates to the technical field of servers, and in particular, to a method, device, server, and storage medium for adjusting and processing host load. Background Art

[0002] The firmware management platform is a platform for monitoring and managing servers. As an independent hardware management system, it is equipped with an independent operating system, network protocol stack, and power management system to ensure that the server can be effectively managed under any circumstances. For example, the baseboard management controller can be used for remote management, monitoring, and fault diagnosis of the server.

[0003] In the related art, the running state of the server is mainly adjusted through a fixed and single energy consumption upper limit policy corresponding to the firmware management platform to reduce energy consumption. However, in the related art, the method of adjusting the running state of the server through a fixed and single energy consumption upper limit policy is prone to system running failures and overly rigid energy consumption adjustment. Summary of the Invention

[0004] The present application provides a method, device, server, and storage medium for adjusting and processing host load, so as to at least solve the problems that in the related art, the method of adjusting the running state of the server through a fixed and single energy consumption upper limit policy is prone to system running failures and overly rigid energy consumption adjustment.

[0005] The present application provides a method for adjusting and processing host load, including:

[0006] Reading a plurality of running data of the host collected by a preset sensor, where the host includes a plurality of components, and each component corresponds to one running data;

[0007] Performing data processing on the plurality of running data to obtain each processed running data;

[0008] Inputting one or more processed running data into a pre-generated target clustering model for load analysis to obtain a load analysis result;

[0009] Performing load adjustment on one or more components according to the load analysis result to obtain one or more adjusted running data;

[0010] Outputting the one or more adjusted running data to a display terminal.

[0011] The present application further provides a device for adjusting and processing host load, including:

[0012] A first reading module, configured to read multiple operation data of a host collected by a preset sensor, where the host includes multiple components, and each component corresponds to one operation data respectively;

[0013] A first processing module, configured to perform data processing on the multiple operation data to obtain each processed operation data;

[0014] An analysis module, configured to input one or more processed operation data into a pre-generated target clustering model for load analysis to obtain a load analysis result;

[0015] A first adjustment module, configured to perform load adjustment on one or more components according to the load analysis result to obtain one or more adjusted operation data;

[0016] An output module, configured to output the one or more adjusted operation data to a display terminal.

[0017] This application also provides a server, including: a memory, configured to store a computer program; a processor, configured to implement the steps of any of the above host load adjustment processing methods when executing the computer program.

[0018] This application also provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and the computer program, when executed by a processor, implements the steps of any of the above host load adjustment processing methods.

[0019] The host load adjustment processing method, device, server, and storage medium provided by the embodiments of this application read multiple operation data of a host collected by a preset sensor; perform data processing on the multiple operation data to obtain each processed operation data; input one or more processed operation data into a pre-generated target clustering model for load analysis to obtain a load analysis result; perform load adjustment on corresponding one or more components according to the load analysis result to obtain one or more adjusted operation data; output the one or more adjusted operation data to a display terminal. This application abandons the fixed and single energy consumption upper limit strategy. As long as the load analysis result is obtained, the operation status of one or more components can be automatically adjusted for load. Therefore, not only the stable operation of the system is ensured, but also the host energy consumption adjustment is made more flexible. Description of the Drawings

[0020] To more clearly illustrate the embodiments of this application, the following will briefly introduce the drawings required for the embodiments. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 Scenario schematic of the host load adjustment processing method provided by the embodiment of this application Figure 1 ;

[0022] Figure 2 Flow schematic of the host load adjustment processing method provided by the embodiment of this application Figure 1 ;

[0023] Figure 3 Flow schematic of the host load adjustment processing method provided by the embodiment of this application Figure 2 ;

[0024] Figure 4 Structural schematic diagram of the host load adjustment processing device provided by the embodiment of this application;

[0025] Figure 5 Structural schematic diagram of the server provided by the embodiment of this application;

[0026] Figure 6 Scenario schematic of the host load adjustment processing method provided by the embodiment of this application Figure 2 。 Detailed implementation manners

[0027] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of this application.

[0028] It should be noted that in the description of this application, the terms "include", "comprise" or any other variant thereof are intended to cover a non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. The terms "first", "second", etc. in this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0029] The firmware management platform is a platform for monitoring and managing servers. As an independent hardware management system, it is equipped with an independent operating system, network protocol stack, and power management system to ensure that the server can be effectively controlled under any circumstances. For example, the baseboard management controller can be used for remote management, monitoring, and fault diagnosis of the server. In related technologies, the running state of the server is mainly adjusted through a fixed and single energy consumption upper limit strategy corresponding to the firmware management platform to reduce energy consumption. For example, when the server temperature reaches the upper limit, the CPU of the server is downclocked to reduce energy consumption. However, in related technologies, the method of adjusting the running state of the server through a fixed and single energy consumption upper limit strategy is prone to system operation failures and overly rigid energy consumption adjustments.

[0030] To solve the above technical problems, the embodiments of the present application propose the following technical concept: The inventor considers multiple running data of the host, performs load analysis on the multiple running data based on the target clustering model to obtain the load analysis result, and performs load adjustment on the corresponding components according to the load analysis result to obtain the corresponding adjusted running data, which not only ensures the normal operation of the system but also makes the host energy consumption adjustment more flexible.

[0031] Combined with the specific application environment architecture or specific hardware architecture on which the execution of the adjustment processing method of the host load depends, the specific application environment architecture or specific hardware architecture is described herein. Refer to Figure 1 , Figure 1 This is a schematic diagram of the scenario of the adjustment processing method of the host load provided by the embodiments of the present application Figure 1 .

[0032] As Figure 1 shown, this scenario includes: server 10 and display terminal 20.

[0033] Among them, server 10 can be an independent server or a cluster composed of multiple servers.

[0034] Server 10 includes at least a preset sensor 101 and a host 102.

[0035] Display terminal 20 can be a mobile phone terminal or a personal computer terminal.

[0036] Server 10 reads multiple operation data of host 102 collected by a preset sensor 101, where host 102 includes multiple components, and each component corresponds to an operation data; processes the multiple operation data to obtain processed operation data; inputs one or more processed operation data into a pre-generated target clustering model for load analysis to obtain a load analysis result; adjusts the load of one or more components according to the load analysis result to obtain one or more adjusted operation data; and outputs one or more adjusted operation data to display terminal 20.

[0037] The following uses specific embodiments to elaborate in detail on the technical solution of this application and how the technical solution of this application solves the above technical problems. These several specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following will describe the embodiments of this application in conjunction with the accompanying drawings.

[0038] Figure 2 Schematic diagram of the process for the host load adjustment processing method provided by the embodiments of this application Figure 1 , as Figure 2 shown, the embodiments of this application provide a host load adjustment processing method, and the method is described in detail as follows:

[0039] S201: Read multiple operation data of the host collected by a preset sensor, where the host includes multiple components, and each component corresponds to an operation data.

[0040] Specifically, in a pre-built management environment, read multiple operation data of the host collected by a preset sensor.

[0041] Among them, the process of building the management environment includes steps a to c:

[0042] Step a: Obtain the source code of the preset firmware management platform.

[0043] In this embodiment, the preset firmware management platform can be a BMC platform or other platforms.

[0044] Among them, BMC is a baseboard management control firmware stack, which is widely used in the management of servers and data centers and provides functions such as remote management, monitoring, and fault diagnosis.

[0045] Step b: Determine whether the preset firmware management platform can run normally according to the source code.

[0046] Step c: If the preset firmware management platform can run normally according to the source code, initialize one or more service platforms and test the preset sensors configured by the preset firmware management platform to complete the construction of the management environment.

[0047] In this embodiment, the service platform may be a Web Server, DBus, phosphor-ipmi-host, or other service platforms.

[0048] Among them, the Web Server is a website server.

[0049] Among them, DBus is an efficient and reliable inter-process communication mechanism, mainly used for message exchange and remote calls between different programs in Linux and other Unix-like systems. The design goal of DBus is to provide a communication method with low latency, low overhead, and high availability.

[0050] Among them, phosphor-ipmi-host is a software package in the BMC firmware, mainly responsible for starting the ipmid daemon process after the BMC powers on and responding to network management commands. It is an important component in the BMC project, implementing the IPMI function, allowing users to remotely manage and monitor hardware devices such as the power supply, temperature, fan speed, voltage, etc. of the server through the network.

[0051] In this embodiment, the preset sensor may be a single sensor or multiple sensors, used to read different operating data of the host.

[0052] In this embodiment, the multiple components may be a central processing unit (CPU), a memory module, and other components.

[0053] In this embodiment, the multiple operating data may be CPU utilization, memory bandwidth, process-level resource occupancy information, temperature, current, and other operating data.

[0054] Among them, CPU utilization refers to the amount of time the central processing unit of a computer spends on executing tasks and processes, usually expressed as a percentage.

[0055] Among them, memory bandwidth refers to the amount of data that a memory module can transfer through multiple memory channels per unit time, usually measured in GB / s.

[0056] Among them, the process-level resource occupancy information is obtained through sensors corresponding to the Linux kernel module.

[0057] Specifically, step S201 specifically includes:

[0058] S2011: Control the preset sensor to collect multiple operating data of the host.

[0059] Specifically, through a preset data collection script, control the preset sensor to collect multiple operating data of the host through a preset interface.

[0060] Among them, the preset interface can be an IPMI interface or other interfaces. The IPMI interface is a standard interface for managing and monitoring computer system hardware.

[0061] S2012: In response to the user's writing operation according to the preset coding language, generate a corresponding timing script.

[0062] In this embodiment, the preset coding language can be Bash or other coding languages.

[0063] In this embodiment, the timing script mainly includes a preset reading function, a preset tool, a preset time period, and other parameters.

[0064] Among them, the preset reading function can be the read_sensor function or other functions.

[0065] Among them, the preset tool can be the ipmitool tool or other tools.

[0066] Among them, the preset time period can be half a minute, one minute, or other time periods.

[0067] S2013: Read multiple running data of the host from the preset sensor according to the timing script.

[0068] Exemplarily, according to the read_sensor function, use the ipmitool tool to read the data of the temperature sensor and the current sensor respectively at one-minute intervals; print the read data to the console to obtain the corresponding temperature data and current data.

[0069] S202: Perform data processing on the multiple running data to obtain the processed running data.

[0070] Specifically, perform cleaning and normalization processing on the multiple running data to obtain the processed running data.

[0071] S203: Input one or more processed running data into the pre-generated target clustering model for load analysis to obtain a load analysis result.

[0072] Exemplarily, input the processed CPU utilization rate into the pre-generated target clustering model through the controller for load analysis to obtain a load analysis result of a light load result or a medium-high load result.

[0073] Exemplarily, the light load result is load_level < 0.5; the medium-high load result is load_level ≥ 0.5.

[0074] S204: Adjust the load of one or more components according to the load analysis result to obtain one or more adjusted operation data.

[0075] In this embodiment, the load analysis result is a light load result or a medium-high load result; correspondingly, step S204 specifically includes:

[0076] S2041: Perform load reduction adjustment on the operating states of one or more components according to the light load result to obtain one or more adjusted operation data; or,

[0077] Specifically, control the controller to perform load reduction adjustment on the operating states of one or more components by means of a preset adjustment function and call a first preset instruction according to the light load result to obtain one or more adjusted operation data.

[0078] Among them, the preset adjustment function is the apply_power_policy function.

[0079] Among them, the first preset instruction is the cpufreq-set -f low instruction.

[0080] Exemplarily, by means of the apply_power_policy function and load_level < 0.5, call the cpufreq-set -f low instruction to reduce the frequency of the CPU and reduce the number of memory channels of the memory bandwidth to obtain the adjusted CPU utilization rate and the adjusted memory bandwidth.

[0081] S2042: Perform load increase adjustment on the operating states of one or more components according to the medium-high load result to obtain one or more adjusted operation data.

[0082] Specifically, control the controller to perform load increase adjustment on the operating states of one or more components by means of a preset adjustment function and call a second preset instruction according to the medium-high load result to obtain one or more adjusted operation data.

[0083] Among them, the second preset instruction is the cpufreq-set -f high instruction.

[0084] Exemplarily, by means of the apply_power_policy function and load_level ≥ 0.5, call the cpufreq-set -f high instruction to increase the frequency of the CPU and increase the number of memory channels of the memory bandwidth to obtain the adjusted CPU utilization rate and the adjusted memory bandwidth.

[0085] In addition, the above adjustment can be performed once per minute.

[0086] S205: Output one or more adjusted operation data to a display terminal.

[0087] Exemplarily, output the adjusted CPU utilization rate and the adjusted memory bandwidth to the display terminal.

[0088] In this embodiment, the display terminal is a Web interface created according to the Web Server.

[0089] Among them, the specific creation process of the Web interface is: Develop an intuitive Web interface according to the Web Server using a front-end framework. The Web interface is used to view the real-time status, historical data, and current power consumption management strategy of the system.

[0090] Among them, the front-end framework can be React, Vue.js, or other frameworks.

[0091] In addition, an API interface can also be set on the display terminal to facilitate external system calls.

[0092] In this embodiment, on the basis of Figure 1 , the specific process of the adjustment processing of the host load can be found in the scenario illustration of the method for adjusting the host load Figure 2 , such as Figure 6 shown.

[0093] In summary, the method for adjusting the host load provided in this embodiment reads multiple operation data of the host collected by a preset sensor; performs data processing on the multiple operation data to obtain each processed operation data; inputs one or more processed operation data into a pre-generated target clustering model for load analysis to obtain a load analysis result; performs load adjustment on one or more corresponding components according to the load analysis result to obtain one or more adjusted operation data; outputs one or more adjusted operation data to a display terminal. This application abandons the fixed and single energy consumption upper limit strategy. Only by obtaining the load analysis result can it automatically perform load adjustment on the operation states of one or more components, which not only ensures the stable operation of the system but also makes the host energy consumption adjustment more flexible.

[0094] In addition, the method for adjusting the host load provided in this embodiment collects multiple operation data of the host through a preset sensor, ensuring an accurate perception of the host operation state.

[0095] The method for adjusting the host load provided in this embodiment performs load adjustment on one or more components according to the load analysis result to obtain one or more adjusted operation data, ensuring that the host reduces energy consumption at low load and meets performance requirements at high load.

[0096] The host load adjustment processing method provided in this embodiment enables the system to be integrated into the existing BMC framework by setting up an API interface, and supports docking with other management tools or systems, enhancing the flexibility and scalability of the system.

[0097] Figure 3 Schematic flow of the host load adjustment processing method provided in the embodiments of this application Figure 2 . In the embodiments of this application, based on the embodiments provided in Figure 2 , a detailed description of the generation process of the target clustering model in step S203 is given. As Figure 3 shown, the method includes:

[0098] S301: Read multiple historical operation data of the host collected by preset sensors.

[0099] In this embodiment, the discussion about the preset sensors and operation data has been given in detail in step S201, and will not be elaborated here.

[0100] S302: Store the multiple historical operation data as a fixed file in a preset database.

[0101] In this embodiment, the preset database is PostgreSQL, InfluxDB or other databases.

[0102] S303: Configure logging for the fixed file to obtain a corresponding log file.

[0103] Specifically, step S303 specifically includes:

[0104] S3031: Configure logging for the fixed file through a preset logger to obtain a log record name and an initial record level.

[0105] In this embodiment, the log record name can be sensor_data.log or other names.

[0106] In this embodiment, the initial record level is INFO.

[0107] S3032: Record the initial record level into the log file through a preset function.

[0108] In this embodiment, the preset function is the log_sensor_data function.

[0109] Specifically, at preset time intervals, the log_sensor_data function receives a piece of data and records the data as an INFO log message into the log file.

[0110] Among them, the preset time interval can be half a minute, one minute or other time intervals.

[0111] S3033: Perform conversion processing on multiple historical operation data in a fixed file to obtain multiple strings.

[0112] S3034: Record multiple strings into a log file through a preset function.

[0113] Exemplarily, record multiple strings into a log file through the log_sensor_data function.

[0114] S304: Retrieve one or more historical operation data from the log file.

[0115] Specifically, retrieve one or more historical operation data from the log file and store the one or more historical operation data as a CSV file.

[0116] S305: Perform data processing on one or more historical operation data to obtain one or more processed historical operation data.

[0117] Specifically, step S305 specifically includes:

[0118] S3051: Perform data cleaning on one or more historical operation data to obtain one or more cleaned historical operation data.

[0119] In this embodiment, data cleaning is used to check whether there are missing values or outliers in the data and perform corresponding processing. For example, the forward filling method can be used to fill in the missing values, or the records containing outliers can be deleted.

[0120] S3052: Perform data feature selection on one or more cleaned historical operation data to obtain one or more selected historical operation data.

[0121] In this embodiment, feature selection is used to select features related to load adjustment from all available sensor data. For example, features such as GPU usage rate and memory bandwidth.

[0122] S3053: Perform normalization processing on one or more selected historical operation data to obtain one or more processed historical operation data.

[0123] In this embodiment, common normalization methods in normalization processing include: Min-Max normalization and Z-Score normalization. Through normalization processing, each feature can be within the same scale range, thereby improving the performance of the clustering model.

[0124] S306: Divide one or more processed historical operation data to obtain a training set and a test set.

[0125] In this embodiment, the ratio of the training set to the test set can be any ratio among 9:1, 8:2 or 7:3, or other ratios.

[0126] S307: Train a preset clustering model according to the training set to obtain a trained clustering model.

[0127] Specifically, step S307 specifically includes:

[0128] S3071: Set one or more parameters of the preset clustering model to initialize the preset clustering model, where any parameter includes multiple working modes.

[0129] In this embodiment, the preset clustering model is a K-Means model, where K-Means is a commonly used unsupervised learning algorithm, suitable for discovering natural groupings or clusters in data.

[0130] In this embodiment, the parameters can be the number of clusters, random seeds or other parameters.

[0131] Exemplarily, set the number-of-clusters parameter of the K-Means model to initialize the K-Means model, where the number-of-clusters parameter includes three working modes.

[0132] S3072: After the preset clustering model is initialized, perform data fitting on the preset clustering model of each working mode according to the training set to obtain each cluster center.

[0133] In addition, a cluster label needs to be assigned to each working mode.

[0134] S3073: Determine the preset clustering model of the target working mode according to each cluster center.

[0135] Specifically, determine the preset clustering model of the target working mode according to each cluster center and the corresponding cluster label.

[0136] S3074: Determine the preset clustering model of the target working mode as the trained preset clustering model.

[0137] S308: Test the trained clustering model according to the test set to complete the test of the trained clustering model.

[0138] Specifically, test the trained clustering model according to the test set to obtain the cluster label of each sample to complete the test of the trained clustering model.

[0139] S309: After testing the trained clustering model, evaluate the trained clustering model to complete the evaluation of the trained clustering model.

[0140] Specifically, evaluate the silhouette coefficient or other clustering metrics of the trained clustering model to measure the clustering effect of the trained clustering model and complete the evaluation of the trained clustering model.

[0141] S310: After completing the evaluation of the trained clustering model, adjust the parameters of the trained clustering model to determine the target clustering model.

[0142] Specifically, step S310 specifically includes:

[0143] S3101: Obtain various parameters.

[0144] Exemplarily, obtain different numbers of clusters and different random seeds.

[0145] S3102: Combine various parameters in different ways to obtain parameter groups.

[0146] Exemplarily, combine each number of clusters and each random seed in different ways to obtain parameter groups.

[0147] S3103: Perform cross-validation on the trained clustering model according to each parameter group to obtain the target parameter group.

[0148] In this embodiment, cross-validation is used to evaluate the performance of the trained clustering model under each parameter group to select the target parameter group that meets the requirements.

[0149] S3104: Adjust the parameters of the trained clustering model according to the target parameter group to determine the target clustering model.

[0150] In addition, save the target clustering model to a model file for subsequent use. It is also necessary to record the process of training, testing, evaluating, and parameter adjustment of the preset clustering model for subsequent analysis and reuse.

[0151] In summary, the method for adjusting and processing the host load provided in this embodiment reads multiple historical operation data of the host collected by a preset sensor; processes one or more historical operation data to obtain one or more processed historical operation data; divides one or more processed historical operation data to obtain a training set and a test set; trains a preset clustering model according to the training set to obtain a trained clustering model; tests the trained clustering model according to the test set to complete the testing of the trained clustering model; after completing the testing of the trained clustering model, evaluates the trained clustering model to complete the evaluation of the trained clustering model; after completing the evaluation of the trained clustering model, adjusts the parameters of the trained clustering model to determine a target clustering model, which can make the target clustering model more in line with actual requirements, make the load analysis more accurate, and lay a foundation for subsequent host energy consumption adjustment.

[0152] In addition, the method for adjusting and processing the host load provided in this embodiment persists the historical operation data by storing multiple historical operation data as fixed files in a preset database.

[0153] In addition, the method for adjusting and processing the host load provided in this embodiment configures log records for the fixed files to obtain corresponding log files, which is convenient for querying the corresponding historical operation data by date.

[0154] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method.

[0155] Figure 4 This is a schematic structural diagram of the device for adjusting and processing the host load provided in the embodiment of the present application. As Figure 4 shown, the embodiment of the present application also provides a device for adjusting and processing the host load, including: a first reading module 401, a first processing module 402, an analysis module 403, a first adjustment module 404, and an output module 405.

[0156] The first reading module 401 is used to read multiple operation data of the host collected by a preset sensor, where the host includes multiple components, and each component corresponds to one operation data;

[0157] The first processing module 402 is used to process the multiple operation data to obtain each processed operation data;

[0158] The analysis module 403 is used to input one or more processed operation data into a pre-generated target clustering model for load analysis to obtain a load analysis result;

[0159] The first adjustment module 404 is configured to perform load adjustment on one or more components according to the load analysis result to obtain one or more adjusted operation data;

[0160] The output module 405 is configured to output the one or more adjusted operation data to a display terminal.

[0161] In a possible implementation, the reading module 401 specifically includes:

[0162] A control unit, configured to control the preset sensor to collect multiple operation data of the host;

[0163] A generating unit, configured to generate a corresponding timing script in response to a writing operation of a user according to a preset coding language;

[0164] A reading unit, configured to read the multiple operation data of the host from the preset sensor according to the timing script.

[0165] In a possible implementation, the device further includes:

[0166] A second reading module, configured to read multiple historical operation data of the host collected by a preset sensor;

[0167] A storage module, configured to store the multiple historical operation data as a fixed file in a preset database;

[0168] A configuration module, configured to perform log record configuration on the fixed file to obtain a corresponding log file;

[0169] An extraction module, configured to extract one or more historical operation data from the log file;

[0170] A second processing module, configured to perform data processing on the one or more historical operation data to obtain one or more processed historical operation data;

[0171] A partitioning module, configured to partition the one or more processed historical operation data to obtain a training set and a test set;

[0172] A training module, configured to train a preset clustering model according to the training set to obtain a trained clustering model;

[0173] A testing module, configured to test the trained clustering model according to the test set to complete the testing of the trained clustering model;

[0174] An evaluation module, configured to evaluate the trained clustering model after testing the trained clustering model, so as to complete the evaluation of the trained clustering model;

[0175] A second adjustment module, configured to adjust parameters of the trained clustering model after completing the evaluation of the trained clustering model, so as to determine a target clustering model.

[0176] In a possible implementation manner, the configuration module specifically includes:

[0177] A configuration unit, configured to perform log recording configuration on the fixed file through a preset logger, so as to obtain a log recording name and an initial recording level;

[0178] A first recording unit, configured to record the initial recording level into a log file through a preset function;

[0179] A processing unit, configured to perform conversion processing on multiple historical operation data in the fixed file, so as to obtain multiple strings;

[0180] A second recording unit, configured to record the multiple strings into a log file through the preset function.

[0181] In a possible implementation manner, the second processing module specifically includes:

[0182] A cleaning unit, configured to perform data cleaning on the one or more historical operation data, so as to obtain one or more cleaned historical operation data;

[0183] A selection unit, configured to perform data feature selection on the one or more cleaned historical operation data, so as to obtain one or more selected historical operation data;

[0184] A processing unit, configured to perform normalization processing on the one or more selected historical operation data, so as to obtain one or more processed historical operation data.

[0185] In a possible implementation manner, the training module specifically includes:

[0186] A setting unit, configured to set one or more parameters of the preset clustering model to initialize the preset clustering model, where any one of the parameters includes multiple working modes;

[0187] A fitting unit, configured to perform data fitting on the preset clustering model of each working mode according to the training set after the preset clustering model is initialized, so as to obtain each clustering center;

[0188] A first determination unit, configured to determine a preset clustering model of a target working mode according to the respective clustering centers;

[0189] A second determination unit, configured to determine the preset clustering model of the target working mode as a trained preset clustering model.

[0190] In a possible implementation manner, the load analysis result is a light load result or a medium-high load result;

[0191] Correspondingly, the first adjustment module 404 specifically includes:

[0192] A first adjustment unit, configured to perform load reduction adjustment on the operating states of one or more components according to the light load result, so as to obtain one or more adjusted operating data; or,

[0193] A second adjustment unit, configured to perform load increase adjustment on the operating states of one or more components according to the medium-high load result, so as to obtain one or more adjusted operating data.

[0194] For the description of the features in the corresponding embodiment of the adjustment processing device for the host load, reference may be made to the relevant description in the corresponding embodiment of the adjustment processing method for the host load, which will not be elaborated here one by one.

[0195] Figure 5 This is a schematic structural diagram of the server provided by this application. As Figure 5 shown, the server provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the server further includes a communication component 503. Among them, the processor 501, the memory 502, and the communication component 503 are connected through a bus.

[0196] In a specific implementation process, at least one processor 501 executes the computer execution instructions stored in the memory 502, so that at least one processor 501 executes the above-mentioned embodiment of the adjustment processing method for the host load.

[0197] For the specific implementation process of the processor 501, reference may be made to the above method embodiment, and its implementation principle and technical effects are similar, which will not be elaborated here in this embodiment.

[0198] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the application can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0199] The memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0200] The bus may be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the buses in the drawings of the present application are not limited to only one bus or one type of bus.

[0201] The embodiments of the present application also provide a computer-readable storage medium, in which a computer program is stored, and the computer program is configured to execute the steps in any of the above embodiments of the method for adjusting the host load when running.

[0202] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs, etc., all of which can store computer programs.

[0203] The embodiments of the present application also provide a computer program product, the above computer program product includes a computer program, and when the computer program is executed by a processor, the steps in any of the above embodiments of the method for adjusting the host load are implemented.

[0204] An embodiment of the present application also provides another computer program product, including a non-volatile computer-readable storage medium. The non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above-described embodiments of the host load adjustment processing method are implemented.

[0205] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0206] The above has introduced in detail a host load adjustment processing method, device, server, and storage medium provided by the present application. Specific examples are used herein to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application. It should be noted that for those of ordinary skill in the art in the technical field, without departing from the principle of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. A method for adjusting and processing the host load, characterized in that Including: Reading multiple operating data of a host collected by a preset sensor, where the host includes multiple components, and each component corresponds to an operating data respectively; Performing data processing on the multiple operating data to obtain each processed operating data; Inputting one or more processed operating data into a pre-generated target clustering model for load analysis to obtain a load analysis result; Adjusting the load of one or more components according to the load analysis result to obtain one or more adjusted operating data; Outputting the one or more adjusted operating data to a display terminal.

2. The method for adjusting and processing the host load according to claim 1, wherein The reading of the multiple operating data of the host collected by the preset sensor includes: Controlling the preset sensor to collect the multiple operating data of the host; Responding to a user's writing operation according to a preset coding language to generate a corresponding timing script; Reading the multiple operating data of the host from the preset sensor according to the timing script.

3. The method for adjusting and processing the host load according to claim 1, wherein, The generation process of the target clustering model includes: Reading multiple historical operating data of a host collected by a preset sensor; Storing the multiple historical operating data as a fixed file in a preset database; Performing log record configuration on the fixed file to obtain a corresponding log file; Retrieving one or more historical operating data from the log file; Performing data processing on the one or more historical operating data to obtain one or more processed historical operating data; Dividing the one or more processed historical operating data to obtain a training set and a test set; Training a preset clustering model according to the training set to obtain a trained clustering model; Testing the trained clustering model according to the test set to complete the testing of the trained clustering model; After completing the testing of the trained clustering model, evaluating the trained clustering model to complete the evaluation of the trained clustering model; After completing the evaluation of the trained clustering model, adjusting the parameters of the trained clustering model to determine a target clustering model.

4. The method for adjusting and processing the host load according to claim 3, wherein The performing log record configuration on the fixed file to obtain a corresponding log file includes: Performing log record configuration on the fixed file through a preset logger to obtain a log record name and an initial record level; Recording the initial record level into the log file through a preset function; Performing conversion processing on the multiple historical operating data in the fixed file to obtain multiple strings; Recording the multiple strings into the log file through the preset function.

5. The method for adjusting and processing the host load according to claim 3, wherein The performing data processing on the one or more historical operating data to obtain one or more processed historical operating data includes: Performing data cleaning on the one or more historical operating data to obtain one or more cleaned historical operating data; Performing data feature selection on the one or more cleaned historical operating data to obtain one or more selected historical operating data; Performing normalization processing on the one or more selected historical operating data to obtain one or more processed historical operating data.

6. The method for adjusting and processing the host load according to claim 3, wherein Training the preset clustering model according to the training set to obtain a trained clustering model includes: Setting one or more parameters of the preset clustering model to initialize the preset clustering model, where any parameter includes multiple working modes; After the preset clustering model is initialized, fitting the data of the preset clustering model of each working mode according to the training set to obtain each clustering center; Determining the preset clustering model of the target working mode according to each clustering center; Determining the preset clustering model of the target working mode as the trained preset clustering model.

7. The method for adjusting and processing the host load according to any one of claims 1 to 6, characterized in that Where the load analysis result is a light load result or a medium-high load result; Correspondingly, adjusting the load of one or more components according to the load analysis result to obtain one or more adjusted operation data, including: Reducing the load of the operating state of one or more components according to the light load result to obtain one or more adjusted operation data; or, Increasing the load of the operating state of one or more components according to the medium-high load result to obtain one or more adjusted operation data.

8. An adjustment processing device for host load, characterized in that Includes: A first reading module for reading a plurality of operation data of a host collected by a preset sensor, where the host includes a plurality of components, and each component corresponds to an operation data respectively; A first processing module for processing the plurality of operation data to obtain each processed operation data; An analysis module for inputting one or more processed operation data into a pre-generated target clustering model for load analysis to obtain a load analysis result; A first adjustment module for adjusting the load of one or more components according to the load analysis result to obtain one or more adjusted operation data; An output module for outputting the one or more adjusted operation data to a display terminal.

9. A server, characterized in that, Includes: A memory for storing a computer program; A processor for implementing the steps of the method for adjusting the host load according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, where the computer program, when executed by a processor, implements the steps of the method for adjusting the host load according to any one of claims 1 to 7.

Images