Cloud platform baseline performance analysis method and device, equipment and storage medium

Abstract

The invention discloses a cloud platform baseline performance analysis method, device and equipment and a storage medium, and relates to the technical field of cloud platform performance analysis, and the method comprises the steps: constructing a container mirror image, and loading a detection model library through a corresponding container instance; creating an initial virtual machine object, loading model data of the hybrid heterogeneous cloud resource model to the initial virtual machine object to obtain target virtual machine objects, and obtaining a reference performance comprehensive value corresponding to each target virtual machine object by using the detection model library; and obtaining target reference performance comprehensive value matrix data based on each reference performance comprehensive value, generating a first target curve according to the historical reference performance comprehensive value matrix data, generating a second target curve by using the target reference performance comprehensive value matrix data, and analyzing the baseline performance of the cloud platform according to the first target curve and the second target curve. The reference performance comprehensive value of the cloud platform under different conditions is obtained by using a container technology, and the baseline performance of the cloud platform can be evaluated by analyzing the change condition of the reference performance comprehensive value.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud platform performance analysis, and particularly to a method, device, equipment and storage medium for analyzing the baseline performance of a cloud platform. Background Art

[0002] A hybrid heterogeneous cloud platform is a comprehensive cloud computing platform that combines multiple cloud service models (such as public cloud, private cloud, edge cloud, etc.) and different technical architectures (such as virtualization, containerization, bare metal, etc.). It can integrate and manage cloud resources from different vendors and different technology stacks, and provide unified resource scheduling, management and service delivery capabilities.

[0003] Currently, in cloud platform products with different architectures, the computing resources, disk resources, and network resource capabilities that provide basic services are different, and their baseline performance performances are also different. It is impossible to effectively measure and analyze the baseline performance of a hybrid heterogeneous cloud platform, nor is it possible to specifically select improvement and optimization methods. Therefore, how to analyze the baseline performance of a hybrid heterogeneous cloud platform has become a technical problem to be solved at present. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method, device, equipment and storage medium for analyzing the baseline performance of a cloud platform, which can obtain the comprehensive value of the benchmark performance of the cloud platform under different conditions by using container technology, and can evaluate the baseline performance of the cloud platform by analyzing the change of the comprehensive value of the benchmark performance. The specific solutions are as follows:

[0005] In a first aspect, the present application provides a method for analyzing the baseline performance of a cloud platform, including:

[0006] Configure a hybrid heterogeneous cloud resource model corresponding to the target cloud platform, obtain a preset script file and a preset base image, and use the preset script file and the preset base image to construct a container image corresponding to the hybrid heterogeneous cloud resource model;

[0007] Upload the container image to the target cloud platform to start a container instance corresponding to the container image, and use a preset analysis control program in the container instance to load a preset detection model library;

[0008] Create an initial virtual machine object according to a target step value corresponding to the hybrid heterogeneous cloud resource model, load the model data of the hybrid heterogeneous cloud resource model into each initial virtual machine object to obtain each target virtual machine object, and use the preset detection model library to obtain a comprehensive value of the benchmark performance corresponding to each target virtual machine object;

[0009] Obtain the target benchmark performance composite value matrix data corresponding to the target cloud platform based on each of the said benchmark performance composite values, generate a first target curve according to the historical benchmark performance composite value matrix data, and generate a second target curve using the target benchmark performance composite value matrix data. Obtain the offset situation of the second target curve according to the first target curve, and analyze the baseline performance of the target cloud platform according to the offset situation.

[0010] Optionally, configuring the hybrid heterogeneous cloud resource model corresponding to the target cloud platform includes:

[0011] Create an initial cloud resource model, configure the model architecture type of the initial cloud resource model, and configure the number of infrastructures corresponding to each model architecture type and the maximum available cloud resource computing scale corresponding to the initial cloud resource model to obtain the initially configured model corresponding to the initial cloud resource model;

[0012] Determine the disk specifications of the cloud disk to be detected and the network type of the network to be detected from the initially configured model, configure the test framework system corresponding to the initially configured model, and set the linearly increasing step value corresponding to the initial cloud resource model to obtain the hybrid heterogeneous cloud resource model corresponding to the target cloud platform; wherein, the test framework system includes test tools, a baseline performance detection model, test scenarios, and test metrics.

[0013] Optionally, using the preset script file and the preset base image to build the container image corresponding to the hybrid heterogeneous cloud resource model includes:

[0014] Download the preset base image from a preset server and download the preset script file from a preset Git repository;

[0015] Use the preset script file to load the hybrid heterogeneous cloud resource model into the preset base image to build an initial image, and tag the initial image to obtain the container image corresponding to the hybrid heterogeneous cloud resource model; wherein, the tag includes the version information of the container image and the environment information matching the container image, and the environment information includes development environment information, production environment information, and test environment information.

[0016] Optionally, uploading the container image to the target cloud platform to start the container instance corresponding to the container image, and using the preset analysis control program in the container instance to load the preset detection model library includes:

[0017] Upload the container image to the target cloud platform, instantiate the container image in the target cloud platform to obtain a container instance corresponding to the container image, and run the preset analysis control program in the container instance to start the container instance;

[0018] Use the preset analysis control program to load the test framework system to complete the loading of the preset detection model library.

[0019] Optionally, creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, loading the model data of the hybrid heterogeneous cloud resource model into each initial virtual machine object to obtain each target virtual machine object, and using the preset detection model library to obtain the corresponding benchmark performance comprehensive value of each target virtual machine object, including:

[0020] Load the maximum available cloud resource computing scale corresponding to the hybrid heterogeneous cloud resource model;

[0021] Create an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and load the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object;

[0022] Perform a benchmark performance test on the current target virtual machine object to obtain the corresponding benchmark performance comprehensive value of the current target virtual machine object, and format and save the corresponding benchmark performance comprehensive value of the current target virtual machine object;

[0023] Judge whether the current computing scale is less than the maximum available cloud resource computing scale. If the current computing scale is less than the maximum available cloud resource computing scale, jump to the step of creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and loading the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object, so as to obtain the corresponding benchmark performance comprehensive value of the new target virtual machine object;

[0024] If the current computing scale is not less than the maximum available cloud resource computing scale, jump to the step of obtaining the target benchmark performance comprehensive value matrix data corresponding to the target cloud platform based on each benchmark performance comprehensive value.

[0025] Optionally, generating a first target curve according to the historical benchmark performance comprehensive value matrix data, generating a second target curve using the target benchmark performance comprehensive value matrix data, and obtaining the offset situation of the second target curve according to the first target curve, including:

[0026] Obtain the historical reference performance comprehensive value matrix data, generate the first target curve according to the historical reference performance comprehensive value matrix data, load the target reference performance comprehensive value matrix data, and generate the second target curve according to the target reference performance comprehensive value matrix data;

[0027] Calculate the positive deviation data and negative deviation data of the second target curve based on the first target curve, and save the positive deviation data and the negative deviation data.

[0028] Optionally, the analysis of the baseline performance of the target cloud platform according to the deviation situation includes:

[0029] Load the positive deviation data and the negative deviation data, and obtain the first ratio of the positive deviation data and the second ratio of the negative deviation data;

[0030] Judge the baseline performance of the target cloud platform according to the first ratio and the second ratio, and generate an evaluation conclusion on the baseline performance of the target cloud platform.

[0031] In a second aspect, the present application provides a cloud platform baseline performance analysis device, including:

[0032] An image building module, configured to configure a hybrid heterogeneous cloud resource model corresponding to the target cloud platform, obtain a preset script file and a preset base image, and build a container image corresponding to the hybrid heterogeneous cloud resource model by using the preset script file and the preset base image;

[0033] A model library loading module, configured to upload the container image to the target cloud platform to start a container instance corresponding to the container image, and load a preset detection model library by using a preset analysis control program in the container instance;

[0034] A virtual machine object creation model, configured to create an initial virtual machine object according to a target step value corresponding to the hybrid heterogeneous cloud resource model, load model data of the hybrid heterogeneous cloud resource model into each initial virtual machine object to obtain each target virtual machine object, and obtain a reference performance comprehensive value corresponding to each target virtual machine object by using the preset detection model library;

[0035] A baseline performance analysis module, configured to obtain target reference performance comprehensive value matrix data corresponding to the target cloud platform based on each reference performance comprehensive value, generate a first target curve according to historical reference performance comprehensive value matrix data, generate a second target curve by using the target reference performance comprehensive value matrix data, obtain the deviation situation of the second target curve according to the first target curve, and analyze the baseline performance of the target cloud platform according to the deviation situation.

[0036] In a third aspect, the present application provides an electronic device, including:

[0037] a memory for storing a computer program;

[0038] a processor for executing the computer program to implement the foregoing cloud platform baseline performance analysis method.

[0039] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program, and when the computer program is executed by a processor, the foregoing cloud platform baseline performance analysis method is implemented.

[0040] In the present application, first, a hybrid heterogeneous cloud resource model corresponding to a target cloud platform is configured, a preset script file and a preset base image are obtained, and a container image corresponding to the hybrid heterogeneous cloud resource model is constructed by using the preset script file and the preset base image. Then, the container image is uploaded to the target cloud platform to start a container instance corresponding to the container image, and a preset analysis control program in the container instance is used to load a preset detection model library. After that, an initial virtual machine object is created according to a target step value corresponding to the hybrid heterogeneous cloud resource model, and model data of the hybrid heterogeneous cloud resource model is loaded into each initial virtual machine object to obtain each target virtual machine object. Then, a benchmark performance comprehensive value corresponding to each target virtual machine object is obtained by using the preset detection model library. Finally, target benchmark performance comprehensive value matrix data corresponding to the target cloud platform is obtained based on each benchmark performance comprehensive value, a first target curve is generated according to historical benchmark performance comprehensive value matrix data, and a second target curve is generated by using the target benchmark performance comprehensive value matrix data. The offset situation of the second target curve is obtained according to the first target curve, and the baseline performance of the target cloud platform is analyzed according to the offset situation. Thus, it can be seen that in the present application, a container image corresponding to a hybrid heterogeneous cloud resource model is constructed through container technology, and a container can be used as a carrier of a virtual machine resource definition and testing tool, providing support for subsequent creation of detection resource virtual machine objects; by loading data of the hybrid heterogeneous cloud resource model into virtual machine objects, and obtaining the benchmark performance comprehensive value of the cloud platform under different conditions, by constructing curves using different benchmark performance comprehensive values and analyzing the change situation of the comprehensive values using the curves, the baseline performance of the hybrid heterogeneous cloud platform can be analyzed, and the performance compliance situation of the cloud platform can be evaluated. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0042] Figure 1 Flow chart of a cloud platform baseline performance analysis method disclosed in this application;

[0043] Figure 2 Schematic diagram of the process of a specific cloud platform baseline performance analysis method disclosed in this application;

[0044] Figure 3 Schematic diagram of the structure of a cloud platform baseline performance analysis device disclosed in this application;

[0045] Figure 4 Schematic diagram of the structure of an electronic device disclosed in this application. Specific implementation manners

[0046] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0047] Currently, in cloud platform products with different architectures, the computing resources, disk resources, and network resource capabilities that provide basic services are different, and their baseline performance performances are also different, making it impossible to effectively measure and analyze the baseline performance of a hybrid heterogeneous cloud platform. For this reason, this application provides a cloud platform baseline performance analysis method, which can obtain the comprehensive benchmark performance value of the cloud platform under different conditions by using container technology, and can evaluate the baseline performance of the cloud platform by analyzing the change of the comprehensive benchmark performance value.

[0048] See Figure 1 As shown, an embodiment of the present invention discloses a cloud platform baseline performance analysis method, including:

[0049] Step S11, configure a hybrid heterogeneous cloud resource model corresponding to the target cloud platform, obtain a preset script file and a preset base image, and use the preset script file and the preset base image to construct a container image corresponding to the hybrid heterogeneous cloud resource model.

[0050] It should be noted that the container technology in this embodiment refers to using a container as the carrier of a virtual machine resource definition and testing tool, providing support for creating a detected resource virtual machine object in the following; the hybrid heterogeneous cloud platform in this embodiment specifically refers to a cloud environment that can provide services for an Arm architecture computing resource pool and an X86 architecture computing resource pool. The process of performing baseline performance analysis on the cloud platform in this embodiment is specifically as Figure 2As shown in the figure, first, it is necessary to define the hybrid heterogeneous cloud resources, purchase the container image according to the hybrid heterogeneous cloud resources, then use the program in the container image to load the detection model library, then create a virtual machine object for detection resources, and finally perform a baseline performance analysis on the virtual machine object to obtain the baseline performance of the cloud platform.

[0051] In this embodiment, the target cloud platform is a hybrid heterogeneous cloud platform, and its corresponding hybrid heterogeneous cloud resources include information such as hybrid architecture types, the number of hybrid architecture infrastructures, the maximum available cloud resource computing scale, cloud disk specifications, network types, baseline performance detection models, and linear increment step values; correspondingly, the process of configuring the hybrid heterogeneous cloud resource model corresponding to the above-mentioned target cloud platform may specifically include: creating an initial cloud resource model, configuring the model architecture type of the initial cloud resource model, and configuring the number of infrastructures corresponding to each model architecture type and the maximum available cloud resource computing scale corresponding to the initial cloud resource model to obtain the initial configuration model corresponding to the initial cloud resource model; determining the cloud disk specifications of the cloud disk to be detected and the network type of the network to be detected from the initial configuration model, configuring the test framework system corresponding to the initial configuration model, and setting the linear increment step value corresponding to the initial cloud resource model to obtain the hybrid heterogeneous cloud resource model corresponding to the target cloud platform; among them, the test framework system includes test tools, baseline performance detection models, test scenarios, and test indicators. The specific process of configuring the hybrid heterogeneous cloud resource model is as follows:

[0052] (1). Create a new hybrid heterogeneous cloud resource model object;

[0053] (2). Configure the hybrid architecture types (i.e., the above-mentioned model architecture types, including X86, Arm, MIPS (Microprocessor without Interlocked Pipelined Stages, that is, a processor architecture), Alpha, etc.), multiple combinations;

[0054] (3). Configure the number of infrastructures for each architecture;

[0055] (4). Set the maximum available cloud resource computing scale;

[0056] (5). Select the cloud disk specifications (SSD (Solid State Drive, that is, solid state drive) resource pool, SATA (Serial Advanced Technology Attachment, that is, a serial hard disk) resource pool, SAS (Serial Attached SCSI, that is, a serial hard disk) resource pool) participating in the performance detection;

[0057] (6). Select the network type participating in the performance detection;

[0058] (7) Configure a baseline performance detection model, test tools, test scenarios, test metrics, etc.;

[0059] (8) Set a linearly increasing step value;

[0060] (9) Save and name the current model object.

[0061] In this embodiment, the process of constructing a container image corresponding to the hybrid heterogeneous cloud resource model by using a preset script file and a preset base image may specifically include: downloading the preset base image from a preset server and downloading the preset script file from a preset Git repository; using the preset script file to load the hybrid heterogeneous cloud resource model into the preset base image to construct an initial image, and tagging the initial image to obtain the container image corresponding to the hybrid heterogeneous cloud resource model; where the tag includes the version information of the container image and the environment information matching the container image, and the environment information includes development environment information, production environment information, and test environment information. It can be understood that constructing a resource container image for baseline performance analysis (i.e., the above container image) includes downloading the system base image (i.e., the above initial image), downloading the build script Dockfile (i.e., the above preset script file), and performing the build operation; the specific implementation process of constructing the resource container image for baseline performance analysis is as follows:

[0062] (1) Access the build server (i.e., the above preset server) and download the system base image;

[0063] (2) Access the Git repository and download the Dockfile file for the build script of the resource container image for baseline performance analysis;

[0064] (3) Perform the build, and the build process is as follows:

[0065] a. Download a performance test tool set, including CPU (Central Processing Unit, i.e., the central processing unit), memory, operating system comprehensive, disk, and network;

[0066] b. Load the current hybrid heterogeneous cloud resource model object;

[0067] c. According to the configured combination of hybrid architecture types, parallel-compile the test tools, control programs, and test scripts corresponding to the combined architecture;

[0068] d. Save the built container image and tag it;

[0069] e. Push the local container image to the remote image repository center.

[0070] In this embodiment, by using a container as the carrier of the virtual machine resource definition and testing tool, it provides support for creating a virtual machine object for detecting resources subsequently, thus laying a foundation for detecting the baseline performance of the hybrid heterogeneous cloud platform.

[0071] Step S12: Upload the container image to the target cloud platform to start the container instance corresponding to the container image, and use the preset analysis control program in the container instance to load the preset detection model library.

[0072] In this embodiment, the process of uploading the container image to the target cloud platform to start the container instance corresponding to the container image and using the preset analysis control program in the container instance to load the preset detection model library may specifically include: uploading the container image to the target cloud platform, instantiating the container image in the target cloud platform to obtain the container instance corresponding to the container image, and running the preset analysis control program in the container instance to start the container instance; using the preset analysis control program to load the test framework system to complete the loading of the preset detection model library; specifically, the above process of uploading and starting the container instance is specifically: uploading the performance analysis resource container image to the target cloud platform; executing to generate a performance analysis resource container instance; running the built-in baseline performance analysis control program; in addition, it can be understood that the process of the performance analysis control program loading the baseline performance detection model library includes information such as test tools, test scenario use cases, test metrics, etc. Specifically, the process of loading the above preset detection model library is as follows:

[0073] (1). Load the baseline performance detection model data;

[0074] (2). Obtain the test tool set (Unixbench, SysBench, stream, fio, Iperf, etc.);

[0075] (3). Obtain the test scenario use cases;

[0076] (4). Obtain the test metric standard data.

[0077] In this implementation, by loading the detection model library, it lays a foundation for the subsequent process of performing benchmark performance detection on the virtual machine object.

[0078] Step S13: Create an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, load the model data of the hybrid heterogeneous cloud resource model into each of the initial virtual machine objects to obtain each target virtual machine object, and use the preset detection model library to obtain the comprehensive baseline performance value corresponding to each of the target virtual machine objects.

[0079] In this embodiment, the process of creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, loading the model data of the hybrid heterogeneous cloud resource model into each initial virtual machine object to obtain each target virtual machine object, and using a preset detection model library to obtain the baseline performance comprehensive value corresponding to each target virtual machine object may specifically include: loading the maximum available cloud resource computing scale corresponding to the hybrid heterogeneous cloud resource model; creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and loading the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object; performing a baseline performance test on the current target virtual machine object to obtain the baseline performance comprehensive value corresponding to the current target virtual machine object, and formatting and saving the baseline performance comprehensive value corresponding to the current target virtual machine object; determining whether the current computing scale is less than the maximum available cloud resource computing scale. If the current computing scale is less than the maximum available cloud resource computing scale, then jump to the step of creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and loading the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object, so as to obtain the baseline performance comprehensive value corresponding to the new target virtual machine object; if the current computing scale is not less than the maximum available cloud resource computing scale, then jump to the step of obtaining the target baseline performance comprehensive value matrix data corresponding to the target cloud platform based on each baseline performance comprehensive value. It can be understood that the performance analysis control program calculates the number of virtual machine objects that need to be created according to the current linearly increasing step number, executes the operation of creating a detection resource virtual machine object, and loads information such as baseline performance detection model data into the newly created resource virtual machine object. The above process of creating a virtual machine object is specifically as follows: calculating the required number of computing resource virtual machine objects (i.e., initial virtual machine objects); batch-creating detection resource virtual machine objects; and loading the baseline performance detection model data into the virtual machine objects to obtain target virtual machine objects.

[0080] After the target virtual machine object is successfully created, it is also necessary to perform a baseline performance test operation on the virtual machine object, including loading a virtual cloud disk, loading a virtual network, reading detection model data, executing a baseline performance test script, obtaining baseline performance test result data, comparing the value test result with the test index standard value in sequence, and calculating the baseline performance comprehensive value; correspondingly, the process of performing a baseline performance (i.e., baseline performance comprehensive value) test on the virtual machine object is specifically as follows:

[0081] (1) Execute the baseline performance test operation;

[0082] (2) Load the virtual cloud disk;

[0083] (3) Load the virtual network;

[0084] (4) Read the detection model data;

[0085] (5), Execute the benchmark performance test script;

[0086] (6), Obtain the benchmark performance test result data;

[0087] (7), Compare the test results with the test index standard values in turn, and record the qualified items and unqualified items;

[0088] (8), Calculate the comprehensive benchmark performance value. Record 1 for qualified test indicators and 0 for unqualified ones. Each test indicator has a different weighting value. According to the test indicator weighting value list, find the weighting value of each test indicator, perform weighted calculation, accumulate the weighted calculation results of all test indicators, and calculate the current comprehensive benchmark performance value according to a certain proportion.

[0089] (9), Save the current comprehensive benchmark test performance value;

[0090] (10), End the current virtual machine object for detecting resources.

[0091] In this embodiment, by performing weighted calculations on various indicators, the influence of important test indicators can be highlighted, making the data of the calculated comprehensive benchmark performance value more accurate.

[0092] Step S14, Obtain the target benchmark performance comprehensive value matrix data corresponding to the target cloud platform based on each of the comprehensive benchmark performance values, generate a first target curve according to the historical benchmark performance comprehensive value matrix data, generate a second target curve using the target benchmark performance comprehensive value matrix data, obtain the offset situation of the second target curve according to the first target curve, and analyze the baseline performance of the target cloud platform according to the offset situation.

[0093] In this embodiment, the process of generating a first target curve according to the historical benchmark performance comprehensive value matrix data, generating a second target curve using the target benchmark performance comprehensive value matrix data, and obtaining the offset situation of the second target curve according to the first target curve may specifically include: obtaining the historical benchmark performance comprehensive value matrix data, generating a first target curve according to the historical benchmark performance comprehensive value matrix data, loading the target benchmark performance comprehensive value matrix data, and generating the second target curve according to the target benchmark performance comprehensive value matrix data; calculating the positive deviation data and negative deviation data of the second target curve based on the first target curve, and saving the positive deviation data and negative deviation data. The above process is specifically as follows:

[0094] (1), Load the comprehensive value matrix data of the resource benchmark performance;

[0095] (2), Plot the benchmark performance mean curve 1 with a linearly increasing step size change;

[0096] (3) Draw the curve 2 of the comprehensive value matrix of the measured baseline performance;

[0097] (4) Calculate the positive deviation and negative deviation of curve 2 relative to curve 1;

[0098] (5) Save the positive deviation data and negative deviation data.

[0099] Correspondingly, in this embodiment, the process of analyzing the baseline performance of the target cloud platform according to the deviation situation may specifically include: loading the positive deviation data and negative deviation data, and obtaining the first ratio of the positive deviation data and the second ratio of the negative deviation data; judging the baseline performance of the target cloud platform according to the first ratio and the second ratio, and generating an evaluation conclusion on the baseline performance of the target cloud platform. The above process is specifically as follows:

[0100] (1) Load the positive deviation data and negative deviation data;

[0101] (2) Analyze the proportion of the positive deviation data and analyze the proportion of the negative deviation data;

[0102] (3) Comprehensively judge the baseline performance of the hybrid heterogeneous cloud platform based on the proportion of the positive deviation data and the proportion of the negative deviation data;

[0103] (4) Generate an evaluation conclusion on the baseline performance of the hybrid heterogeneous cloud platform.

[0104] In this embodiment, by evaluating the comprehensive value of the baseline performance of the hybrid heterogeneous cloud platform according to the deviation situation of the curve, a method for analyzing the baseline performance of the hybrid heterogeneous cloud platform is formed, which effectively meets the performance analysis requirements of the heterogeneous cloud platform, realizes the automation of the heterogeneous cloud platform performance analysis, reduces the complexity of the hybrid heterogeneous cloud platform performance analysis operation, and improves the operation efficiency and operation stability of the heterogeneous cloud platform performance analysis.

[0105] It can be seen that in this application, by constructing a container image corresponding to the hybrid heterogeneous cloud resource model through container technology, the container can be used as the carrier of the virtual machine resource definition and testing tool, providing support for the subsequent creation of the detected resource virtual machine object; by loading the data of the hybrid heterogeneous cloud resource model into the virtual machine object and obtaining the comprehensive value of the baseline performance of the cloud platform under different conditions, by using different comprehensive values of the baseline performance to construct a curve and using the curve to analyze the change of the comprehensive value, the baseline performance of the hybrid heterogeneous cloud platform can be analyzed and the compliance of the cloud platform performance can be evaluated.

[0106] See Figure 3 As shown, an embodiment of the present invention discloses a cloud platform baseline performance analysis device, including:

[0107] The mirror building module 11 is used to configure a hybrid heterogeneous cloud resource model corresponding to a target cloud platform, obtain a preset script file and a preset base image, and build a container image corresponding to the hybrid heterogeneous cloud resource model by using the preset script file and the preset base image;

[0108] The model library loading module 12 is used to upload the container image to the target cloud platform to start a container instance corresponding to the container image, and load a preset detection model library by using a preset analysis control program in the container instance;

[0109] The virtual machine object creation model 13 is used to create an initial virtual machine object according to a target step value corresponding to the hybrid heterogeneous cloud resource model, load model data of the hybrid heterogeneous cloud resource model into each initial virtual machine object to obtain each target virtual machine object, and obtain a benchmark performance comprehensive value corresponding to each target virtual machine object by using the preset detection model library;

[0110] The baseline performance analysis module 14 is used to obtain target baseline performance comprehensive value matrix data corresponding to the target cloud platform based on each benchmark performance comprehensive value, generate a first target curve according to historical baseline performance comprehensive value matrix data, generate a second target curve by using the target baseline performance comprehensive value matrix data, obtain the offset situation of the second target curve according to the first target curve, and analyze the baseline performance of the target cloud platform according to the offset situation.

[0111] It can be seen that the present application builds a container image corresponding to a hybrid heterogeneous cloud resource model through container technology, can use the container as a carrier of a virtual machine resource definition and testing tool, and provides support for subsequent creation of a detection resource virtual machine object; by loading the data of the hybrid heterogeneous cloud resource model into the virtual machine object, obtaining the benchmark performance comprehensive value of the cloud platform under different conditions, building curves by using different benchmark performance comprehensive values, and analyzing the change situation of the comprehensive value by using the curves, the baseline performance of the hybrid heterogeneous cloud platform can be analyzed, and the performance compliance situation of the cloud platform can be evaluated.

[0112] In some specific embodiments, the mirror building module 11 may specifically include:

[0113] The model architecture configuration unit is used to create an initial cloud resource model, configure the model architecture type of the initial cloud resource model, configure the number of infrastructure corresponding to each model architecture type and the maximum available cloud resource computing scale corresponding to the initial cloud resource model, so as to obtain an initial configuration model corresponding to the initial cloud resource model;

[0114] A step value setting unit, configured to determine the disk specification of the disk to be detected and the network type of the network to be detected from the initial configuration model, configure a test framework system corresponding to the initial configuration model, and set a linearly increasing step value corresponding to the initial cloud resource model, so as to obtain the hybrid heterogeneous cloud resource model corresponding to the target cloud platform; wherein, the test framework system includes test tools, a baseline performance detection model, test scenarios, and test metrics.

[0115] In some specific embodiments, the image building module 11 may specifically include:

[0116] A data download unit, configured to download the preset basic image from a preset server and download the preset script file from a preset Git repository;

[0117] An image acquisition unit, configured to use the preset script file to load the hybrid heterogeneous cloud resource model into the preset basic image to build an initial image, and tag the initial image to obtain the container image corresponding to the hybrid heterogeneous cloud resource model; wherein, the tag includes version information of the container image and environment information matching the container image, and the environment information includes development environment information, production environment information, and test environment information.

[0118] In some specific embodiments, the model library loading module 12 may specifically include:

[0119] A program running unit, configured to upload the container image to the target cloud platform, instantiate the container image in the target cloud platform to obtain a container instance corresponding to the container image, and run the preset analysis control program in the container instance to start the container instance;

[0120] A test framework loading unit, configured to use the preset analysis control program to load the test framework system to complete the loading of the preset detection model library.

[0121] In some specific embodiments, the virtual machine object creation module 13 may specifically include:

[0122] A computing scale loading unit, configured to load the maximum available cloud resource computing scale corresponding to the hybrid heterogeneous cloud resource model;

[0123] A virtual machine object acquisition unit, configured to create an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and load the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object;

[0124] A performance testing unit is used to perform a benchmark performance test on the current target virtual machine object to obtain a comprehensive benchmark performance value corresponding to the current target virtual machine object, and format and save the comprehensive benchmark performance value corresponding to the current target virtual machine object;

[0125] A first step jump unit is used to determine whether the current computing scale is less than the maximum available cloud resource computing scale. If the current computing scale is less than the maximum available cloud resource computing scale, it jumps to the step of creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and loading the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object, so as to obtain the comprehensive benchmark performance value corresponding to the new target virtual machine object;

[0126] A second step jump unit is used to, if the current computing scale is not less than the maximum available cloud resource computing scale, jump to the step of obtaining the target benchmark performance comprehensive value matrix data corresponding to the target cloud platform based on each of the comprehensive benchmark performance values.

[0127] In some specific embodiments, the baseline performance analysis module 14 may specifically include:

[0128] A curve generation unit is used to obtain the historical comprehensive benchmark performance value matrix data, generate the first target curve according to the historical comprehensive benchmark performance value matrix data, and load the target comprehensive benchmark performance value matrix data, and generate the second target curve according to the target comprehensive benchmark performance value matrix data;

[0129] A deviation data calculation unit is used to calculate the positive deviation data and the negative deviation data of the second target curve based on the first target curve, and save the positive deviation data and the negative deviation data.

[0130] In some specific embodiments, the baseline performance analysis module 14 may specifically include:

[0131] A deviation data loading unit is used to load the positive deviation data and the negative deviation data, and obtain the first ratio of the positive deviation data and the second ratio of the negative deviation data;

[0132] A performance evaluation unit is used to judge the baseline performance of the target cloud platform according to the first ratio and the second ratio, and generate a baseline performance evaluation conclusion of the target cloud platform.

[0133] Furthermore, an embodiment of the present application also discloses an electronic device, Figure 4 It is a structural diagram of an electronic device 20 shown according to an exemplary embodiment. The content in the figure should not be regarded as any limitation on the scope of use of the present application.

[0134] Figure 4 This is a schematic structural diagram of an electronic device 20 provided by an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. Among them, the memory 22 is used to store a computer program, and the computer program is loaded and executed by the processor 21 to implement the relevant steps in the cloud platform baseline performance analysis method disclosed in any of the foregoing embodiments. In addition, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0135] In this embodiment, the power supply 23 is used to provide a working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of the present application, and no specific limitation is imposed on it here; the input / output interface 25 is used to obtain external input data or output data to the outside, and its specific interface type can be selected according to specific application needs, and no specific limitation is made here.

[0136] In addition, as a carrier for resource storage, the memory 22 may be a read-only memory, a random access memory, a disk, or an optical disc, etc., and the resources stored thereon may include an operating system 221, a computer program 222, etc., and the storage method may be short-term storage or permanent storage.

[0137] Among them, the operating system 221 is used to manage and control each hardware device and the computer program 222 on the electronic device 20, and it may be Windows Server, Netware, Unix, Linux, etc. In addition to the computer program that can be used to complete the cloud platform baseline performance analysis method executed by the electronic device 20 disclosed in any of the foregoing embodiments, the computer program 222 may further include a computer program that can be used to complete other specific tasks.

[0138] Furthermore, the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the cloud platform baseline performance analysis method disclosed above. For the specific steps of this method, reference may be made to the corresponding content disclosed in the foregoing embodiments, and details are not described herein again.

[0139] In the present specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other. For the apparatuses disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For related parts, reference can be made to the description in the method part.

[0140] Those skilled in the art can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of the examples 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 this application.

[0141] The steps of the methods or algorithms described in combination with the embodiments disclosed herein can be directly implemented by hardware, software modules executed by a processor, or a combination of the two. The software modules can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0142] Finally, it should also be noted that in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover 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 also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0143] The technical solutions provided in this application have been introduced in detail above. Specific examples are used herein to elaborate on the principles and implementation manners of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A cloud platform baseline performance analysis method, characterized in that: include: Configure a hybrid heterogeneous cloud resource model corresponding to the target cloud platform, obtain a preset script file and a preset basic image, and use the preset script file and the preset basic image to build a container image corresponding to the hybrid heterogeneous cloud resource model; Uploading the container image to the target cloud platform to start the container instance corresponding to the container image, and using the preset analysis control program in the container instance to load the preset detection model library; Creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, loading the model data of the hybrid heterogeneous cloud resource model into each of the initial virtual machine objects to obtain each target virtual machine object, and using the preset detection model library to obtain the benchmark performance comprehensive value corresponding to each of the target virtual machine objects; Based on each of the benchmark performance comprehensive values, target benchmark performance comprehensive value matrix data corresponding to the target cloud platform is obtained, a first target curve is generated according to the historical benchmark performance comprehensive value matrix data, and a second target curve is generated using the target benchmark performance comprehensive value matrix data, an offset of the second target curve is obtained according to the first target curve, and the baseline performance of the target cloud platform is analyzed according to the offset.

2. The cloud platform baseline performance analysis method according to claim 1, characterized in that: The configuration of the hybrid heterogeneous cloud resource model corresponding to the target cloud platform includes: Creating an initial cloud resource model, configuring the model architecture type of the initial cloud resource model, and configuring the number of infrastructures corresponding to each model architecture type and the maximum available cloud resource computing scale corresponding to the initial cloud resource model, so as to obtain an initial configuration model corresponding to the initial cloud resource model; Determine the cloud disk specifications of the cloud disk to be tested and the network type of the network to be tested from the initial configuration model, configure the test framework system corresponding to the initial configuration model, and set the linear incremental step value corresponding to the initial cloud resource model to obtain the hybrid heterogeneous cloud resource model corresponding to the target cloud platform; wherein the test framework system includes test tools, baseline performance detection models, test scenarios and test indicators.

3. The cloud platform baseline performance analysis method according to claim 2, characterized in that: The using the preset script file and the preset basic image to construct the container image corresponding to the hybrid heterogeneous cloud resource model includes: Download the preset basic image from the preset server, and download the preset script file from the preset Git repository; The hybrid heterogeneous cloud resource model is loaded into the preset basic image using the preset script file to build an initial image, and the initial image is labeled to obtain the container image corresponding to the hybrid heterogeneous cloud resource model; wherein the label includes version information of the container image and environment information matching the container image, and the environment information includes development environment information, production environment information and test environment information.

4. The cloud platform baseline performance analysis method according to claim 2, characterized in that: The uploading of the container image to the target cloud platform to start the container instance corresponding to the container image, and using the preset analysis control program in the container instance to load the preset detection model library, includes: Uploading the container image to the target cloud platform, instantiating the container image in the target cloud platform to obtain a container instance corresponding to the container image, and running the preset analysis control program in the container instance to start the container instance; The preset analysis control program is used to load the test framework system to complete the loading of the preset detection model library.

5. The cloud platform baseline performance analysis method according to claim 2, characterized in that: The step of creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, loading the model data of the hybrid heterogeneous cloud resource model into each of the initial virtual machine objects to obtain each target virtual machine object, and using the preset detection model library to obtain the benchmark performance comprehensive value corresponding to each of the target virtual machine objects, includes: Loading the maximum available cloud resource computing scale corresponding to the hybrid heterogeneous cloud resource model; Creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and loading the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object; Performing a benchmark performance test on the current target virtual machine object to obtain a benchmark performance comprehensive value corresponding to the current target virtual machine object, and formatting and saving the benchmark performance comprehensive value corresponding to the current target virtual machine object; Determine whether the current computing scale is smaller than the maximum available cloud resource computing scale. If the current computing scale is smaller than the maximum available cloud resource computing scale, jump to the step of creating an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, and load the model data of the hybrid heterogeneous cloud resource model into the current initial virtual machine object to obtain the current target virtual machine object, so as to obtain the comprehensive benchmark performance value corresponding to the new target virtual machine object; If the current computing scale is not less than the maximum available cloud resource computing scale, jump to the step of obtaining the target benchmark performance comprehensive value matrix data corresponding to the target cloud platform based on each benchmark performance comprehensive value.

6. The cloud platform baseline performance analysis method according to any one of claims 1 to 5, characterized in that: The step of generating a first target curve according to the historical benchmark performance comprehensive value matrix data, generating a second target curve using the target benchmark performance comprehensive value matrix data, and obtaining the offset of the second target curve according to the first target curve includes: Acquire the historical benchmark performance comprehensive value matrix data, generate the first target curve according to the historical benchmark performance comprehensive value matrix data, and load the target benchmark performance comprehensive value matrix data, and generate the second target curve according to the target benchmark performance comprehensive value matrix data; Positive deviation data and negative deviation data of the second target curve are calculated based on the first target curve, and the positive deviation data and the negative deviation data are saved.

7. The cloud platform baseline performance analysis method according to claim 6, characterized in that: The analyzing the baseline performance of the target cloud platform according to the offset situation includes: Loading the positive deviation data and the negative deviation data, and acquiring a first ratio of the positive deviation data and a second ratio of the negative deviation data; The baseline performance of the target cloud platform is determined according to the first ratio and the second ratio, and a baseline performance evaluation conclusion of the target cloud platform is generated.

8. A cloud platform baseline performance analysis device, characterized in that: include: An image building module is used to configure a hybrid heterogeneous cloud resource model corresponding to the target cloud platform, obtain a preset script file and a preset basic image, and use the preset script file and the preset basic image to build a container image corresponding to the hybrid heterogeneous cloud resource model; A model library loading module, used to upload the container image to the target cloud platform to start the container instance corresponding to the container image, and use the preset analysis control program in the container instance to load the preset detection model library; A virtual machine object creation module, used to create an initial virtual machine object according to the target step value corresponding to the hybrid heterogeneous cloud resource model, load the model data of the hybrid heterogeneous cloud resource model into each of the initial virtual machine objects to obtain each target virtual machine object, and use the preset detection model library to obtain the benchmark performance comprehensive value corresponding to each of the target virtual machine objects; A baseline performance analysis module is used to obtain target benchmark performance comprehensive value matrix data corresponding to the target cloud platform based on each of the benchmark performance comprehensive values, generate a first target curve according to the historical benchmark performance comprehensive value matrix data, and generate a second target curve using the target benchmark performance comprehensive value matrix data, obtain the offset of the second target curve according to the first target curve, and analyze the baseline performance of the target cloud platform according to the offset.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the cloud platform baseline performance analysis method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store a computer program, which, when executed by a processor, implements the cloud platform baseline performance analysis method as described in any one of claims 1 to 7.

Images