A resource synchronization method, a multi-cloud management system, a device, an apparatus, and a medium

Abstract

The application provides a resource synchronization method, a multi-cloud management system, a device, equipment and a medium. The method comprises the following steps: determining target original resources to which virtual resources of a target server are mapped according to mapping relationships between the virtual resources of each server in a distributed server cluster and original resources of a multi-cloud management platform; reading alarm data of the target original resources from a local historical database of the target server; obtaining a failure probability of the target original resources in a process of performing resource synchronization according to the alarm data and a first early warning model, the first early warning model being a model for predicting failure of the original resources; and performing resource synchronization between the target server and the multi-cloud management platform in a case where the failure probability is less than a preset threshold. The application realizes early warning of resource synchronization failure by predicting a synchronization failure probability by using the alarm data and the first early warning model before performing resource synchronization.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a resource synchronization method, a multi-cloud management system, an apparatus, a device, and a medium. Background Technology

[0002] With the continuous development of information technology and the increasing sophistication of the network society, more and more projects are choosing to adopt multi-cloud management platforms to manage different types of cloud platforms or local servers in order to pursue more efficient service performance. Multi-cloud management platforms, through a unified interface, achieve centralized scheduling and management of resources, reduce business operating costs, and ensure system security and reliability.

[0003] However, during resource synchronization between a multi-cloud management platform and its managed servers, issues such as synchronization failure or data loss can easily occur due to network anomalies or resource errors. Therefore, it is necessary to develop a resource synchronization method to improve the success rate of resource synchronization between the multi-cloud management platform and the servers. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention provide a resource synchronization method, a multi-cloud management system, an apparatus, a device, and a medium to overcome or at least partially solve the above problems.

[0005] This invention provides a resource synchronization method applied to a target server in a distributed server cluster. The method includes:

[0006] Based on the mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform, the target original resources mapped to the virtual resources of the target server are determined.

[0007] Read the alarm data of the original target resource from the local historical database of the target server;

[0008] Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during the resource synchronization process is obtained. The first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform.

[0009] If the failure probability is less than a preset threshold, resource synchronization is performed between the target server and the multi-cloud management platform.

[0010] Optionally, the method further includes:

[0011] An initial early warning model is obtained from the multi-cloud management platform. The initial early warning model is obtained by the multi-cloud management platform by training a preset model based on the alarm data of the original resources of the multi-cloud management platform.

[0012] Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during resource synchronization is obtained, including:

[0013] Based on the alarm data of the target's original resources, the model parameters of the initial early warning model are corrected to obtain the first early warning model;

[0014] The alarm data of the target original resource is input into the first early warning model to obtain the fault probability.

[0015] Optionally, based on the alarm data of the target original resource, the model parameters of the initial early warning model are corrected to obtain a first early warning model, including:

[0016] Based on the alarm data of the target original resource, the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment are obtained, where the second moment is the adjacent moment after the first moment.

[0017] Based on the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment, the model parameters of the initial early warning model are corrected to obtain the first early warning model.

[0018] Optionally, after determining the target original resource mapped to the virtual resource of the target server, the method further includes:

[0019] Other servers that are associated with the original resources mapped by the virtual resources and the target original resources are identified as the associated servers of the target server.

[0020] Request the associated server to read the alarm data of the original resources that are associated from the local historical database of the associated server;

[0021] Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during resource synchronization is obtained, including:

[0022] Based on the alarm data of the target original resource and the alarm data of the related original resources, the model parameters of the initial early warning model are corrected to obtain the first early warning model;

[0023] The alarm data of the target original resource is input into the first early warning model to obtain the fault probability.

[0024] Optionally, after correcting the model parameters of the initial early warning model to obtain the first early warning model, the method further includes:

[0025] The model parameters of the first early warning model are sent to the multi-cloud management platform so that the multi-cloud management platform can update the initial early warning model based on the model parameters sent by each server in the distributed server cluster.

[0026] The updated initial warning model is obtained from the multi-cloud management platform for use in the next resource synchronization between the target server and the multi-cloud management platform.

[0027] Optionally, the method further includes:

[0028] If the probability of failure is greater than the preset threshold, a failure message is generated and sent to the terminal corresponding to the target server.

[0029] The second aspect of this embodiment also provides a multi-cloud management system, which includes a multi-cloud management platform and a distributed server cluster;

[0030] The multi-cloud management platform is used to create a mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform, and to send the mapping relationship to each server in the server cluster.

[0031] Each server in the distributed server cluster is used to determine the target original resource mapped by the virtual resource of the server according to the received mapping relationship; read the alarm data of the target original resource from the local historical database; obtain the failure probability of the target original resource during the resource synchronization process based on the alarm data of the target original resource and the first early warning model, wherein the first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform; and perform resource synchronization with the multi-cloud management platform when the failure probability is less than a preset threshold.

[0032] A third aspect of this embodiment also provides a resource synchronization device, the device comprising:

[0033] The determination module is used to determine the target original resource mapped to the virtual resource of the target server based on the mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform.

[0034] The reading module is used to read the alarm data of the target's original resources from the local historical database of the target server;

[0035] The fault probability determination module is used to obtain the fault probability of the target original resource failing during the resource synchronization process based on the alarm data of the target original resource and the first early warning model. The first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform.

[0036] The synchronization module is used to synchronize resources between the target server and the multi-cloud management platform when the failure probability is less than a preset threshold.

[0037] Optionally, the device further includes:

[0038] The model acquisition module is used to acquire an initial early warning model from the multi-cloud management platform. The initial early warning model is obtained by the multi-cloud management platform by training a preset model based on the alarm data of the original resources of the multi-cloud management platform.

[0039] The fault probability determination module includes:

[0040] The first correction submodule is used to correct the model parameters of the initial early warning model based on the alarm data of the target original resource to obtain the first early warning model;

[0041] The first input submodule is used to input the alarm data of the target original resource into the first early warning model to obtain the fault probability.

[0042] Optionally, the first correction submodule includes:

[0043] The first correction unit is used to obtain, based on the alarm data of the target original resource, the failure probability of the target original resource at a first moment, the probability of the target original resource transitioning from normal to failure at a second moment, and the probability of the target original resource transitioning from failure to normal at a second moment, wherein the second moment is an adjacent moment after the first moment.

[0044] The second correction unit is used to correct the model parameters of the initial early warning model based on the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment, so as to obtain the first early warning model.

[0045] Optionally, the device further includes:

[0046] The associated server determination module is used to determine other servers that are associated with the target original resource and the original resource mapped by the virtual resource as associated servers of the target server.

[0047] The associated alarm data reading module is used to request the associated server to read the alarm data of the original resources that are associated from the local historical database of the associated server;

[0048] The fault probability determination module includes:

[0049] The second correction submodule is used to correct the model parameters of the initial early warning model based on the alarm data of the target original resource and the alarm data of the related original resources, so as to obtain the first early warning model.

[0050] The second input submodule is used to input the alarm data of the target original resource into the first early warning model to obtain the fault probability.

[0051] Optionally, the device further includes:

[0052] The update module is used to send the model parameters of the first early warning model to the multi-cloud management platform, so that the multi-cloud management platform updates the initial early warning model according to the model parameters sent by each server in the distributed server cluster.

[0053] The acquisition module is used to acquire the updated initial warning model from the multi-cloud management platform for the next resource synchronization between the target server and the multi-cloud management platform.

[0054] Optionally, the device further includes:

[0055] The fault warning module is used to generate and send fault information to the terminal corresponding to the target server when the fault probability is greater than the preset threshold.

[0056] The fourth aspect of this embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps in the resource synchronization method described in the first aspect of this embodiment.

[0057] The fifth aspect of this embodiment also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps in the resource synchronization method as described in the first aspect of this embodiment.

[0058] This invention provides a resource synchronization method, a multi-cloud management system, an apparatus, a device, and a medium. The method includes: determining the target original resource mapped to the virtual resource of the target server based on the mapping relationship between the virtual resources of each server in a distributed server cluster and the original resources of the multi-cloud management platform; reading alarm data of the target original resource from the local historical database of the target server; obtaining the failure probability of the target original resource during resource synchronization based on the alarm data and a first early warning model, where the first early warning model is a model used to predict the failure of the original resource; and performing resource synchronization between the target server and the multi-cloud management platform when the failure probability is less than a preset threshold. This invention improves the success rate of resource synchronization by predicting the synchronization failure probability using alarm data and a first early warning model before resource synchronization, and selecting whether to perform resource synchronization based on the determined failure probability. Attached Figure Description

[0059] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This is a flowchart of the steps of a resource synchronization method provided in an embodiment of the present invention;

[0061] Figure 2 This is a schematic diagram of a multi-cloud management system provided in an embodiment of the present invention;

[0062] Figure 3 This is a schematic diagram of the structure of a resource synchronization device provided in an embodiment of the present invention;

[0063] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0064] Exemplary embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0065] To facilitate understanding of the technical solution of the present invention, a brief explanation of the relevant knowledge involved in the present invention will be given first.

[0066] A server cluster refers to a group of servers working together to provide the same service. From the client's perspective, it appears as a single server, but it is actually a loosely coupled collection of computing nodes consisting of two or more servers. Clustering can utilize multiple computers for parallel computing, achieving high computing speeds. It is a solution for improving the overall computing power of a server, and can also use multiple computers for backup. Through a parallel or distributed system composed of interconnected servers, the failure of any one machine ensures that the entire system continues to operate normally. Clustering reduces the number of single points of failure and achieves high availability of clustered resources.

[0067] Abstraction is a form of encapsulation, while virtualization is replication based on abstraction. Virtualization refers to abstracting one or more resources into one or more new resources through spatial partitioning, time-sharing, and abstract simulation. In a computer scenario, the objects of abstraction and virtualization are various hardware and software resources. Virtualization abstracts lower-level resources into a form of resource for use by higher-level layers. Virtualization abstracts one or more resources into one or more new resources through spatial partitioning, time-sharing, and abstract simulation. The abstracted resources can be physical or logical. Virtualization can be multi-layered or nested.

[0068] In various embodiments of the present invention, "resources" include software resources and hardware resources, including but not limited to VMs (virtual machines), hosts, disks, networks (including physical networks and virtual networks), firewalls, routers (including physical routers and virtual routers), floating IPs, public IPs, and other configuration files, interface files, ports, virtual machine image files, snapshots, disk data, hard disk capacity, CPU processing power, etc., used to define the cloud platform or the containers deployed on the cloud platform.

[0069] A multi-cloud management platform refers to a platform that manages the entire lifecycle of complex heterogeneous IT infrastructure resources. It includes functional modules such as resource management, operation management, and maintenance management, and comprehensively covers the digital transformation scenarios of government and enterprise customers in finance, energy, and telecommunications.

[0070] This invention provides a resource synchronization method applied to a target server in a distributed server cluster, as described above. Figure 1 , Figure 1 A flowchart illustrating the steps of a resource synchronization method provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method includes:

[0071] Step S101: Based on the mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform, determine the target original resources mapped to the virtual resources of the target server.

[0072] In this embodiment, the multi-cloud management platform manages all servers in the server cluster, playing a unified scheduling and control role over all resources. The multi-cloud management platform deploys original resources, which are abstracted and transformed into virtual resources through resource virtualization, and then synchronized to all servers in the distributed server cluster. This embodiment describes a method for synchronizing resources between a multi-cloud management platform and one or more servers in a distributed server cluster, where each server in the cluster is independent. The mapping relationship can represent the conversion relationship between original resources and virtual resources. Generally, the multi-cloud management platform virtualizes the original resources; specifically, FusionSphere can be used to implement open server virtualization in a cloud computing environment. During resource virtualization, the multi-cloud management platform generates the corresponding mapping relationship, which is calculated using the following formula:

[0073] Abstract mathematical expression: y = FA(x)

[0074] Where FA (Function for Abstract) is the abstract mapping function; x is the original resource being abstracted; and y is the abstracted virtual resource.

[0075] Virtualized mathematical expression: (y0, y1, y2……y n-1 ) = FV(x0,x1,x2……x m-1 )

[0076] Where FV (Function for Virtualization) is the virtualization mapping function; x0 to x m-1 y0 represents the original resource, m represents the quantity of the original resource; y0 to y n-1 Let n be the number of virtual resources.

[0077] For each y, establish an independent mapping relating to all x:

[0078] y j =FV j (x0,x1,x2……x m-1 )

[0079] Where j is any index of the virtualized resource y (0 ≤ j < n); FV j For the virtualization mapping function of y.

[0080] By splitting FV j We can obtain the virtualization mapping function for each x and each y:

[0081] y j =FV j i (x i ) = FA j i (x i )

[0082] Therefore, virtualization is a collection of single-resource abstractions, which can be represented as:

[0083] X = FA -1 (y)

[0084] Based on the above mapping relationship, virtual resources can be traced back to their original mapping resources. Specifically, one virtual resource may map to N original resources, and one original resource may map to M virtual resources. The multi-cloud management platform sends this mapping relationship to the target server, which then determines all the target original resources mapped to its own virtual resources based on the mapping relationship.

[0085] Step S102: Read the alarm data of the target original resource from the local historical database of the target server.

[0086] The alarm data is data stored in the local historical database of the target server, representing abnormal information about the target original resource, and may include historical synchronization data and synchronization failure data of the target original resource.

[0087] Step S103: Based on the alarm data of the target original resource and the first early warning model, obtain the failure probability of the target original resource during the resource synchronization process. The first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform.

[0088] This embodiment utilizes a first early warning model to predict the probability of failure of the original resource that needs to be synchronized. This enables the prediction of resource synchronization failure between the target server and the multi-cloud management platform, and determines the probability that the target original resource will change from a normal state to an abnormal state during the resource synchronization process at the current moment, i.e., the probability of synchronization failure or failure during the synchronization process.

[0089] In this embodiment, the formula for the first early warning model can be expressed as: X(k+1)=X(k)×P

[0090] Where X(k) represents the state vector of the target original resource at time t = k, and the state vector is a trend analysis and prediction for the target original resource. P represents the one-step transition probability matrix, and X(k+1) represents the state vector of the target original resource at time t = k+1. Specifically, the monitoring indicators of this first early warning model may include: the network device anomaly probability of the target original resource, the port network quality anomaly probability, and the host anomaly probability associated with the VLAN pool. Among them, the network device anomaly probability includes: memory anomaly probability, CPU anomaly probability, and disk size anomaly probability; the port network quality anomaly probability includes: packet loss probability and latency probability; the host anomaly probability associated with the VLAN pool includes: host memory anomaly probability, host CPU anomaly probability, and host disk size anomaly probability. Thus, the first early warning model can calculate the anomaly probability values ​​of the above indicators based on the alarm data, and then fit them through the weights of each indicator to obtain the comprehensive probability that the target original resource will fail.

[0091] In one embodiment, the method further includes:

[0092] An initial early warning model is obtained from the multi-cloud management platform. The initial early warning model is obtained by the multi-cloud management platform by training a preset model based on the alarm data of the original resources of the multi-cloud management platform.

[0093] In this embodiment, the multi-cloud management platform establishes an initial early warning model based on the alarm data of the original resources to monitor the status of the original resources. Then, the multi-cloud management platform sends the initial early warning model to each server in the server cluster.

[0094] Correspondingly, step S103, based on the alarm data of the target original resource and the first early warning model, obtains the failure probability of the target original resource failing during resource synchronization, including:

[0095] Step S1031: Based on the alarm data of the target original resource, the model parameters of the initial early warning model are corrected to obtain the first early warning model;

[0096] Step S1032: Input the alarm data of the target original resource into the first early warning model to obtain the fault probability.

[0097] Because of the differences between servers in a server cluster, a single early warning model (initial early warning model) cannot be directly used for probability prediction. This embodiment proposes to revise the model parameters of the initial early warning model based on alarm data stored in the target server's local database, and then use the revised first early warning model to perform fault probability prediction to obtain more accurate prediction results.

[0098] Step S104: If the failure probability is less than a preset threshold, perform resource synchronization between the target server and the multi-cloud management platform.

[0099] Before synchronizing resources between the server and the multi-cloud management platform, this embodiment of the invention uses alarm data and a first early warning model to predict the probability of synchronization failure. Based on the determined failure probability, it selects whether to perform resource synchronization. Specifically, if the calculated failure probability is less than a preset threshold, resource synchronization between the target server and the multi-cloud management platform is then performed. This can be achieved through the OpenAPI interface. Therefore, this embodiment achieves early warning of synchronization failures, improving the success rate of resource synchronization.

[0100] In one embodiment, the method further includes:

[0101] If the probability of failure is greater than the preset threshold, a failure message is generated and sent to the terminal corresponding to the target server.

[0102] In this embodiment, a preset probability threshold is determined based on historical data. When the failure probability output by the first early warning model exceeds this threshold, it indicates a high probability of failure during the current resource synchronization process. The synchronization can be abandoned, and corresponding failure information is generated and sent to the terminal corresponding to the target server, thereby notifying the relevant staff of the anomaly in the resource synchronization process. Furthermore, the analysis results for this failure can also be sent simultaneously. Specifically, the monitoring indicators of the first early warning model include: the probability of network device anomalies in the original resource, the probability of port network quality anomalies, and the probability of host anomalies associated with the VLAN pool. When sending failure information, the anomaly information for each indicator can be sent synchronously to assist staff in determining the cause of the synchronization failure and facilitate subsequent work.

[0103] In one embodiment, based on the alarm data of the target original resource, the model parameters of the initial early warning model are corrected to obtain a first early warning model, including:

[0104] Based on the alarm data of the target original resource, the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment are obtained, where the second moment is the adjacent moment after the first moment.

[0105] Based on the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment, the model parameters of the initial early warning model are corrected to obtain the first early warning model.

[0106] For example, based on the target original resource, the total data volume is 1, with abnormal alarm data accounting for 30% (0.3) and normal data accounting for 70% (0.7). This means the probability of the target original resource failing at the first moment is 0.3. Furthermore, analysis of the alarm data shows that 60% of the abnormal alarm data may remain abnormal, and 40% may return to normal. Therefore, the probability of the target original resource transitioning from failure to normal at the second moment is 0.6. Of the 70% normal data, 70% may remain normal, while 30% may become abnormal alarms. Therefore, the probability of the target original resource transitioning from normal to failure at the second moment is 0.3. Thus, the probability of the next abnormal alarm data occurrence is calculated as P1 = 0.3 x 0.6 + 0.3 x 0.7 = 0.39 x 100 = 39%; the probability of the next normal data occurrence is P2 = 3 x 0.4 + 7 x 0.7 = 0.61 x 100 = 61%. Therefore, by modifying the model parameters of the initial early warning model based on the probability values ​​calculated from the alarm data, the parameters in the one-step transition probability matrix of the initial early warning model formula can be modified, thus obtaining the first early warning model.

[0107] In one embodiment, after determining the target original resource mapped to the virtual resource of the target server, the method further includes:

[0108] Other servers that are associated with the original resources mapped by the virtual resources and the target original resources are identified as the associated servers of the target server.

[0109] Request the associated server to read the alarm data of the original resources that are associated from the local historical database of the associated server;

[0110] Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during resource synchronization is obtained, including:

[0111] Based on the alarm data of the target original resource and the alarm data of the related original resources, the model parameters of the initial early warning model are corrected to obtain the first early warning model;

[0112] The alarm data of the target original resource is input into the first early warning model to obtain the fault probability.

[0113] In this embodiment, an associated server refers to a server whose virtual resource mapping is associated with the original resources of the target server. For example, the original resources mapped to the virtual resources of target server A include a portion of the hard drive capacity of hard drive B, while the original resources mapped to the virtual resources of server C also include another portion of the hard drive capacity of hard drive B. Therefore, there is a certain correlation between the virtual resources of target server A and server C. During the resource synchronization process of target server A, the state of the associated original resources involved in server C needs to be considered. This embodiment proposes that when correcting the initial warning model, alarm data from the target server and associated servers are used to correct the model parameters. This ensures that the corrected warning model can fully consider the impact of associated servers on the resource synchronization of the target server, thereby making the obtained fault probability prediction results more accurate and reliable.

[0114] In one embodiment, after correcting the model parameters of the initial early warning model to obtain the first early warning model, the method further includes:

[0115] The model parameters of the first early warning model are sent to the multi-cloud management platform so that the multi-cloud management platform can update the initial early warning model based on the model parameters sent by each server in the distributed server cluster.

[0116] The updated initial warning model is obtained from the multi-cloud management platform for use in the next resource synchronization between the target server and the multi-cloud management platform.

[0117] In this embodiment, each server in the distributed server cluster uses alarm data from its local database to correct the parameters of the initial early warning model. The resulting first early warning model parameters are then resent to the multi-cloud management platform. The multi-cloud management platform can then update the initial early warning model based on the parameters sent by each server, and use the updated initial early warning model for the next resource synchronization. This achieves iterative optimization of the early warning model. Through distributed model training, while synchronizing resources between the servers and the multi-cloud management platform, the privacy of local server data is protected from leakage. Furthermore, distributed model training and parameter correction significantly reduce the pressure on various local data to be uniformly trained through the initial early warning model on the multi-cloud management platform servers.

[0118] This embodiment also provides a multi-cloud management system, referencing... Figure 2 , Figure 2 The diagram illustrates the distributed architecture of a multi-cloud management system, such as... Figure 2 As shown, the multi-cloud management system includes a multi-cloud management platform and a distributed server cluster;

[0119] The multi-cloud management platform is used to create a mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform, and to send the mapping relationship to each server in the server cluster.

[0120] Each server in the distributed server cluster is used to determine the target original resource mapped by the virtual resource of the server according to the received mapping relationship; read the alarm data of the target original resource from the local historical database; obtain the failure probability of the target original resource during the resource synchronization process based on the alarm data of the target original resource and the first early warning model, wherein the first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform; and perform resource synchronization with the multi-cloud management platform when the failure probability is less than a preset threshold.

[0121] This embodiment also provides a resource synchronization device, referring to... Figure 3 , Figure 3 A schematic diagram of a resource synchronization device is shown, such as... Figure 3 As shown, the device includes:

[0122] The determination module is used to determine the target original resource mapped to the virtual resource of the target server based on the mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform.

[0123] The reading module is used to read the alarm data of the target's original resources from the local historical database of the target server;

[0124] The fault probability determination module is used to obtain the fault probability of the target original resource failing during the resource synchronization process based on the alarm data of the target original resource and the first early warning model. The first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform.

[0125] The synchronization module is used to synchronize resources between the target server and the multi-cloud management platform when the failure probability is less than a preset threshold.

[0126] In one embodiment, the device further includes:

[0127] The model acquisition module is used to acquire an initial early warning model from the multi-cloud management platform. The initial early warning model is obtained by the multi-cloud management platform by training a preset model based on the alarm data of the original resources of the multi-cloud management platform.

[0128] The fault probability determination module includes:

[0129] The first correction submodule is used to correct the model parameters of the initial early warning model based on the alarm data of the target original resource to obtain the first early warning model;

[0130] The first input submodule is used to input the alarm data of the target original resource into the first early warning model to obtain the fault probability.

[0131] In one embodiment, the first correction submodule includes:

[0132] The first correction unit is used to obtain, based on the alarm data of the target original resource, the failure probability of the target original resource at a first moment, the probability of the target original resource transitioning from normal to failure at a second moment, and the probability of the target original resource transitioning from failure to normal at a second moment, wherein the second moment is an adjacent moment after the first moment.

[0133] The second correction unit is used to correct the model parameters of the initial early warning model based on the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment, so as to obtain the first early warning model.

[0134] In one embodiment, the device further includes:

[0135] The associated server determination module is used to determine other servers that are associated with the target original resource and the original resource mapped by the virtual resource as associated servers of the target server.

[0136] The associated alarm data reading module is used to request the associated server to read the alarm data of the original resources that are associated from the local historical database of the associated server;

[0137] The fault probability determination module includes:

[0138] The second correction submodule is used to correct the model parameters of the initial early warning model based on the alarm data of the target original resource and the alarm data of the related original resources, so as to obtain the first early warning model.

[0139] The second input submodule is used to input the alarm data of the target original resource into the first early warning model to obtain the fault probability.

[0140] In one embodiment, the device further includes:

[0141] The update module is used to send the model parameters of the first early warning model to the multi-cloud management platform, so that the multi-cloud management platform updates the initial early warning model according to the model parameters sent by each server in the distributed server cluster.

[0142] The acquisition module is used to acquire the updated initial warning model from the multi-cloud management platform for the next resource synchronization between the target server and the multi-cloud management platform.

[0143] In one embodiment, the device further includes:

[0144] The fault warning module is used to generate and send fault information to the terminal corresponding to the target server when the fault probability is greater than the preset threshold.

[0145] This invention also provides an electronic device, with reference to... Figure 4 , Figure 4 This is a schematic diagram of the structure of the electronic device proposed in an embodiment of the present invention. Figure 4 As shown, the electronic device 100 includes a memory 110 and a processor 120. The memory 110 and the processor 120 are connected via a bus for communication. The memory 110 stores a computer program that can run on the processor 120 to implement the steps in the resource synchronization method disclosed in the embodiments of the present invention.

[0146] This invention also provides a computer-readable storage medium storing a computer program / instructions thereon, which, when executed by a processor, implements the steps of a resource synchronization method disclosed in this invention.

[0147] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0148] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, apparatuses, electronic devices, and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0149] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0150] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0151] The resource synchronization method, multi-cloud management system, apparatus, device, and medium provided by the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A resource synchronization method, characterized in that, The method, applied to a target server in a distributed server cluster, includes: Based on the mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform, the target original resources mapped to the virtual resources of the target server are determined. Read the alarm data of the original target resource from the local historical database of the target server; Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during the resource synchronization process is obtained. The first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform. If the failure probability is less than a preset threshold, resource synchronization is performed between the target server and the multi-cloud management platform.

2. The resource synchronization method according to claim 1, characterized in that, The method further includes: An initial early warning model is obtained from the multi-cloud management platform. The initial early warning model is obtained by the multi-cloud management platform by training a preset model based on the alarm data of the original resources of the multi-cloud management platform. Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during resource synchronization is obtained, including: Based on the alarm data of the target's original resources, the model parameters of the initial early warning model are corrected to obtain the first early warning model; The alarm data of the target original resource is input into the first early warning model to obtain the fault probability.

3. The resource synchronization method according to claim 2, characterized in that, Based on the alarm data of the target's original resources, the model parameters of the initial early warning model are corrected to obtain a first early warning model, including: Based on the alarm data of the target original resource, the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment are obtained, where the second moment is the adjacent moment after the first moment. Based on the failure probability of the target original resource at the first moment, the probability of the target original resource transitioning from normal to failure at the second moment, and the probability of the target original resource transitioning from failure to normal at the second moment, the model parameters of the initial early warning model are corrected to obtain the first early warning model.

4. The resource synchronization method according to claim 2, characterized in that, After determining the target original resource mapped to the virtual resource of the target server, the process further includes: Other servers that are associated with the original resources mapped by the virtual resources and the target original resources are identified as the associated servers of the target server. Request the associated server to read the alarm data of the original resources that are associated from the local historical database of the associated server; Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during resource synchronization is obtained, including: Based on the alarm data of the target original resource and the alarm data of the related original resources, the model parameters of the initial early warning model are corrected to obtain the first early warning model; The alarm data of the target original resource is input into the first early warning model to obtain the fault probability.

5. The resource synchronization method according to any one of claims 2-4, characterized in that, After correcting the model parameters of the initial early warning model to obtain the first early warning model, the process further includes: The model parameters of the first early warning model are sent to the multi-cloud management platform so that the multi-cloud management platform can update the initial early warning model based on the model parameters sent by each server in the distributed server cluster. The updated initial warning model is obtained from the multi-cloud management platform for use in the next resource synchronization between the target server and the multi-cloud management platform.

6. The resource synchronization method according to claim 1, characterized in that, The method further includes: If the probability of failure is greater than the preset threshold, a failure message is generated and sent to the terminal corresponding to the target server.

7. A multi-cloud management system, characterized in that, The multi-cloud management system includes a multi-cloud management platform and a distributed server cluster; The multi-cloud management platform is used to create a mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform, and to send the mapping relationship to each server in the server cluster. Each server in the distributed server cluster is used to determine the target original resource mapped by the virtual resource of the server according to the received mapping relationship. Read the alarm data of the target original resource from the local historical database; Based on the alarm data of the target original resource and the first early warning model, the failure probability of the target original resource failing during resource synchronization is obtained. The first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform. If the failure probability is less than a preset threshold, resource synchronization with the multi-cloud management platform is performed.

8. A resource synchronization device, characterized in that, The device includes: The determination module is used to determine the target original resource mapped to the virtual resource of the target server based on the mapping relationship between the virtual resources of each server in the distributed server cluster and the original resources of the multi-cloud management platform. The reading module is used to read the alarm data of the target's original resources from the local historical database of the target server; The fault probability determination module is used to obtain the fault probability of the target original resource failing during the resource synchronization process based on the alarm data of the target original resource and the first early warning model. The first early warning model is a model used to predict the failure of the original resource of the multi-cloud management platform. The synchronization module is used to synchronize resources between the target server and the multi-cloud management platform when the failure probability is less than a preset threshold.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the resource synchronization method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps in the resource synchronization method as described in any one of claims 1 to 6.

Images