Cluster node fault prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a cluster node fault prediction method and device, electronic equipment and a storage medium, and relates to the technical field of data processing, and the main scheme comprises the following steps: collecting operation data of a target node in a cluster; inputting the operation data into a fault prediction model to obtain a fault probability corresponding to the target node; and when it is determined that the fault probability is greater than a preset fault threshold, determining that the target node has a fault. Compared with the prior art, the fault probability that the target node has the fault is predicted based on the fault prediction model according to the operation data generated by the target node, and when it is determined that the fault probability is larger than the preset fault threshold value, it is determined that the target node has the fault. Therefore, the fault of the target node is predicted, and the target node which may have a fault is repaired in advance by predicting the fault possibility of the target node in advance, so that the stability of node service operation is ensured.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a cluster node failure prediction method and device, electronic equipment, and storage medium. Background Art

[0002] Kubernetes, also known as K8s, is an open-source platform for managing containerized applications. A Kubernetes cluster is a distributed system consisting of multiple nodes, each of which hosts a container runtime for deploying and running containerized applications. Node failure detection is crucial in a Kubernetes cluster, as node failures can cause application interruptions or failures, resulting in service unavailability and a degraded user experience. Summary of the Invention

[0003] This disclosure provides a cluster node failure prediction method and device, electronic device, and storage medium. Its main purpose is to achieve early prediction of node failures in a Kubernetes cluster to ensure the stability of service operation.

[0004] According to a first aspect of the present disclosure, a method for predicting cluster node failure is provided, comprising:

[0005] Collect the running data of the target node in the cluster;

[0006] Inputting the operating data into a fault prediction model to obtain a fault probability corresponding to the target node;

[0007] When it is determined that the failure probability is greater than a preset failure threshold, it is determined that a failure exists in the target node.

[0008] Optionally, after determining that the target node has a fault, the method further includes:

[0009] determining a fault level for the fault according to a preset fault level rule;

[0010] Output a corresponding early warning prompt according to the fault level.

[0011] Optionally, after determining that the target node has a fault, the method includes:

[0012] Collecting operating data of all nodes in the cluster within a preset time period;

[0013] Outputting the failure probabilities corresponding to the operating data of all nodes according to the fault prediction model, and dividing the operating data of all nodes into failure samples or non-fault samples;

[0014] The fault prediction model is trained by regression using the fault samples or non-fault samples.

[0015] Optionally, training the fault prediction model includes:

[0016] Obtaining original operation data, wherein the original operation data includes various indicator parameters and log data when the target node is running;

[0017] Preprocessing the original operating data to obtain training operating data;

[0018] The fault prediction model is trained based on the training operation data until it is determined that the output result of the fault prediction model meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0019] Optionally, after preprocessing the training data, the method further includes:

[0020] Dividing the training running data into a test set and a training set according to a preset ratio;

[0021] The training of the fault prediction model based on the training operation data includes:

[0022] Training the fault prediction model based on the training operation data in the training set;

[0023] Performing an accuracy test on the trained fault prediction model based on the training operation data in the test set;

[0024] When it is determined that the test result does not meet the preset accuracy condition, the fault prediction model continues to be trained until the test result meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0025] Optionally, preprocessing the original operating data to obtain training operating data includes:

[0026] Searching for data to be deleted in the original operating data, where the data to be deleted includes at least one of the following: abnormal data, missing data, and standardized data;

[0027] Execute deletion processing on the data to be deleted.

[0028] According to a second aspect of the present disclosure, a device for predicting cluster node failure is provided, comprising:

[0029] A first collection unit, configured to collect operating data of a target node in the cluster;

[0030] An input unit, configured to input the operating data into a fault prediction model to obtain a fault probability corresponding to the target node;

[0031] The first determining unit is configured to determine that a fault exists in the target node when it is determined that the fault probability is greater than a preset fault threshold.

[0032] Optionally, the device further includes:

[0033] a second determining unit, configured to determine a fault level for the fault according to a preset fault level rule;

[0034] The output unit is used to output a corresponding early warning prompt according to the fault level.

[0035] Optionally, the device includes:

[0036] A second collection unit is used to collect operating data of all nodes in the cluster within a preset time period;

[0037] A first dividing unit is configured to output the fault probabilities corresponding to the operation data of all nodes according to the fault prediction model, and divide the operation data of all nodes into fault samples or non-fault samples;

[0038] The first training unit is configured to perform regression training on the fault prediction model using the fault samples or non-fault samples.

[0039] Optionally, the device further includes:

[0040] An acquisition unit, configured to acquire original operation data, wherein the original operation data includes various indicator parameters and log data when the target node is running;

[0041] a processing unit, configured to preprocess the original operating data to obtain operating data for training;

[0042] The second training unit is used to train the fault prediction model based on the training operation data until it is determined that the output result of the fault prediction model meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0043] Optionally, the device further includes:

[0044] A second division unit is used to divide the training operation data into a test set and a training set according to a preset ratio;

[0045] The second training unit is further configured to:

[0046] Training the fault prediction model based on the training operation data in the training set;

[0047] Performing an accuracy test on the trained fault prediction model based on the training operation data in the test set;

[0048] When it is determined that the test result does not meet the preset accuracy condition, the fault prediction model continues to be trained until the test result meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0049] Optionally, the processing unit includes:

[0050] a search module, configured to search for data to be deleted in the original operation data, wherein the data to be deleted includes at least one of the following: abnormal data, missing data, and standardized data;

[0051] The deletion module is used to perform deletion processing on the data to be deleted.

[0052] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0053] at least one processor; and

[0054] a memory communicatively connected to the at least one processor; wherein,

[0055] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect.

[0056] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described in the first aspect.

[0057] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method as described in the first aspect above.

[0058] The present disclosure provides a cluster node failure prediction method and device, electronic device, and storage medium. The method collects operating data of a target node in a cluster; inputs the operating data into a fault prediction model to obtain a corresponding failure probability of the target node; and determines that a fault exists in the target node when the failure probability is determined to be greater than a preset failure threshold. Compared to related technologies, the method predicts the failure probability of the target node based on the operating data generated by the target node and the fault prediction model. When the failure probability is determined to be greater than the preset failure threshold, the target node is determined to be faulty, thereby achieving a prediction of a failure of the target node. By predicting the possibility of a failure of the target node in advance, the target node that may fail can be repaired in advance, thereby ensuring the stability of node service operation.

[0059] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0061] Figure 1 A schematic diagram of a process for predicting cluster node failures provided by an embodiment of the present disclosure;

[0062] Figure 2 A schematic diagram of the structure of a cluster node failure prediction device provided by an embodiment of the present disclosure;

[0063] Figure 3 A schematic diagram of another cluster node failure prediction device provided by an embodiment of the present disclosure

[0064] Figure 4 A schematic block diagram of an exemplary electronic device 300 provided in accordance with an embodiment of the present disclosure. DETAILED DESCRIPTION

[0065] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0066] The following describes a cluster node failure prediction method, apparatus, electronic device, and storage medium according to embodiments of the present disclosure with reference to the accompanying drawings.

[0067] Figure 1 A schematic flow chart of a cluster node failure prediction method provided by an embodiment of the present disclosure.

[0068] like Figure 1 As shown, the method comprises the following steps:

[0069] Step 101: Collect operating data of target nodes in the cluster.

[0070] As a refinement of step 101 above, it is important to understand that the cluster is a distributed system composed of multiple target nodes, each of which is a computer entity. To observe the operating status of each target node in the cluster, operating data of the target node is collected. This operating data includes, but is not limited to, metrics such as CPU utilization, memory utilization, disk input and output, network bandwidth, Pod response time, request failure rate, and data in various log files. The Pod is the smallest unit of K8s scheduling. A service consists of one or more Pods, and a Pod corresponds to a specific set of computing tasks on a computer.

[0071] Step 102: Input the operating data into a fault prediction model to obtain a fault probability corresponding to the target node.

[0072] As a refinement of the above-mentioned step 102, in order to predict the failure probability of the target node, the operating data obtained by step 101 is input into the fault prediction model, and the fault prediction model is obtained based on the training of the deep learning model, and the fault prediction model can predict the failure probability of the target node corresponding to the operating data based on the operating data.

[0073] In a possible implementation of this embodiment, the value of the failure probability is 0%-100%.

[0074] Step 103: When it is determined that the failure probability is greater than a preset failure threshold, it is determined that a failure exists in the target node.

[0075] As a refinement of step 103 above, in order to determine whether the target node is about to fail based on the failure probability, the relationship between the failure probability and the preset failure threshold is determined. The preset failure threshold is a probability threshold set based on actual scenarios. This embodiment does not limit the value range of the preset failure threshold. When it is determined that the failure probability is greater than the preset failure threshold, it is determined that the target object is about to fail. The above description is merely exemplary and does not constitute a limitation of the present disclosure.

[0076] The present disclosure provides a cluster node failure prediction method that collects operating data of a target node in a cluster; inputs the operating data into a fault prediction model to obtain a failure probability corresponding to the target node; and determines that the target node is faulty if the failure probability is greater than a preset failure threshold. Compared to related technologies, the method predicts the failure probability of the target node based on the operating data generated by the target node and the fault prediction model; and determines that the target node is faulty if the failure probability is greater than the preset failure threshold, thereby achieving prediction of the target node failure. By predicting the possibility of the target node failure in advance, the target node that may fail can be repaired in advance, thereby ensuring the stability of node service operation.

[0077] As a refinement of the above embodiment, after determining that there is a fault in the target node, the method can also adopt but is not limited to the following implementation methods, for example: determining the fault level of the fault according to preset fault level rules; outputting a corresponding early warning prompt according to the fault level.

[0078] In order to achieve a reasonable allocation of early warning resources, different early warning prompts are allocated according to the severity of the fault, specifically including: determining the fault level of the fault according to the preset fault level rules, the preset fault level rules can be based on the fault probability division, based on the location of the fault in the target node as hardware or software division, based on the importance of the service undertaken by the target node, etc., which is not limited in this embodiment. The early warning prompt corresponding to the determined fault level is output, that is, different prompt content is output according to the fault level, and different early warning prompt methods are used. The early warning prompt methods include: one or more methods of telephone, text message and instant messaging software for early warning prompts. The above description of the early warning prompt method is only exemplary, and this embodiment does not limit it.

[0079] As a refinement of the above embodiment, when the failure probability is determined to be greater than or equal to the preset failure threshold, the target node is determined to be non-failure-prone. After completing the prediction of whether the target node will fail, the operating data generated by the target node in the next time period is collected, and then the determination of whether the target node has failed is again made. In one possible implementation of this embodiment, the time period is 30 minutes. The above description is merely illustrative, and this embodiment does not limit the time period.

[0080] As a refinement of the above embodiment, after determining that the target node has a fault, the method can also adopt but is not limited to the following implementation methods, for example: collecting the operating data of all nodes in the cluster within a preset time period; outputting the failure probabilities corresponding to the operating data of all nodes according to the fault prediction model, and dividing the operating data of all nodes into fault samples or non-fault samples; and using the fault samples or non-fault samples to perform regression training on the fault prediction model.

[0081] In order to achieve continuous iterative training of the fault prediction model after deployment to improve the accuracy of the prediction results of the fault prediction model, this embodiment provides a scenario description, collecting the operating data of all nodes in the cluster within 24 hours, and the 24 hours is the preset time period. The collected operating data is used as a training sample for training the fault prediction model. Before using the operating data to train the fault prediction model, it is necessary to output the failure probabilities corresponding to the operating data of all nodes according to the fault prediction model, and divide the operating data of the nodes into fault samples or non-fault samples, that is, based on the fault prediction model, the failure probability of a node is predicted according to the operating data corresponding to the node. If it is determined that the node will not fail, it is a non-fault sample, otherwise it is a fault sample.

[0082] As a refinement of the above embodiment, in order to obtain the fault prediction model, a deep learning model needs to be trained, and the adopted method includes but is not limited to the following implementation methods, for example: obtaining original operation data, the original operation data including various indicator parameters and log data during the operation of the target node; preprocessing the original operation data to obtain training operation data; training the fault prediction model based on the training operation data until it is determined that the output result of the fault prediction model meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0083] The original operation data is the training data for training the fault prediction model. This embodiment aims to obtain the fault prediction model by constructing a neural network model and training the constructed neural network model. The original operation data includes but is not limited to indicators such as CPU utilization, memory utilization, disk input and output, network bandwidth, Pod response time, request failure rate, data in various log files, etc. The training operation data is obtained by performing enhancement processing on the original operation data, and the fault prediction model is trained based on the training operation data to improve the quality of the fault prediction model obtained by training. In the process of training the fault prediction model, it is necessary to test the accuracy of the output result of the fault prediction model, and obtain the trained fault prediction model after determining that the output result of the fault prediction model meets the preset accuracy condition. The preset accuracy condition can be an accuracy threshold.

[0084] In a possible implementation of this embodiment, the deep learning algorithm may be, but is not limited to, a decision tree, a random forest, a support vector machine, or the like.

[0085] As a refinement of the above embodiment, after preprocessing the training data, the method can also adopt but is not limited to the following implementation methods, for example: dividing the training operation data into a test set and a training set according to a preset ratio; the training of the fault prediction model based on the training operation data includes: training the fault prediction model based on the training operation data in the training set; performing an accuracy test on the trained fault prediction model based on the training operation data in the test set; when it is determined that the test result does not meet the preset accuracy condition, continuing to train the fault prediction model until the test result meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0086] As a refinement of the above embodiment, in order to more clearly illustrate the process of training the fault prediction model, the process includes: preprocessing the original operating data to obtain the training operating data, dividing the training operating data into a test set and a training set according to a preset ratio, the training operating data in the test set is used to test the accuracy of the output result of the fault prediction model, and the training operating data in the training set is used to train the fault prediction model, and the division of the training operating data is based on a preset ratio. For example, the ratio of the training operating data in the test set to the training operating data in the training set is 2:8. The above ratio is only exemplary, and this embodiment does not limit the selection of the preset ratio. The training of the fault prediction model based on the training operating data includes: training the fault prediction model based on the training operating data in the training set; performing an accuracy test on the trained fault prediction model based on the training operating data in the test set; when it is determined that the test result does not meet the preset accuracy condition, continuing to train the fault prediction model until the test result meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0087] As a refinement of the above embodiment, when preprocessing the original operating data to obtain operating data for training, the following implementation methods can be adopted but are not limited to, for example: searching for data to be deleted in the original operating data, the data to be deleted including at least one of the following: abnormal data, missing data and standardized data; performing deletion processing on the data to be deleted.

[0088] In order to improve the quality of the original operation data and thus improve the quality of the fault prediction model obtained by training, abnormal data, missing data, and standardized data in the original operation data are deleted to enhance the original operation data.

[0089] In summary, this embodiment can achieve the following effects:

[0090] 1. By predicting the possibility of failure of the target node in advance, the target node that may fail is repaired in advance, thereby ensuring the stability of node service operation.

[0091] 2. Effectively provide early warning of node failures, reduce the possibility of node failures, and reduce economic losses.

[0092] 3. Through machine learning algorithms, various types of collected data are labeled and trained to obtain a fault prediction model. Based on the prediction results of the fault prediction model, node fault warning information is issued.

[0093] 4. By continuously collecting and iterating operational data, the accuracy of the fault prediction model is improved, forming a closed-loop platform. Efficient design is made in terms of usage, achieving automatic and continuous updates and iterations, and having high practical value.

[0094] Corresponding to the aforementioned cluster node failure prediction method, the present invention also provides a cluster node failure prediction device. Since the device embodiment of the present invention corresponds to the aforementioned method embodiment, details not disclosed in the device embodiment can be referred to the aforementioned method embodiment and will not be further described in this invention.

[0095] Figure 2 A schematic diagram of a cluster node failure prediction device provided by an embodiment of the present disclosure is shown in FIG. Figure 2 Shown, including:

[0096] The first collection unit 21 is used to collect the operating data of the target node in the cluster;

[0097] An input unit 22, configured to input the operating data into a fault prediction model to obtain a fault probability corresponding to the target node;

[0098] The first determining unit 23 is configured to determine that a fault exists in the target node when it is determined that the fault probability is greater than a preset fault threshold.

[0099] The present disclosure provides a cluster node failure prediction device that collects operating data of a target node in a cluster; inputs the operating data into a fault prediction model to obtain a failure probability corresponding to the target node; and determines that the target node has a fault if the failure probability is greater than a preset fault threshold. Compared to related technologies, the present disclosure predicts the failure probability of the target node based on the operating data generated by the target node and the fault prediction model; and determines that the target node has a fault if the failure probability is greater than the preset fault threshold, thereby achieving a prediction of the target node's failure. By predicting the possibility of the target node's failure in advance, the target node that may have a fault can be repaired in advance, thereby ensuring the stability of node service operation.

[0100] Figure 3 A schematic diagram of the structure of another cluster node failure prediction device provided by an embodiment of the present disclosure, such as Figure 3 As shown, the device also includes:

[0101] A second determining unit 24 is configured to determine a fault level for the fault according to a preset fault level rule;

[0102] The output unit 25 is configured to output a corresponding warning prompt according to the fault level.

[0103] Furthermore, in a possible implementation of this embodiment, as Figure 3 As shown, the device includes:

[0104] A second collection unit 26 is configured to collect operating data of all nodes in the cluster within a preset time period;

[0105] A first dividing unit 27 is configured to output the fault probabilities corresponding to the operation data of all nodes according to the fault prediction model, and divide the operation data of all nodes into fault samples or non-fault samples;

[0106] The first training unit 28 is configured to perform regression training on the fault prediction model using the fault samples or non-fault samples.

[0107] Furthermore, in a possible implementation of this embodiment, as Figure 3 As shown, the device also includes:

[0108] An acquisition unit 29 is configured to acquire original operation data, wherein the original operation data includes various indicator parameters and log data when the target node is running;

[0109] The processing unit 210 is configured to pre-process the original operation data to obtain training operation data;

[0110] The second training unit 211 is configured to train the fault prediction model based on the training operation data until it is determined that the output result of the fault prediction model meets a preset accuracy condition, thereby obtaining the trained fault prediction model.

[0111] Furthermore, in a possible implementation of this embodiment, as Figure 3 As shown, the device also includes:

[0112] A second division unit 212 is configured to divide the training operation data into a test set and a training set according to a preset ratio;

[0113] The second training unit 211 is further configured to:

[0114] Training the fault prediction model based on the training operation data in the training set;

[0115] Performing an accuracy test on the trained fault prediction model based on the training operation data in the test set;

[0116] When it is determined that the test result does not meet the preset accuracy condition, the fault prediction model continues to be trained until the test result meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

[0117] Furthermore, in a possible implementation of this embodiment, as Figure 3 As shown, the processing unit 210 includes:

[0118] A search module 2101 is configured to search for data to be deleted in the original operation data, where the data to be deleted includes at least one of the following: abnormal data, missing data, and standardized data;

[0119] The deletion module 2102 is configured to execute deletion processing on the data to be deleted.

[0120] It should be noted that the above explanation of the method embodiment is also applicable to the device of this embodiment, and the principles are the same, which is not limited in this embodiment.

[0121] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0122] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0123] like Figure 4 As shown, the device 300 includes a computing unit 301, which can perform various appropriate actions and processes according to a computer program stored in a ROM (Read-Only Memory) 302 or a computer program loaded from a storage unit 308 into a RAM (Random Access Memory) 303. Various programs and data required for the operation of the device 300 can also be stored in the RAM 303. The computing unit 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An I / O (Input / Output) interface 305 is also connected to the bus 304.

[0124] Various components in device 300 are connected to I / O interface 305, including: an input unit 306, such as a keyboard, mouse, etc.; an output unit 307, such as various types of displays, speakers, etc.; a storage unit 308, such as a magnetic disk, optical disk, etc.; and a communication unit 309, such as a network card, modem, wireless communication transceiver, etc. The communication unit 309 allows device 300 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0125] Computing unit 301 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of computing unit 301 include, but are not limited to, a CPU (Central Processing Unit), a GPU (Graphic Processing Unit), various specialized AI (Artificial Intelligence) computing chips, various computing units that run machine learning model algorithms, a DSP (Digital Signal Processor), and any suitable processor, controller, microcontroller, etc. Computing unit 301 performs the various methods and processes described above, such as the cluster node failure prediction method. For example, in some embodiments, the cluster node failure prediction method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 308. In some embodiments, part or all of the computer program can be loaded and / or installed onto device 300 via ROM 302 and / or communication unit 309. When the computer program is loaded into RAM 303 and executed by computing unit 301, one or more steps of the method described above can be performed. Alternatively, in other embodiments, the computing unit 301 may be configured to execute the aforementioned cluster node failure prediction method in any other appropriate manner (for example, by means of firmware).

[0126] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application-Specific Standard Products), SOCs (System on Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0127] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0128] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, RAM, ROM, EPROM (Electrically Programmable Read-Only-Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0129] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0130] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: LAN (Local Area Network), WAN (Wide Area Network), the Internet, and blockchain networks.

[0131] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact via a communication network. This client-server relationship is established by computer programs running on the respective computers, establishing a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host, a host product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosts and VPS services ("Virtual Private Servers" or simply "VPS"). The server may also be a server in a distributed system or a server integrated with blockchain.

[0132] It's important to note that artificial intelligence (AI) is the study of how computers can simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). This encompasses both hardware and software technologies. AI hardware technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily encompass computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graphs.

[0133] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0134] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A cluster node failure prediction method, characterized in that: include: Collect the running data of the target nodes in the cluster; Inputting the operation data into a fault prediction model to obtain a fault probability corresponding to the target node; When it is determined that the failure probability is greater than a preset failure threshold, it is determined that a failure exists in the target node.

2. The method according to claim 1, characterized in that After determining that the target node has a fault, the method further includes: Determine a fault level for the fault according to a preset fault level rule; Output a corresponding early warning prompt according to the fault level.

3. The method according to claim 1, characterized in that After determining that the target node has a fault, the method includes: Collecting operation data of all nodes in the cluster within a preset time period; Outputting the failure probabilities corresponding to the operation data of all nodes according to the fault prediction model, and dividing the operation data of all nodes into failure samples or non-fault samples; The fault prediction model is trained by regression using the fault samples or non-fault samples.

4. The method according to claim 1, characterized in that Training the fault prediction model includes: Acquire original operation data, wherein the original operation data includes various indicator parameters and log data when the target node is running; Preprocessing the original operation data to obtain training operation data; The fault prediction model is trained based on the training operation data until it is determined that the output result of the fault prediction model meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

5. The method according to claim 4, characterized in that After preprocessing the training data, the method further includes: Dividing the training operation data into a test set and a training set according to a preset ratio; The training of the fault prediction model based on the training operation data includes: Training the fault prediction model based on the training operation data in the training set; Performing an accuracy test on the trained fault prediction model based on the training operation data in the test set; When it is determined that the test result does not meet the preset accuracy condition, the fault prediction model continues to be trained until the test result meets the preset accuracy condition, thereby obtaining the trained fault prediction model.

6. The method according to claim 4, characterized in that The preprocessing of the original operation data to obtain the operation data for training includes: Searching for data to be deleted in the original running data, wherein the data to be deleted includes at least one of the following: abnormal data, missing data, and standardized data; A deletion process is performed on the data to be deleted.

7. A cluster node failure prediction device, characterized in that: include: A first collection unit, used to collect operation data of a target node in the cluster; An input unit, used to input the operation data into a fault prediction model to obtain a fault probability corresponding to the target node; The first determining unit is configured to determine that a fault exists in the target node when it is determined that the fault probability is greater than a preset fault threshold.

8. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-6.

10. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images