Cloud management platform resource processing method and device, computer equipment, storage medium and computer program product

Abstract

The invention relates to a cloud management platform resource processing method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: in response to resource request information sent by a terminal, determining target health condition information of a bare metal server to be analyzed in a cloud management platform; determining the to-be-analyzed bare metal server of which the target health condition information is normal, and taking the to-be-analyzed bare metal server as a candidate bare metal server; according to the computing resource information of the candidate bare metal servers, the storage resource information of the storage devices associated with the candidate bare metal servers and the bandwidth resource information of the network devices associated with the candidate bare metal servers, screening out a target bare metal server corresponding to the resource request information from the candidate bare metal servers; and determining a target virtual resource corresponding to the resource request information in the cloud management platform, and deploying the target virtual resource in the target bare metal server. By adopting the method, the resource processing accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a cloud management platform resource processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] At present, in order to realize various resource requests of users, how to accurately process resources according to resource requests is crucial.

[0003] In traditional technology, during the process of resource processing, the method of processing is generally adopted based on the computing resources of bare metal servers; however, the factors considered in this method are relatively simple, resulting in low accuracy of resource processing. Summary of the invention

[0004] Based on this, it is necessary to provide a cloud management platform resource processing method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of resource processing in response to the above technical problems.

[0005] In a first aspect, the present application provides a cloud management platform resource processing method, including:

[0006] In response to the resource request information sent by the terminal, obtaining the physical status information of the bare metal server to be analyzed in the cloud management platform;

[0007] Determining target health status information of the bare metal server to be analyzed according to the physical status information;

[0008] From the bare metal servers to be analyzed, a bare metal server to be analyzed whose target health status information is normal is selected as a candidate bare metal server;

[0009] Obtain computing resource information of the candidate bare metal server, storage resource information of a storage device associated with the candidate bare metal server, and bandwidth resource information of a network device associated with the candidate bare metal server;

[0010] Filtering out a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information;

[0011] Determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

[0012] In one of the embodiments, after determining the target virtual resource corresponding to the resource request information in the cloud management platform and deploying the target virtual resource in the target bare metal server, the method further includes:

[0013] Obtaining current operating information of the target bare metal server;

[0014] In a case where the current operation information is abnormal operation information, determining distance information between the other bare metal servers and the target bare metal server; the other bare metal servers are used to represent any bare metal server among the candidate bare metal servers except the target bare metal server;

[0015] Selecting a bare metal server with the smallest distance information from the other bare metal servers as a backup bare metal server;

[0016] Migrate the target virtual resource to the standby bare metal server.

[0017] In one embodiment, migrating the target virtual resource to the standby bare metal server includes:

[0018] Obtaining first operating system information of the target bare metal server and second operating system information of the standby bare metal server;

[0019] When the first operating system information and the second operating system information are different, adjusting the configuration information of the target virtual resource to obtain the target virtual resource after the configuration information is adjusted; the target virtual resource after the configuration information is adjusted matches the standby bare metal server;

[0020] The target virtual resource after the configuration information is adjusted is migrated to the standby bare metal server.

[0021] In one of the embodiments, the physical status information includes temperature information, fan speed information, and power supply information of the bare metal server to be analyzed;

[0022] Determining target health status information of the bare metal server to be analyzed according to the physical status information includes:

[0023] Extracting a first eigenvector corresponding to the temperature information, a second eigenvector corresponding to the fan speed information, and a third eigenvector corresponding to the power supply information;

[0024] The first feature vector, the second feature vector, and the third feature vector are fused to obtain a fused feature vector corresponding to the bare metal server to be analyzed;

[0025] Inputting the fused feature vector into a pre-trained first health status prediction model to obtain a first prediction probability of the bare metal server to be analyzed under preset health status information;

[0026] Inputting the fused feature vector into a pre-trained second health status prediction model to obtain a second prediction probability of the bare metal server to be analyzed under the preset health status information;

[0027] The first prediction probability and the second prediction probability of the bare metal server to be analyzed under the preset health status information are respectively merged to obtain a target prediction probability of the bare metal server to be analyzed under the preset health status information;

[0028] From each of the preset health status information, the preset health status information with the maximum target prediction probability is screened out as the target health status information.

[0029] In one of the embodiments, the step of selecting a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information includes:

[0030] For each of the candidate bare metal servers, normalize the computing resource information, the storage resource information, and the bandwidth resource information to obtain processed computing resource information, processed storage resource information, and processed bandwidth resource information corresponding to each of the candidate bare metal servers;

[0031] Determine, according to the resource request information, a first weight corresponding to the processed computing resource information, a second weight corresponding to the processed storage resource information, and a third weight corresponding to the processed bandwidth resource information;

[0032] According to the first weight, the second weight, and the third weight, the processed computing resource information, the processed storage resource information, and the processed bandwidth resource information are fused to obtain fused resource information corresponding to each of the candidate bare metal servers;

[0033] From the candidate bare metal servers, a candidate bare metal server with the largest fusion resource information is selected as the target bare metal server.

[0034] In one embodiment, determining the target virtual resource corresponding to the resource request information in the cloud management platform includes:

[0035] Determining candidate virtual resources corresponding to the resource request information;

[0036] Determining resource utilization corresponding to each of the candidate virtual resources;

[0037] From the candidate virtual resources, a candidate virtual resource with the highest resource utilization rate is determined as the target virtual resource.

[0038] In a second aspect, the present application also provides a cloud management platform resource processing device, including:

[0039] A status acquisition module, used to obtain physical status information of the bare metal server to be analyzed in the cloud management platform in response to the resource request information sent by the terminal;

[0040] A status determination module, configured to determine target health status information of the bare metal server to be analyzed according to the physical status information;

[0041] A server screening module is used to screen out the bare metal servers to be analyzed whose target health status information is normal from the bare metal servers to be analyzed as candidate bare metal servers;

[0042] An information acquisition module, used to acquire computing resource information of the candidate bare metal server, storage resource information of a storage device associated with the candidate bare metal server, and bandwidth resource information of a network device associated with the candidate bare metal server;

[0043] A target screening module, configured to screen out a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information;

[0044] The resource deployment module is used to determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

[0045] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0046] In response to the resource request information sent by the terminal, obtaining the physical status information of the bare metal server to be analyzed in the cloud management platform;

[0047] Determining target health status information of the bare metal server to be analyzed according to the physical status information;

[0048] From the bare metal servers to be analyzed, a bare metal server to be analyzed whose target health status information is normal is selected as a candidate bare metal server;

[0049] Obtain computing resource information of the candidate bare metal server, storage resource information of a storage device associated with the candidate bare metal server, and bandwidth resource information of a network device associated with the candidate bare metal server;

[0050] Filtering out a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information;

[0051] Determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

[0052] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0053] In response to the resource request information sent by the terminal, obtaining the physical status information of the bare metal server to be analyzed in the cloud management platform;

[0054] Determining target health status information of the bare metal server to be analyzed according to the physical status information;

[0055] From the bare metal servers to be analyzed, a bare metal server to be analyzed whose target health status information is normal is selected as a candidate bare metal server;

[0056] Obtain computing resource information of the candidate bare metal server, storage resource information of a storage device associated with the candidate bare metal server, and bandwidth resource information of a network device associated with the candidate bare metal server;

[0057] Filtering out a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information;

[0058] Determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

[0059] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0060] In response to the resource request information sent by the terminal, obtaining the physical status information of the bare metal server to be analyzed in the cloud management platform;

[0061] Determining target health status information of the bare metal server to be analyzed according to the physical status information;

[0062] From the bare metal servers to be analyzed, a bare metal server to be analyzed whose target health status information is normal is selected as a candidate bare metal server;

[0063] Obtain computing resource information of the candidate bare metal server, storage resource information of a storage device associated with the candidate bare metal server, and bandwidth resource information of a network device associated with the candidate bare metal server;

[0064] Filtering out a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information;

[0065] Determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

[0066] The cloud management platform resource processing method, apparatus, computer equipment, storage medium and computer program product described above firstly obtain physical status information of the bare metal server to be analyzed in the cloud management platform in response to resource request information sent by the terminal, and determine the target health status information of the bare metal server to be analyzed based on the physical status information, and then select the bare metal server to be analyzed whose target health status information is normal from each bare metal server to be analyzed as a candidate bare metal server, and then obtain the computing resource information of the candidate bare metal server, the storage resource information of the storage device associated with the candidate bare metal server, and the bandwidth resource information of the network device associated with the candidate bare metal server, and then select the target bare metal server corresponding to the resource request information from each candidate bare metal server based on the computing resource information, the storage resource information and the bandwidth resource information, and finally determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server. In this way, in the process of resource processing, after responding to the resource request information sent by the terminal, the bare metal server to be analyzed is screened multiple times based on the target health status information, computing resource information, storage resource information of the associated storage device and bandwidth resource information of the associated network device of the bare metal server, so that the target bare metal server matching the resource request information can be obtained more accurately, so that the target bare metal server finally determined can more accurately match the target virtual resource corresponding to the resource request information in the cloud management platform, and then the target virtual resource can be more accurately deployed in the target bare metal server, which is conducive to improving the accuracy of resource processing; moreover, the entire resource processing process takes into account multiple sources of information such as target health status information, computing resource information, storage resource information and bandwidth resource information, avoiding the defect of low accuracy of resource processing caused by the traditional technology of processing based on the computing resources of the bare metal server, which considers relatively single factors, thereby improving the accuracy of resource processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 A schematic diagram of a process of a cloud management platform resource processing method in one embodiment;

[0069] Figure 2 A schematic diagram of a flow chart of steps for migrating target virtual resources to a standby bare metal server in one embodiment;

[0070] Figure 3 A schematic diagram of a flow chart of a cloud management platform resource processing method in another embodiment;

[0071] Figure 4 It is a structural block diagram of a cloud management platform resource processing device in one embodiment;

[0072] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0073] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0074] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0075] In an exemplary embodiment, Figure 1 As shown, a cloud management platform resource processing method is provided. This embodiment uses the method applied to a server as an example for illustration; it can be understood that the method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be but is not limited to various personal computers, laptops, smart phones and tablets; the server can be implemented as an independent server or a server cluster composed of multiple servers. In this embodiment, the method includes the following steps:

[0076] Step S101, in response to resource request information sent by a terminal, obtaining physical status information of a bare metal server to be analyzed in a cloud management platform.

[0077] The resource request information is used to indicate the total amount of resources requested by the terminal, such as a 16-core CPU (Central Processing Unit) and 64 GB of memory.

[0078] The cloud management platform refers to a comprehensive software tool for managing cloud computing resources. Cloud computing resources include actual resources (such as bare metal servers) and virtual resources (such as virtual machines and containers). It should be noted that the cloud management platform can achieve seamless communication and functional integration between the cloud management platform and the BMC (Baseboard Management Controller).

[0079] The bare metal server to be analyzed refers to the bare metal server included in the cloud management platform.

[0080] The physical status information refers to information used to characterize the physical status of the bare metal server to be analyzed, including temperature information, fan speed information, and power supply information of the bare metal server to be analyzed.

[0081] Exemplarily, the server receives resource request information sent by the terminal through a network path between the server and the terminal, and verifies the legitimacy of the resource request information (such as checking whether the data format is correct, whether necessary fields are missing, etc.), and obtains a verification result corresponding to the resource request information; if the verification result is passed, the server responds to the resource request information sent by the terminal, and determines the bare metal server included in the cloud management platform as the bare metal server to be analyzed; then, the server obtains the temperature information of the bare metal server to be analyzed through a thermistor sensor associated with the bare metal server to be analyzed, obtains the fan speed information of the bare metal server to be analyzed through a Hall effect sensor associated with the bare metal server to be analyzed, and obtains the power supply information of the bare metal server to be analyzed through a current sensor and a voltage sensor associated with the bare metal server to be analyzed; then, the server combines the temperature information, fan speed information and power supply information of the bare metal server to be analyzed in a preset combination method to obtain the physical status information of the bare metal server to be analyzed.

[0082] Step S102: Determine target health status information of the bare metal server to be analyzed according to the physical status information.

[0083] The target health status information is used to indicate the health status of the bare metal server to be analyzed, including normal and faulty.

[0084] Exemplarily, the server queries the correspondence between the physical status information and the health status information based on the physical status information, and obtains the health status information corresponding to the physical status information as the target health status information of the bare metal server to be analyzed.

[0085] Step S103: Filter out the bare metal servers to be analyzed whose target health status information is normal from the bare metal servers to be analyzed as candidate bare metal servers.

[0086] The candidate bare metal server refers to a bare metal server to be analyzed whose target health status information is normal.

[0087] Exemplarily, the server determines the target health status information of the bare metal servers to be analyzed; then, the server selects the bare metal servers to be analyzed whose target health status information is normal from the bare metal servers to be analyzed, and uses these bare metal servers to be analyzed as candidate bare metal servers.

[0088] Step S104, obtaining computing resource information of the candidate bare metal server, storage resource information of the storage device associated with the candidate bare metal server, and bandwidth resource information of the network device associated with the candidate bare metal server.

[0089] The computing resource information is used to represent the memory value of the candidate bare metal server, such as 512 GB.

[0090] The storage resource information is used to represent the storage capacity value of the storage device associated with the candidate bare metal server, such as 12 TB.

[0091] The bandwidth resource information is used to represent the network bandwidth value of the network device associated with the candidate bare metal server, such as 60 Gbps.

[0092] Exemplarily, the server obtains connection information of a candidate bare metal server, and based on the connection information, determines a storage device connected to the candidate bare metal server as a storage device associated with the candidate bare metal server, and based on the connection information, determines a network device connected to the candidate bare metal server as a network device associated with the candidate bare metal server; then, the server obtains computing resource information of the candidate bare metal server, storage resource information of the storage device associated with the candidate bare metal server, and bandwidth resource information of the network device associated with the candidate bare metal server.

[0093] Step S105 , screening out a target bare metal server corresponding to the resource request information from each candidate bare metal server according to the computing resource information, the storage resource information, and the bandwidth resource information.

[0094] The target bare metal server refers to the bare metal server used to deploy virtual resources.

[0095] Exemplarily, the server screens out candidate bare metal servers whose computing resource information, storage resource information and bandwidth resource information all meet preset conditions from each candidate bare metal server based on the computing resource information, storage resource information and bandwidth resource information, and uses these candidate bare metal servers as target bare metal servers corresponding to the resource request information; for example, the server screens out candidate bare metal servers whose computing resource information is greater than preset computing resource information, whose storage resource information is greater than preset storage resource information and whose bandwidth resource information is greater than preset bandwidth resource information from each candidate bare metal server, and uses these candidate bare metal servers as target bare metal servers corresponding to the resource request information.

[0096] Step S106, determining the target virtual resource corresponding to the resource request information in the cloud management platform, and deploying the target virtual resource in the target bare metal server.

[0097] The target virtual resource refers to the virtual resource in the cloud management platform that matches the resource request information.

[0098] Exemplarily, the server determines, from among the candidate virtual resources, a candidate virtual resource corresponding to the resource request information in the cloud management platform as the target virtual resource; then, the server queries the correspondence between the resource type and the deployment method based on the resource type of the target virtual resource, and obtains the deployment method corresponding to the resource type of the target virtual resource as the target deployment method; then, the server deploys the target virtual resource in the target bare metal server according to the target deployment method.

[0099] In the above-mentioned cloud management platform resource processing method, first, in response to the resource request information sent by the terminal, the physical status information of the bare metal server to be analyzed in the cloud management platform is obtained, and based on the physical status information, the target health status information of the bare metal server to be analyzed is determined, and then from each bare metal server to be analyzed, the bare metal server to be analyzed whose target health status information is normal is screened out as a candidate bare metal server, and then, the computing resource information of the candidate bare metal server, the storage resource information of the storage device associated with the candidate bare metal server, and the bandwidth resource information of the network device associated with the candidate bare metal server are obtained, and then, based on the computing resource information, the storage resource information and the bandwidth resource information, the target bare metal server corresponding to the resource request information is screened out from each candidate bare metal server, and finally, the target virtual resource corresponding to the resource request information in the cloud management platform is determined, and the target virtual resource is deployed in the target bare metal server. In this way, in the process of resource processing, after responding to the resource request information sent by the terminal, the bare metal server to be analyzed is screened multiple times based on the target health status information, computing resource information, storage resource information of the associated storage device and bandwidth resource information of the associated network device of the bare metal server, so that the target bare metal server matching the resource request information can be obtained more accurately, so that the target bare metal server finally determined can more accurately match the target virtual resource corresponding to the resource request information in the cloud management platform, and then the target virtual resource can be more accurately deployed in the target bare metal server, which is conducive to improving the accuracy of resource processing; moreover, the entire resource processing process takes into account multiple sources of information such as target health status information, computing resource information, storage resource information and bandwidth resource information, avoiding the defect of low accuracy of resource processing caused by the traditional technology of processing based on the computing resources of the bare metal server, which considers relatively single factors, thereby improving the accuracy of resource processing.

[0100] In an exemplary embodiment, Figure 2 As shown, the above step S106, in determining the target virtual resource corresponding to the resource request information in the cloud management platform, and deploying the target virtual resource in the target bare metal server, specifically includes the following contents:

[0101] Step S201, obtaining the current operation information of the target bare metal server.

[0102] Step S202, when the current operation information is abnormal operation information, determine the distance information between other bare metal servers and the target bare metal server; the other bare metal servers are used to represent any bare metal server among the candidate bare metal servers except the target bare metal server.

[0103] Step S203: Filter out the bare metal server with the smallest distance information from other bare metal servers as a backup bare metal server.

[0104] Step S204: Migrate the target virtual resources to the standby bare metal server.

[0105] The current operation information refers to the operation information of the target bare metal server at the current time, including normal operation information and abnormal operation information.

[0106] The abnormal operation information is used to indicate abnormal situations that occur during the operation of the target bare metal server, such as the target bare metal server needs maintenance, upgrade or failure.

[0107] The distance information is used to indicate the logical distance between other bare metal servers and the target bare metal server.

[0108] The standby bare metal server is used to represent a bare metal server having the smallest distance information to the target bare metal server.

[0109] Exemplarily, the server obtains the current log information of the target bare metal server, extracts the current operating information of the target bare metal server from the current log information, and judges the current operating information; then, the server uses any bare metal server other than the target bare metal server among the candidate bare metal servers as other bare metal servers, and when it is judged that the current operating information is abnormal operating information, the server obtains the first location information corresponding to the target bare metal server and the second location information corresponding to the other bare metal servers, and determines the distance information between the other bare metal servers and the target bare metal server based on the first location information and the second location information; then, the server selects the bare metal server with the smallest distance information from the other bare metal servers, and uses the bare metal server as a backup bare metal server; then, the server migrates the target virtual resources to the backup bare metal server.

[0110] In this embodiment, when it is detected that the current operating information of the target bare metal server is abnormal operating information, by determining the distance information between other bare metal servers and the target bare metal server, and screening out the backup bare metal server with the smallest distance information, rapid fault switching can be achieved, thereby ensuring the operating stability of the target virtual resources.

[0111] In an exemplary embodiment, the above step S204, migrating the target virtual resources to the backup bare metal server, specifically includes the following contents: obtaining the first operating system information of the target bare metal server and the second operating system information of the backup bare metal server; when the first operating system information and the second operating system information are different, adjusting the configuration information of the target virtual resource to obtain the target virtual resource after the configuration information is adjusted; matching the target virtual resource after the configuration information is adjusted with the backup bare metal server; migrating the target virtual resource after the configuration information is adjusted to the backup bare metal server.

[0112] The first operating system information is used to represent the operating system information of the target bare metal server, which may be a Linux (an operating system) operating system or a Windows (an operating system) operating system.

[0113] The second operating system information is used to represent the operating system information of the standby bare metal server, which may be a Linux operating system or a Windows operating system.

[0114] The configuration information of the target virtual resource includes network configuration information and storage configuration information of the target virtual resource.

[0115] Exemplarily, the server obtains the first server information of the target bare metal server, and extracts the operating system information of the target bare metal server from the first server information as the first operating system information, and obtains the second server information of the backup bare metal server, and extracts the operating system information of the backup bare metal server from the second server information as the second operating system information; then, the server compares the first operating system information and the second operating system information; when the first operating system information and the second operating system information are different, the server adjusts the network configuration information and storage configuration information of the target virtual resource to obtain the target virtual resource with the adjusted configuration information that matches the backup bare metal server, and migrates the target virtual resource with the adjusted configuration information to the backup bare metal server; when the first operating system information and the second operating system information are the same, the server directly migrates the target virtual resource to the backup bare metal server.

[0116] In this embodiment, when the first operating system information of the target bare metal server and the second operating system information of the backup bare metal server are different, the configuration information of the target virtual resources is adjusted so that the target virtual resources can be compatible with the operating system environment of the backup bare metal server, thereby ensuring that the virtual resources can run normally on the backup server.

[0117] In an exemplary embodiment, the physical status information includes temperature information, fan speed information, and power supply information of the bare metal server to be analyzed.

[0118] Then, the above step S102, based on the physical status information, determines the target health status information of the bare metal server to be analyzed, specifically including the following contents: extracting the first eigenvector corresponding to the temperature information, the second eigenvector corresponding to the fan speed information, and the third eigenvector corresponding to the power supply information; fusing the first eigenvector, the second eigenvector, and the third eigenvector to obtain the fused eigenvector corresponding to the bare metal server to be analyzed; inputting the fused eigenvector into a pre-trained first health status prediction model to obtain the first prediction probability of the bare metal server to be analyzed under the preset health status information; inputting the fused eigenvector into a pre-trained second health status prediction model to obtain the second prediction probability of the bare metal server to be analyzed under the preset health status information; fusing the first prediction probability and the second prediction probability of the bare metal server to be analyzed under the preset health status information respectively to obtain the target prediction probability of the bare metal server to be analyzed under the preset health status information; and screening out the preset health status information with the largest target prediction probability from each preset health status information as the target health status information.

[0119] The temperature information is used to represent the CPU temperature value of the bare metal server to be analyzed.

[0120] The fan speed information is used to indicate the speed value of the cooling fan of the bare metal server to be analyzed.

[0121] The power supply information is used to represent the input power supply parameters of the bare metal server to be analyzed, such as voltage, current, etc.

[0122] The first eigenvector is used to represent the characterization vector corresponding to the temperature information.

[0123] The second eigenvector is used to represent the characterization vector corresponding to the fan speed information.

[0124] The third eigenvector is used to represent the characterization vector corresponding to the power information.

[0125] The fused feature vector refers to a feature vector obtained by fusing the first feature vector, the second feature vector and the third feature vector.

[0126] The first health status prediction model refers to a deep learning model that can predict the health status information of the bare metal server, such as a convolutional neural network model.

[0127] The preset health status information refers to the preset health status information, including normal and fault.

[0128] The first predicted probability is used to represent the predicted probability of the bare metal server to be analyzed predicted by the first health status prediction model under the preset health status information.

[0129] The second health status prediction model refers to a time series model that can predict the health status information of the bare metal server, such as an autoregressive integrated moving average model.

[0130] The second predicted probability is used to represent the predicted probability of the bare metal server to be analyzed predicted by the second health status prediction model under the preset health status information.

[0131] The target prediction probability refers to the prediction probability obtained by fusing the first prediction probability and the second prediction probability of the bare metal server to be analyzed under the preset health status information.

[0132] Exemplarily, the server uses temperature information as main data, and uses fan speed information and power supply information as auxiliary data, and inputs them into the feature extraction model for feature extraction processing, and obtains a feature vector corresponding to the temperature information as a first feature vector; then, the server uses fan speed information as main data, and uses temperature information and power supply information as auxiliary data, and inputs them into the feature extraction model for feature extraction processing, and obtains a feature vector corresponding to the fan speed information as a second feature vector; then, the server uses power supply information as main data, and uses temperature information and fan speed information as auxiliary data, and inputs them into the feature extraction model for feature extraction processing, and obtains a feature vector corresponding to the power supply information as a third feature vector; then, the server performs weighted summation processing on the first feature vector, the second feature vector, and the third feature vector according to the weight corresponding to the first feature vector, the weight corresponding to the second feature vector, and the weight corresponding to the third feature vector. A fused feature vector corresponding to the bare metal server to be analyzed is obtained; then, the server inputs the fused feature vector into a pre-trained first health status prediction model to obtain a first prediction probability of the bare metal server to be analyzed under preset health status information, and inputs the fused feature vector into a pre-trained second health status prediction model to obtain a second prediction probability of the bare metal server to be analyzed under preset health status information; then, the server performs weighted summation processing on the first prediction probability and the second prediction probability of the bare metal server to be analyzed under preset health status information according to the model weight of the first health status prediction model and the model weight of the second health status prediction model, respectively, to obtain a target prediction probability of the bare metal server to be analyzed under the preset health status information; then, the server screens out the preset health status information with the largest target prediction probability from each preset health status information, and uses the preset health status information as the target health status information.

[0133] In this embodiment, by extracting the feature vectors corresponding to the temperature information, fan speed information and power supply information and performing fusion processing, and combining the first health status prediction model and the second health status prediction model, the target prediction probability of the bare metal server to be analyzed under the preset health status information can be obtained more accurately, and then the target health status information of the bare metal server to be analyzed can be more accurately determined.

[0134] In an exemplary embodiment, the above step S105, based on the computing resource information, storage resource information and bandwidth resource information, selects the target bare metal server corresponding to the resource request information from each candidate bare metal server, specifically including the following contents: for each candidate bare metal server, respectively normalize the computing resource information, storage resource information and bandwidth resource information to obtain the processed computing resource information, processed storage resource information and processed bandwidth resource information corresponding to each candidate bare metal server; according to the resource request information, determine the first weight corresponding to the processed computing resource information, the second weight corresponding to the processed storage resource information and the third weight corresponding to the processed bandwidth resource information; according to the first weight, the second weight and the third weight, fuse the processed computing resource information, the processed storage resource information and the processed bandwidth resource information to obtain the fused resource information corresponding to each candidate bare metal server; from each candidate bare metal server, filter out the candidate bare metal server with the largest fused resource information as the target bare metal server.

[0135] The processed computing resource information refers to the computing resource information after normalization.

[0136] The processed storage resource information refers to the storage resource information after normalization processing.

[0137] The processed bandwidth resource information refers to the bandwidth resource information after normalization processing.

[0138] It should be noted that the processed computing resource information, the processed storage resource information, and the processed bandwidth resource information are all values ​​between 0 and 1, such as 0.5, 0.2, and the like.

[0139] The first weight is used to indicate the importance of the processed computing resource information.

[0140] The second weight is used to indicate the importance of the processed storage resource information.

[0141] The third weight is used to indicate the importance of the processed bandwidth resource information.

[0142] The fused resource information is used to represent a value obtained by fusion processing of the processed computing resource information, the processed storage resource information and the processed bandwidth resource information.

[0143] Exemplarily, the server performs denoising processing on the computing resource information, storage resource information, and bandwidth resource information for each candidate bare metal server, respectively, to obtain the denoised computing resource information, denoised storage resource information, and denoised bandwidth resource information corresponding to each candidate bare metal server; then, the server performs normalization processing on the denoised computing resource information, denoised storage resource information, and denoised bandwidth resource information for each candidate bare metal server, respectively, to obtain the processed computing resource information, processed storage resource information, and processed bandwidth resource information corresponding to each candidate bare metal server; then, the server extracts key resources from the resource request information. The server receives source request information, and determines, according to the key resource request information, a first weight corresponding to the processed computing resource information, a second weight corresponding to the processed storage resource information, and a third weight corresponding to the processed bandwidth resource information; then, the server performs weighted summation processing on the processed computing resource information, the processed storage resource information, and the processed bandwidth resource information according to the first weight, the second weight, and the third weight, to obtain the fusion resource information corresponding to each candidate bare metal server; then, the server selects the candidate bare metal server with the largest fusion resource information from each candidate bare metal server, and uses the candidate bare metal server as the target bare metal server.

[0144] In this embodiment, by normalizing the computing resource information, storage resource information, and bandwidth resource information of each candidate bare metal server, resource information of different types and magnitudes can be compared on the same scale, avoiding resource evaluation deviations caused by different dimensions, thereby enabling a more reasonable evaluation of the bare metal servers.

[0145] In an exemplary embodiment, the above step S106 determines the target virtual resource corresponding to the resource request information in the cloud management platform, and specifically includes the following contents: determining the candidate virtual resources corresponding to the resource request information; determining the resource utilization corresponding to each candidate virtual resource; and determining the candidate virtual resource with the highest resource utilization from each candidate virtual resource as the target virtual resource.

[0146] The candidate virtual resources are used to represent the combination of virtual machines and containers, for example, 3 containers and 4 virtual machines, or 6 containers and 2 virtual machines.

[0147] The resource utilization rate is used to indicate the resource utilization efficiency of the candidate virtual resources. For example, the resource utilization rate of 3 containers and 4 virtual machines is 60%, or the resource utilization rate of 6 containers and 2 virtual machines is 80%.

[0148] Exemplarily, the server determines the combination of virtual machines and containers based on the resource request information as the candidate virtual resources corresponding to the resource request information; then, the server determines the CPU utilization and memory utilization corresponding to each candidate virtual resource, both as the resource utilization corresponding to each candidate virtual resource; then, the server determines the candidate virtual resource with the highest resource utilization from each candidate virtual resource, and uses the candidate virtual resource as the target virtual resource.

[0149] In this embodiment, by determining the candidate virtual resources corresponding to the resource request information, a set of virtual resources that may meet the requirements can be screened out based on actual business needs, and the virtual resource with the highest resource utilization rate is selected as the target virtual resource, ensuring that the selected resources can meet business needs to the greatest extent, avoiding over-configuration of resources, and helping to improve overall resource utilization efficiency.

[0150] In an exemplary embodiment, Figure 3 As shown, another cloud management platform resource processing method is provided, which is described by taking the method applied to a server as an example, and includes the following steps:

[0151] Step S301, in response to the resource request information sent by the terminal, obtain the physical status information of the bare metal server to be analyzed in the cloud management platform; the physical status information includes the temperature information, fan speed information and power supply information of the bare metal server to be analyzed.

[0152] Step S302, extracting a first eigenvector corresponding to the temperature information, a second eigenvector corresponding to the fan speed information, and a third eigenvector corresponding to the power supply information; fusing the first eigenvector, the second eigenvector, and the third eigenvector to obtain a fused eigenvector corresponding to the bare metal server to be analyzed.

[0153] Step S303: Input the fused feature vector into a pre-trained first health status prediction model to obtain a first prediction probability of the bare metal server to be analyzed under the preset health status information; input the fused feature vector into a pre-trained second health status prediction model to obtain a second prediction probability of the bare metal server to be analyzed under the preset health status information.

[0154] Step S304: The first prediction probability and the second prediction probability of the bare metal server to be analyzed under the preset health status information are respectively merged to obtain the target prediction probability of the bare metal server to be analyzed under the preset health status information.

[0155] Step S305: Filter out the preset health status information with the highest target prediction probability from each preset health status information as the target health status information.

[0156] Step S306: Filter out the bare metal servers to be analyzed whose target health status information is normal from the bare metal servers to be analyzed as candidate bare metal servers.

[0157] Step S307, obtaining computing resource information of the candidate bare metal server, storage resource information of the storage device associated with the candidate bare metal server, and bandwidth resource information of the network device associated with the candidate bare metal server.

[0158] Step S308, for each candidate bare metal server, normalize the computing resource information, storage resource information, and bandwidth resource information to obtain processed computing resource information, processed storage resource information, and processed bandwidth resource information corresponding to each candidate bare metal server.

[0159] Step S309: Determine, according to the resource request information, a first weight corresponding to the processed computing resource information, a second weight corresponding to the processed storage resource information, and a third weight corresponding to the processed bandwidth resource information.

[0160] Step S310: According to the first weight, the second weight and the third weight, the processed computing resource information, the processed storage resource information and the processed bandwidth resource information are integrated to obtain integrated resource information corresponding to each candidate bare metal server.

[0161] Step S311, selecting a candidate bare metal server with the largest fusion resource information from each candidate bare metal server as the target bare metal server.

[0162] Step S312, determine the candidate virtual resources corresponding to the resource request information; determine the resource utilization rate corresponding to each candidate virtual resource; and determine the candidate virtual resource with the highest resource utilization rate from among the candidate virtual resources as the target virtual resource.

[0163] Step S313: deploy the target virtual resources in the target bare metal server.

[0164] In the above-mentioned cloud management platform resource processing method, during the process of resource processing, after responding to the resource request information sent by the terminal, the bare metal server to be analyzed is screened multiple times based on the target health status information, computing resource information, storage resource information of the associated storage device and bandwidth resource information of the associated network device of the bare metal server, so that the target bare metal server matching the resource request information can be obtained more accurately, so that the target bare metal server finally determined can more accurately match the target virtual resource corresponding to the resource request information in the cloud management platform, and then the target virtual resource can be more accurately deployed in the target bare metal server, which is conducive to improving the accuracy of resource processing; moreover, the entire resource processing process takes into account multiple sources of information such as target health status information, computing resource information, storage resource information and bandwidth resource information, avoiding the defect of low accuracy of resource processing caused by the method of processing based on the computing resources of the bare metal server in the traditional technology, which takes into account relatively single factors, thereby improving the accuracy of resource processing.

[0165] In an exemplary embodiment, in order to more clearly illustrate the cloud management platform resource processing method provided by the embodiment of the present application, the cloud management platform resource processing method is specifically described with a specific embodiment below. In one embodiment, the present application also provides a software and hardware integrated cloud management platform and its management and control method. In the process of resource processing, by developing a special communication protocol and interface module, seamless communication and functional integration between the cloud management platform and the BMC are achieved, and the cloud management platform is allowed to directly call the BMC function. At the same time, the hardware information collected by the BMC is integrated into the decision-making process of cloud resource management. Based on the comprehensive monitoring data of multiple resources, a unique resource dynamic allocation algorithm is developed. According to business needs, resource performance and hardware health status, the resource allocation and scheduling strategy is optimized to achieve efficient resource utilization and overall system performance improvement. Combined with the hardware fault diagnosis capability of the BMC and the software fault detection function of the cloud management platform, a comprehensive fault intelligent diagnosis and processing mechanism is formed. When a fault occurs, the platform can quickly determine whether it is a hardware fault or a software fault, and automatically take corresponding processing measures according to the fault type, such as automatically restarting the virtual machine, switching to a backup bare metal server, and remotely resetting the hardware. Specifically including the following contents:

[0166] The overall architecture of this embodiment consists of a hardware layer, a BMC layer, and a cloud management platform layer.

[0167] 1. Hardware layer:

[0168] This layer includes the basic hardware devices that constitute the cloud environment, such as bare metal servers, storage devices, and network devices (routers, switches, firewalls, etc.). Storage devices include hard disks, SSDs (Solid State Drives), storage arrays, etc. Bare metal servers are servers with physical hardware that provide a physical operating environment for virtual machines and containers. These servers are equipped with various sensors, such as temperature sensors, voltage sensors, and current sensors, to monitor the physical status of the hardware. The server controller can receive control instructions from the BMC layer to implement operations such as power on / off, restart, and hardware reset. Storage devices are responsible for storing various data in the cloud platform, including virtual machine images, user data, and container storage volumes. Different types of storage devices can be used in combination according to performance and requirements, such as high-performance SSDs for storing data that requires high read and write speeds, and large-capacity hard disks for storing infrequently used data. Network devices ensure network connectivity and communication within the cloud platform, provide network bandwidth, routing, switching, and other functions, and are the basis for communication between cloud resources and between cloud resources and external networks.

[0169] 2. BMC layer:

[0170] BMC is an important part of every bare metal server. Its main function is to directly manage and monitor the server's hardware. BMC communicates with the bare metal server at the hardware layer through the IPMI (Intelligent Platform Management Interface) protocol. IPMI is an industry-standard hardware management interface that allows system administrators to manage, monitor and maintain servers locally and remotely, even when the operating system is not installed or running. Redfish (an industry standard specification) is an emerging open standard that provides a RESTful (Representational State Transfer, a software design principle) interface based on HTTP (HyperText Transfer Protocol) for easier integration and use. It includes the following features:

[0171] (1) Power management: You can remotely turn on, shut down, or restart the bare metal server and perform power cycling on the server. When a server fails, you can try to perform simple troubleshooting through power management.

[0172] (2) Temperature monitoring: The temperature sensors inside the server monitor the temperature of key components (such as CPU, memory, hard disk, etc.) in real time. When the temperature exceeds the set threshold, an alarm can be triggered or appropriate cooling measures can be taken.

[0173] (3) Hardware log records: Record various hardware operations and events, such as hardware errors, replacement of hardware components, firmware updates, etc., to provide a historical basis for troubleshooting and system maintenance.

[0174] (4) Hardware fault diagnosis: It can detect hardware component failures, such as memory errors, hard disk failures, CPU anomalies, etc., and feed back the fault information to the upper-level cloud management platform layer so that appropriate processing measures can be taken.

[0175] 3. Cloud management platform layer:

[0176] This layer is responsible for managing virtual resources such as virtual machines and containers, and integrates the functions of the BMC layer to achieve comprehensive management and control of the entire cloud platform resources.

[0177] (1) Virtual machine management: The virtualization platform is used to create, delete, migrate, allocate resources (CPU, memory, storage, network resources), create and restore snapshots of virtual machines. For example, when a user requests to create a new virtual machine, the cloud management platform will select appropriate resources from the resource pool at the hardware layer and start the virtual machine based on the resource requirements specified by the user (such as 4-core CPU, 8GB memory, 100GB storage).

[0178] (2) Container management: Use the container orchestration tool Kubernetes to implement operations such as container creation, destruction, scaling, and service discovery. Containers are a lightweight virtualization technology that has a faster startup speed and higher resource utilization than virtual machines, making them suitable for microservice architectures.

[0179] (3) Integrated BMC function: Through custom plug-ins or drivers, the cloud management platform layer establishes communication with the BMC layer to obtain the hardware information of the bare metal server and perform hardware operations. For example, when the power status of the bare metal server needs to be adjusted, the cloud management platform can call the power management function of the BMC; when obtaining the resource status of the bare metal server, the hardware information fed back by the BMC will be taken into consideration in resource management, making the deployment and scheduling of virtual machines or containers more reasonable.

[0180] 4. User interface layer:

[0181] It provides users with an entry point for operating and managing the cloud platform. One is the web interface. Users can log in to the platform through a browser and intuitively view the status of cloud resources (such as available resources, allocated resources, and resource usage), create resource requests (such as applying to create a new virtual machine or container), perform resource operations (such as migrating virtual machines, adjusting container resources), and view operation history and system logs. The other is the API (Application Programming Interface) interface: It allows users to call the platform's functions programmatically, providing developers with a more flexible operation method that can be integrated with the company's automated processes and other systems. For example, developers can automatically create virtual machines through API calls and integrate them into the CI (Continuous Integration) / CD (Continuous Delivery) process to achieve automated deployment.

[0182] The specific functions are as follows:

[0183] 1. Unified resource management:

[0184] Users do not need to switch between different management tools, and can operate bare metal, virtual machines, containers, networks, and storage resources through a unified interface or API. For example, users can specify the required resource type (such as creating an application environment with 2 containers and 1 virtual machine) in a resource application interface, and the platform will automatically handle resource allocation and scheduling.

[0185] When allocating resources, the platform will comprehensively consider the availability and performance of various resources. For example, when allocating virtual machines and containers to bare metal servers, not only CPU and memory resources are considered, but also factors such as the available space of storage devices and the bandwidth of network devices are combined to ensure the resource balance of the entire system. When a user requests to allocate storage resources for a container, the platform will automatically find and allocate space on the storage device, and allocate the corresponding network bandwidth to the container based on the network load.

[0186] 2. Fine hardware management:

[0187] BMC monitors the hardware status in real time and feeds the information back to the cloud management platform, which displays the information on the resource management interface. For example, users can see the CPU temperature curve, fan speed, power status and other information of the bare metal server on the interface, so as to understand the hardware health status at any time.

[0188] In addition to basic power management, more complex hardware control can also be achieved. When the server temperature is too high, the BMC can automatically adjust the fan speed according to the preset strategy; if the CPU load is low, the BMC can reduce the CPU frequency to reduce power consumption; when the server needs maintenance, the BMC can perform a hardware reset operation.

[0189] Based on the hardware usage and performance indicators, the platform can provide users with hardware upgrade suggestions. For example, when the I / O (Input / Output) performance of a storage device reaches a bottleneck, the platform will prompt the user to consider adding a new storage device or replacing it with a storage device with better performance.

[0190] 3. Resource optimization:

[0191] Virtual machines and containers: When imbalanced resource usage is detected for virtual machines or containers, such as a virtual machine whose CPU utilization is persistently too high, the platform can migrate it to a bare metal server with more available CPU resources. For containers, the platform can automatically scale the capacity based on load conditions.

[0192] Bare metal server: Dynamically adjust hardware resources based on hardware health and performance indicators, such as shutting down some CPU cores to reduce power consumption when the load is low; enabling more hardware resources (such as opening more memory channels) when the load is high.

[0193] Storage devices: When the I / O load of a storage device is too high, some data can be migrated to other storage devices for load balancing. Based on storage usage, storage can be automatically tiered to store hot data on high-performance storage devices and cold data on large-capacity storage devices.

[0194] Network equipment: Automatically adjust network routing based on network traffic and latency, and distribute traffic to different network paths to improve network performance.

[0195] 4. Cross-platform resource scheduling:

[0196] When a bare metal server needs maintenance, upgrade, or fails, the platform will automatically migrate the virtual machines on that server to other available bare metal servers, ensuring that the running status of the virtual machines is not affected during the migration process. For example, using the live migration technology of virtual machines, the virtual machines can be migrated to other servers with sufficient resources without interrupting services.

[0197] For containers, you can migrate containers from one node to another according to the container orchestration rules and resource conditions. For example, when the node where the container is located is resource-constrained, the container is migrated to a node with more abundant resources, and the service discovery information is updated to ensure the normal operation of the service.

[0198] When scheduling resources across platforms, we take into account factors such as hardware performance and operating system compatibility of different platforms. For example, when migrating a virtual machine from a Linux-based bare metal server to a Windows-based bare metal server, we ensure the compatibility of the operating system and applications, and adjust the network and storage configurations so that the migrated virtual machine can run normally.

[0199] In the above embodiment, during the resource processing process, after responding to the resource request information sent by the terminal, the bare metal server to be analyzed is screened multiple times based on the target health status information, computing resource information, storage resource information of the associated storage device, and bandwidth resource information of the associated network device of the bare metal server, so that the target bare metal server matching the resource request information can be obtained more accurately, so that the target bare metal server finally determined can more accurately match the target virtual resource corresponding to the resource request information in the cloud management platform, and then the target virtual resource can be more accurately deployed in the target bare metal server, which is conducive to improving the accuracy of resource processing; moreover, the entire resource processing process takes into account multiple sources of information such as target health status information, computing resource information, storage resource information, and bandwidth resource information, avoiding the defect of low accuracy of resource processing caused by the relatively single factor considered in the traditional technology for processing based on the computing resources of the bare metal server, thereby improving the accuracy of resource processing. At the same time, this embodiment deeply integrates the resource management logic of the cloud management platform with the hardware management function of the BMC; when performing resource scheduling, the cloud management platform can call the hardware information of the BMC and use the performance and status of the hardware as an important basis for decision-making; when hardware abnormalities occur, the BMC can promptly transmit the fault information to the cloud management platform for overall fault handling and resource adjustment; when scheduling resources, combined with the hardware health information provided by the BMC, avoid allocating resources to servers with hardware risks; when there is a risk of hardware failure, it will actively adjust resource allocation to avoid affecting the business; when hardware fails, the monitoring mechanism of the cloud management can promptly report the fault information to the user, thereby improving the efficiency of fault discovery.

[0200] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0201] Based on the same inventive concept, the embodiment of the present application also provides a cloud management platform resource processing device for implementing the cloud management platform resource processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more cloud management platform resource processing device embodiments provided below can refer to the limitations of the cloud management platform resource processing method above, and will not be repeated here.

[0202] In an exemplary embodiment, Figure 4 As shown, a cloud management platform resource processing device is provided, including: a state acquisition module 401, a status determination module 402, a server screening module 403, an information acquisition module 404, a target screening module 405 and a resource deployment module 406, wherein:

[0203] The status acquisition module 401 is used to obtain the physical status information of the bare metal server to be analyzed in the cloud management platform in response to the resource request information sent by the terminal.

[0204] The status determination module 402 is used to determine target health status information of the bare metal server to be analyzed based on the physical status information.

[0205] The server screening module 403 is used to screen out the bare metal servers to be analyzed whose target health status information is normal from the bare metal servers to be analyzed as candidate bare metal servers.

[0206] The information acquisition module 404 is used to acquire computing resource information of the candidate bare metal server, storage resource information of the storage device associated with the candidate bare metal server, and bandwidth resource information of the network device associated with the candidate bare metal server.

[0207] The target screening module 405 is used to screen out a target bare metal server corresponding to the resource request information from each candidate bare metal server according to the computing resource information, the storage resource information and the bandwidth resource information.

[0208] The resource deployment module 406 is used to determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

[0209] In an exemplary embodiment, the cloud management platform resource processing device also includes a resource migration module, which is used to obtain the current operating information of the target bare metal server; when the current operating information is abnormal operating information, determine the distance information between other bare metal servers and the target bare metal server; other bare metal servers are used to represent any bare metal server among the candidate bare metal servers except the target bare metal server; from other bare metal servers, select the bare metal server with the smallest distance information as a backup bare metal server; and migrate the target virtual resources to the backup bare metal server.

[0210] In an exemplary embodiment, the resource migration module is also used to obtain the first operating system information of the target bare metal server and the second operating system information of the backup bare metal server; when the first operating system information and the second operating system information are different, the configuration information of the target virtual resource is adjusted to obtain the target virtual resource after the configuration information is adjusted; the target virtual resource after the configuration information is adjusted is matched with the backup bare metal server; and the target virtual resource after the configuration information is adjusted is migrated to the backup bare metal server.

[0211] In an exemplary embodiment, the status determination module 402 is also used to extract a first eigenvector corresponding to the temperature information, a second eigenvector corresponding to the fan speed information, and a third eigenvector corresponding to the power supply information; fuse the first eigenvector, the second eigenvector, and the third eigenvector to obtain a fused eigenvector corresponding to the bare metal server to be analyzed; input the fused eigenvector into a pre-trained first health status prediction model to obtain a first prediction probability of the bare metal server to be analyzed under preset health status information; input the fused eigenvector into a pre-trained second health status prediction model to obtain a second prediction probability of the bare metal server to be analyzed under preset health status information; fuse the first prediction probability and the second prediction probability of the bare metal server to be analyzed under preset health status information respectively to obtain a target prediction probability of the bare metal server to be analyzed under preset health status information; and select the preset health status information with the largest target prediction probability from each preset health status information as the target health status information.

[0212] In an exemplary embodiment, the target screening module 405 is also used to normalize the computing resource information, storage resource information and bandwidth resource information for each candidate bare metal server, respectively, to obtain the processed computing resource information, processed storage resource information and processed bandwidth resource information corresponding to each candidate bare metal server; determine the first weight corresponding to the processed computing resource information, the second weight corresponding to the processed storage resource information and the third weight corresponding to the processed bandwidth resource information according to the resource request information; fuse the processed computing resource information, the processed storage resource information and the processed bandwidth resource information according to the first weight, the second weight and the third weight to obtain the fused resource information corresponding to each candidate bare metal server; and filter out the candidate bare metal server with the largest fused resource information from each candidate bare metal server as the target bare metal server.

[0213] In an exemplary embodiment, the resource deployment module 406 is also used to determine the candidate virtual resources corresponding to the resource request information; determine the resource utilization corresponding to each candidate virtual resource; and determine the candidate virtual resource with the highest resource utilization from among the candidate virtual resources as the target virtual resource.

[0214] Each module in the above-mentioned cloud management platform resource processing device can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0215] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as physical state information and target health status information. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a cloud management platform resource processing method is implemented.

[0216] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0217] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0218] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0219] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0220] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0221] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0222] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A cloud management platform resource processing method, characterized in that: The method comprises: In response to the resource request information sent by the terminal, obtaining the physical status information of the bare metal server to be analyzed in the cloud management platform; Determining target health status information of the bare metal server to be analyzed according to the physical status information; From the bare metal servers to be analyzed, a bare metal server to be analyzed whose target health status information is normal is selected as a candidate bare metal server; Obtain computing resource information of the candidate bare metal server, storage resource information of a storage device associated with the candidate bare metal server, and bandwidth resource information of a network device associated with the candidate bare metal server; Filtering out a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information; Determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

2. The method according to claim 1, characterized in that After determining the target virtual resource corresponding to the resource request information in the cloud management platform and deploying the target virtual resource in the target bare metal server, the method further includes: Obtaining current operating information of the target bare metal server; In a case where the current operation information is abnormal operation information, determining distance information between the other bare metal servers and the target bare metal server; the other bare metal servers are used to represent any bare metal server among the candidate bare metal servers except the target bare metal server; Selecting a bare metal server with the smallest distance information from the other bare metal servers as a backup bare metal server; Migrate the target virtual resource to the standby bare metal server.

3. The method according to claim 2, characterized in that Migrating the target virtual resource to the standby bare metal server includes: Obtaining first operating system information of the target bare metal server and second operating system information of the standby bare metal server; When the first operating system information and the second operating system information are different, adjusting the configuration information of the target virtual resource to obtain the target virtual resource after the configuration information is adjusted; the target virtual resource after the configuration information is adjusted matches the standby bare metal server; The target virtual resource after the configuration information is adjusted is migrated to the standby bare metal server.

4. The method according to claim 1, characterized in that: The physical status information includes temperature information, fan speed information, and power supply information of the bare metal server to be analyzed; Determining target health status information of the bare metal server to be analyzed according to the physical status information includes: Extracting a first eigenvector corresponding to the temperature information, a second eigenvector corresponding to the fan speed information, and a third eigenvector corresponding to the power supply information; The first feature vector, the second feature vector, and the third feature vector are fused to obtain a fused feature vector corresponding to the bare metal server to be analyzed; Inputting the fused feature vector into a pre-trained first health status prediction model to obtain a first prediction probability of the bare metal server to be analyzed under preset health status information; Inputting the fused feature vector into a pre-trained second health status prediction model to obtain a second prediction probability of the bare metal server to be analyzed under the preset health status information; The first prediction probability and the second prediction probability of the bare metal server to be analyzed under the preset health status information are respectively merged to obtain a target prediction probability of the bare metal server to be analyzed under the preset health status information; From each of the preset health status information, the preset health status information with the maximum target prediction probability is screened out as the target health status information.

5. The method according to claim 1, characterized in that The step of selecting a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information includes: For each of the candidate bare metal servers, normalize the computing resource information, the storage resource information, and the bandwidth resource information to obtain processed computing resource information, processed storage resource information, and processed bandwidth resource information corresponding to each of the candidate bare metal servers; Determine, according to the resource request information, a first weight corresponding to the processed computing resource information, a second weight corresponding to the processed storage resource information, and a third weight corresponding to the processed bandwidth resource information; According to the first weight, the second weight, and the third weight, the processed computing resource information, the processed storage resource information, and the processed bandwidth resource information are fused to obtain fused resource information corresponding to each of the candidate bare metal servers; From the candidate bare metal servers, a candidate bare metal server with the largest fusion resource information is selected as the target bare metal server.

6. The method according to any one of claims 1 to 5, characterized in that: The determining of the target virtual resource corresponding to the resource request information in the cloud management platform includes: Determining candidate virtual resources corresponding to the resource request information; Determining resource utilization corresponding to each of the candidate virtual resources; From the candidate virtual resources, a candidate virtual resource with the highest resource utilization rate is determined as the target virtual resource.

7. A cloud management platform resource processing device, characterized in that: The device comprises: A status acquisition module, used to obtain physical status information of the bare metal server to be analyzed in the cloud management platform in response to the resource request information sent by the terminal; A status determination module, configured to determine target health status information of the bare metal server to be analyzed according to the physical status information; A server screening module is used to screen out the bare metal servers to be analyzed whose target health status information is normal from the bare metal servers to be analyzed as candidate bare metal servers; An information acquisition module, used to acquire computing resource information of the candidate bare metal server, storage resource information of a storage device associated with the candidate bare metal server, and bandwidth resource information of a network device associated with the candidate bare metal server; A target screening module, configured to screen out a target bare metal server corresponding to the resource request information from each of the candidate bare metal servers according to the computing resource information, the storage resource information, and the bandwidth resource information; The resource deployment module is used to determine the target virtual resource corresponding to the resource request information in the cloud management platform, and deploy the target virtual resource in the target bare metal server.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images