Cloud resource association relationship mining method, device, equipment and medium

Abstract

The application provides a cloud resource association relationship mining method, device, equipment and medium, which can be used in the field of computer technology. The method comprises the following steps: obtaining a plurality of target alarm data in a target system, classifying the plurality of target alarm data to obtain a horizontal alarm data set, performing an inverse processing on the horizontal alarm data set, storing the horizontal alarm data set after the inverse processing by bits to obtain a vertical alarm data set, performing data mining processing on the vertical alarm data set to obtain a plurality of first frequent item sets, performing data mining processing on the plurality of first frequent item sets according to a priori principle to obtain a plurality of target frequent item sets, determining the association relationship between a plurality of cloud resources in the target system according to a plurality of device identifiers in the plurality of target frequent item sets, and storing the association relationship in a configuration management database; thereby improving the accuracy and efficiency of mining the association relationship between the cloud resources.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, device, and medium for mining cloud resource association relationships. Background Technology

[0002] With the rapid development of cloud computing, cloud architecture is increasingly being used in large-scale systems. Cloud architecture involves multiple layers such as service layer, platform layer, component layer, network layer, and host hardware layer, which makes understanding the asset configuration relationship of the current system extremely complex, resulting in the inability to accurately and quickly find the relationship between Internet information resources.

[0003] Existing methods for mining cloud resource relationships involve operations and maintenance personnel manually entering information about resources with relationships, or by traversing the configuration management database (CMDB) in the resource instance backend and comparing the fields of all resources that need to be associated, and associating resources with the same fields.

[0004] However, existing methods for mining relationships based on manual data entry suffer from inaccuracies in manually entered data; methods for mining relationships based on resource instance backend equivalence judgment rely on the rationality of CMDB model field design and are inefficient when the number of resources is large. Summary of the Invention

[0005] This application provides a method, apparatus, device, and medium for mining cloud resource association relationships, in order to solve the problems of low accuracy and low mining efficiency in existing cloud resource association relationship mining methods.

[0006] Firstly, this application provides a method for mining cloud resource association relationships, the method comprising:

[0007] Acquire multiple target alarm data within the target system, and classify the multiple target alarm data to obtain a horizontal alarm dataset. The target alarm data includes device identifiers and historical alarm time identifiers corresponding to multiple cloud resources within the target system.

[0008] The horizontal alarm dataset is inverted and then stored bitwise to obtain a vertical alarm dataset, which includes a subset of historical alarm time identifiers corresponding to each cloud resource.

[0009] Data mining is performed on the vertical alarm dataset to obtain multiple first frequent itemsets, and data mining is performed on the multiple first frequent itemsets according to prior principles to obtain multiple target frequent itemsets.

[0010] Based on multiple device identifiers in the multiple target frequent item sets, the association relationships between multiple cloud resources in the target system are determined, and the association relationships are stored in the configuration management database.

[0011] Optionally, before acquiring the target alarm data corresponding to multiple cloud resources within the target system, the method further includes:

[0012] Acquire first alarm data from multiple systems, the first alarm data including system identifiers, device identifiers and historical alarm time identifiers corresponding to multiple cloud resources within the multiple systems;

[0013] Based on the system identifiers corresponding to multiple first alarm data, the multiple first alarm data are classified and processed to obtain target alarm data corresponding to multiple cloud resources within the target system.

[0014] Optionally, the step of classifying the multiple target alarm data to obtain a horizontal alarm dataset includes:

[0015] Based on the historical alarm time identifiers corresponding to the multiple target alarm data, the multiple target alarm data are classified and processed according to a preset time interval to obtain the horizontal alarm dataset corresponding to the target system. The horizontal alarm dataset includes a subset of device identifiers corresponding to each preset time interval.

[0016] Optionally, the step of performing data mining processing based on the vertical alarm dataset to obtain multiple first frequent itemsets includes:

[0017] Based on the number of device identifiers corresponding to multiple cloud resources and the preset support threshold, determine the preset support threshold corresponding to the target system, and determine the first support number within the subset of historical alarm time identifiers corresponding to each cloud resource;

[0018] Based on multiple first support numbers and the preset support number threshold, the multiple first frequent itemsets are obtained. The first frequent itemsets include device identifiers corresponding to cloud resources whose first support number is greater than the preset support number threshold, and a subset of historical alarm time identifiers corresponding to the device identifiers.

[0019] Optionally, the step of performing data mining processing on the plurality of first frequent itemsets according to prior principles to obtain a plurality of target frequent itemsets includes:

[0020] The intersection of the historical alarm time identifier subsets corresponding to multiple first frequent itemsets is performed to obtain multiple candidate second frequent itemsets and the second support number corresponding to each candidate second frequent itemset. The candidate second frequent itemset includes the first intersection time identifiers within the historical alarm time identifier subsets corresponding to any two first frequent itemsets, and the second support number is used to indicate the number of first intersection time identifiers within the corresponding candidate second frequent itemset.

[0021] Based on the plurality of second support numbers and a preset support number threshold, at least one second frequent itemset is obtained, wherein the second frequent itemset is a candidate second frequent itemset whose second support number is greater than the preset support number threshold;

[0022] The intersection of the first intersection time identifiers within the at least one second frequent itemset is performed to obtain at least one candidate third frequent itemset and a third support number corresponding to each candidate third frequent itemset. The candidate third frequent itemset includes the second intersection time identifiers within the first intersection time identifiers corresponding to any three different second frequent itemsets. The third support number is used to indicate the number of second intersection time identifiers within the corresponding candidate third frequent itemset.

[0023] Determine whether the number of third support items for each candidate third frequent itemset is zero;

[0024] If so, the mining process of the vertical alarm dataset is terminated, and the second frequent itemset is taken as the target frequent itemset.

[0025] Optionally, the method further includes:

[0026] If there exists at least one candidate third frequent itemset with a non-zero third support number, then data mining processing is performed on the at least one candidate third frequent itemset according to the prior principle until the support number of the obtained frequent itemset is zero, and all frequent itemsets with non-zero support numbers are taken as the target frequent itemset.

[0027] Secondly, this application provides a cloud resource association mining apparatus, the apparatus comprising:

[0028] The acquisition module is used to acquire alarm data from multiple targets within the target system.

[0029] The processing module is used to classify and process the multiple target alarm data to obtain a horizontal alarm dataset. The target alarm data includes device identifiers and historical alarm time identifiers corresponding to multiple cloud resources in the target system.

[0030] The processing module is further configured to perform inverted indexing on the horizontal alarm dataset and perform bit storage on the inverted horizontal alarm dataset to obtain a vertical alarm dataset, wherein the vertical alarm dataset includes a subset of historical alarm time identifiers corresponding to each cloud resource.

[0031] The processing module is further configured to perform data mining processing based on the vertical alarm dataset to obtain multiple first frequent itemsets, and perform data mining processing on the multiple first frequent itemsets according to prior principles to obtain multiple target frequent itemsets.

[0032] The processing module is further configured to determine the association relationship between multiple cloud resources in the target system based on multiple device identifiers in the multiple target frequent item sets, and store the association relationship in the configuration management database.

[0033] Optionally, the acquisition module is further configured to acquire first alarm data of multiple systems, wherein the first alarm data includes system identifiers, device identifiers and historical alarm time identifiers corresponding to multiple cloud resources within the multiple systems;

[0034] The processing module is further configured to classify the multiple first alarm data according to the system identifiers corresponding to the multiple first alarm data, so as to obtain target alarm data corresponding to multiple cloud resources in the target system.

[0035] Optionally, the processing module is further configured to classify the multiple target alarm data according to the historical alarm time identifiers corresponding to the multiple target alarm data and according to a preset time interval to obtain a horizontal alarm dataset corresponding to the target system, wherein the horizontal alarm dataset includes a subset of device identifiers corresponding to each preset time interval.

[0036] Optionally, the processing module is further configured to determine the preset support threshold corresponding to the target system based on the number of device identifiers corresponding to multiple cloud resources and the preset support threshold, and to determine the first support number within the subset of historical alarm time identifiers corresponding to each cloud resource.

[0037] The processing module is further configured to obtain the plurality of first frequent itemsets based on the plurality of first support numbers and the preset support number threshold, wherein the first frequent itemsets include device identifiers corresponding to cloud resources whose first support numbers are greater than the preset support number threshold and a subset of historical alarm time identifiers corresponding to the device identifiers.

[0038] Optionally, the device further includes: a determination module;

[0039] The processing module is further configured to perform intersection processing on the historical alarm time identifier subsets corresponding to multiple first frequent itemsets to obtain multiple candidate second frequent itemsets and a second support number corresponding to each candidate second frequent itemset. The candidate second frequent itemset includes the first intersection time identifiers within the historical alarm time identifier subsets corresponding to any two first frequent itemsets, and the second support number is used to indicate the number of first intersection time identifiers within the corresponding candidate second frequent itemset.

[0040] The processing module is further configured to obtain at least one second frequent itemset based on the plurality of second support numbers and a preset support number threshold, wherein the second frequent itemset is a candidate second frequent itemset whose second support number is greater than the preset support number threshold;

[0041] The processing module is further configured to perform intersection processing on the first intersection time identifiers in the at least one second frequent itemset to obtain at least one candidate third frequent itemset and a third support number corresponding to each candidate third frequent itemset. The candidate third frequent itemset includes the second intersection time identifiers in the first intersection time identifiers corresponding to any three different second frequent itemsets. The third support number is used to indicate the number of second intersection time identifiers in the corresponding candidate third frequent itemset.

[0042] The judgment module is used to determine whether the number of third support items corresponding to each candidate third frequent itemset is zero;

[0043] The processing module is further configured to terminate the mining process of the vertical alarm dataset when the number of third support items corresponding to each candidate third frequent itemset is zero, and to use the second frequent itemset as the target frequent itemset.

[0044] Optionally, the processing module is further configured to perform data mining processing on the at least one candidate third frequent itemset according to a priori principles when the third support number corresponding to at least one candidate third frequent itemset is not zero, until the support number corresponding to the obtained frequent itemset is zero, and to take all frequent itemsets with non-zero support numbers as the target frequent itemset.

[0045] Thirdly, this application provides a device for mining cloud resource association relationships, comprising:

[0046] Memory;

[0047] processor;

[0048] The memory stores computer-executed instructions;

[0049] The processor executes computer execution instructions stored in the memory to implement the cloud resource association mining method as described in the first aspect and various possible implementations of the first aspect above.

[0050] Fourthly, this application provides a computer storage medium storing a computer program thereon, the computer program being executed by a processor to implement the cloud resource association mining method as described in the first aspect and various possible implementations of the first aspect above.

[0051] The cloud resource correlation mining method provided in this application obtains multiple target alarm data within a target system, categorizes these data to obtain a horizontal alarm dataset, performs inverted indexing on the horizontal alarm dataset, and stores the inverted horizontal alarm dataset by bit to obtain a vertical alarm dataset. Data mining is then performed on the vertical alarm dataset to obtain multiple first frequent itemsets. Based on prior principles, data mining is then performed on these first frequent itemsets to obtain multiple target frequent itemsets. Based on multiple device identifiers within these target frequent itemsets, the correlation between multiple cloud resources within the target system is determined, and the correlation is stored in a configuration management database. This method improves the accuracy and efficiency of mining correlation between cloud resources. Attached Figure Description

[0052] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0053] Figure 1 The process of the cloud resource association mining method provided in this application Figure 1 ;

[0054] Figure 2 The process of the cloud resource association mining method provided in this application Figure 2 ;

[0055] Figure 3 The process of the cloud resource association mining method provided in this application Figure 3 ;

[0056] Figure 4 A schematic diagram of the data mining process provided in this application;

[0057] Figure 5 A schematic diagram of the cloud resource association mining device provided in this application;

[0058] Figure 6 A schematic diagram of the structure of the cloud resource association mining equipment provided in this application.

[0059] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0061] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.

[0062] In this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0063] With the rapid development of cloud computing, cloud architecture is increasingly being used in large-scale systems. Cloud architecture involves multiple layers such as service layer, platform layer, component layer, network layer, and host hardware layer, which makes understanding the asset configuration relationship of the current system extremely complex, resulting in the inability to accurately and quickly find the relationship between Internet information resources.

[0064] There are two existing methods for mining cloud resource relationships. One method is to batch sort and enter each configuration item with a relationship in the database, which is then manually entered by operations and maintenance personnel through the platform. The other method is to traverse the configuration management database (CMDB) in the resource instance backend, compare the fields of all resources that need to be associated, and associate resources with the same fields.

[0065] However, existing methods for mining relationships based on manual data entry suffer from inaccuracies in manually entered data; methods for mining relationships based on resource instance backend equivalence judgment rely on the rationality of CMDB model field design, and are inefficient when the number of resources is large because they require pairwise comparisons of all resources that need to be associated.

[0066] To address the aforementioned issues, this application provides a method for mining cloud resource relationships. This method involves acquiring multiple target alarm data from a target system, classifying these alarm data to obtain a horizontal alarm dataset, inverting the horizontal alarm dataset, and storing it in bitwise storage to obtain a vertical alarm dataset. Data mining is then performed on the vertical alarm dataset until multiple frequent itemsets are obtained. Based on all device identifiers within these frequent itemsets, the relationships between multiple cloud resources within the target system are determined, and these relationships are stored in a configuration management database. This method improves the accuracy and efficiency of mining relationships between cloud resources.

[0067] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0068] Figure 1 This is the flowchart of the cloud resource association mining method provided in the embodiments of this application. Figure 1 .like Figure 1 As shown, the cloud resource association mining method provided in this embodiment includes:

[0069] S101. Obtain multiple target alarm data within the target system, and classify the multiple target alarm data to obtain a horizontal alarm dataset.

[0070] The target system refers to a cloud-based architecture system, which includes, for example, a service layer, a platform layer, a component layer, a network layer, and a host hardware layer. The target alarm data refers to the alarm data generated when multiple cloud resource devices in the target system are in an abnormal state. The target alarm data includes the device identifiers and historical alarm time identifiers corresponding to the multiple cloud resources in the target system. Cloud resources refer to the configuration resources of multiple cloud-based layers in the target system. The device identifiers corresponding to the cloud resources can be, for example, service 1, platform 2, component 3, network 4, etc., and the historical alarm time identifiers can be, for example, T1, T2, T3, T4, etc.

[0071] Understandably, when the target system is in an idle state, target alarm data corresponding to all cloud resources within the target system can be retrieved from the database. The obtained target alarm data is then categorized according to the corresponding device identifier and historical alarm time identifier to obtain a target alarm dataset in a horizontal data format, i.e., a horizontal alarm dataset. The horizontal alarm dataset includes multiple subsets, such as T1 (Service 1, Platform 3, Hardware 8), T2 (Service 4, Platform 5, Component 9, Network 7), and T3 (Service 2, Service 3, Platform 7, Platform 4, Component 1, Network 1). Subset T1 refers to cloud resources within the target system with device identifiers of Service 1, Platform 3, and Hardware 8 being in an abnormal state and generating corresponding target alarm data when the historical alarm time identifier is T1. Subset T2 refers to cloud resources within the target system with device identifiers of Service 4, Platform 5, Component 9, and Network 7 being in an abnormal state and generating corresponding target alarm data when the historical alarm time identifier is T2.

[0072] S102. Perform inverted indexing on the horizontal alarm dataset and perform bit storage on the inverted horizontal alarm dataset to obtain the vertical alarm dataset.

[0073] The vertical alarm dataset includes a subset of historical alarm timestamps corresponding to each cloud resource. Inverted indexing refers to performing matrix transformation on all subsets of the horizontal data format in the horizontal alarm dataset of the target system to obtain the alarm dataset in the vertical data format. The inverted alarm dataset is then stored in bits, that is, the alarm dataset is stored in memory space as a bit sequence using a bit set. Since the bit set wraps the bits in bytes, the purpose of storing the inverted alarm dataset in bits is to save storage space, optimize subsequent processing speed, and improve the efficiency of mining cloud resource correlation relationships.

[0074] S103. Perform data mining processing on the vertical alarm dataset to obtain multiple first frequent itemsets, and perform data mining processing on the multiple first frequent itemsets according to the prior principle to obtain multiple target frequent itemsets.

[0075] The vertical alarm dataset refers to the vertical alarm dataset in BitSet format after bit storage processing, which includes a subset of historical alarm time identifiers corresponding to each cloud resource. For example, multiple subsets can be Service 1 (T1, T2, T4), Network 2 (T1, T3), Platform 7 (T2, T3, T4, T8), etc.

[0076] For example, the ECLAT algorithm can be used to perform set operation-based data mining processing on the vertical alarm dataset to obtain multiple frequent itemsets. A frequent itemset is a subset of historical alarm time identifiers in the vertical alarm dataset where the number of historical alarm time identifiers is greater than a preset threshold. Therefore, the itemsets corresponding to multiple cloud resources obtained after performing the first set operation on the vertical alarm dataset are the first frequent itemsets. The itemsets of the first frequent itemset are consistent with the number of device identifiers corresponding to all cloud resources in the target system.

[0077] The a priori principle states that if an itemset is frequent, then all its subsets must also be frequent. Conversely, if an itemset is infrequent, then all its supersets must also be infrequent. Therefore, based on the a priori principle, iterative data mining can be performed on the multiple first frequent itemsets until multiple target frequent itemsets are obtained. The target frequent itemsets are those whose frequency of occurrence in all frequent itemsets meets a preset frequency, and these frequent itemsets are used as target frequent itemsets that can reflect the correlation between multiple cloud resources within the target system.

[0078] S104. Based on the multiple device identifiers in the multiple target frequent item sets, determine the association relationship between multiple cloud resources in the target system, and store the association relationship in the configuration management database.

[0079] The target frequent itemset is used to represent the device identifiers among multiple cloud resources with related relationships. This is because the target frequent itemset is obtained through data mining of a vertical alarm dataset. If multiple devices simultaneously generate alarm data at the same historical alarm time, it indicates a relationship between these devices. Therefore, all possible relationships between cloud resources within the target system can be constructed based on the target frequent itemset. These relationships are then stored in a configuration management database to facilitate subsequent calls to the database's interface for logically establishing relationships between cloud resources within the target system.

[0080] Optionally, after constructing all possible associations between multiple cloud resources based on the multiple target frequent itemsets, it is also possible to calculate the confidence level corresponding to each association, and determine whether the multiple confidence levels are greater than the preset minimum confidence level based on the multiple confidence levels and the preset minimum confidence level, and establish the associations that are greater than the preset minimum confidence level as reasonable associations. The preset minimum confidence level is, for example, 0.6.

[0081] The cloud resource correlation mining method provided in this application embodiment obtains multiple target alarm data within a target system, categorizes these data to obtain a horizontal alarm dataset, performs inverted indexing on the horizontal alarm dataset, and stores the inverted horizontal alarm dataset bitwise to obtain a vertical alarm dataset. Data mining is then performed on the vertical alarm dataset to obtain multiple first frequent itemsets. Based on prior principles, data mining is then performed on these first frequent itemsets to obtain multiple target frequent itemsets. Based on multiple device identifiers within these target frequent itemsets, the correlation between multiple cloud resources within the target system is determined, and the correlation is stored in a configuration management database. This improves the accuracy and efficiency of mining the correlation between cloud resources.

[0082] Figure 2 The process of the cloud resource association mining method provided in the embodiments of this application Figure 2 This embodiment is... Figure 1 Based on the examples, the method for mining cloud resource relationships is described in detail. For example... Figure 2 As shown, the cloud resource association mining method provided in this embodiment includes:

[0083] S201. Obtain first alarm data from multiple systems. The first alarm data includes system identifiers, device identifiers, and historical alarm time identifiers corresponding to multiple cloud resources within the multiple systems.

[0084] When multiple systems exist, each system corresponds to multiple cloud resources. When each cloud resource generates alarm data, the alarm data includes the system identifier, device identifier, and historical alarm time identifier corresponding to the cloud resource.

[0085] S202. Based on the system identifiers corresponding to the multiple first alarm data, classify the multiple first alarm data to obtain target alarm data corresponding to multiple cloud resources in the target system.

[0086] The target system is the system whose cloud resource relationships need to be mined. The first alarm data is the alarm data corresponding to all systems. Therefore, it is necessary to determine the system identifier of the target system based on the system identifier corresponding to all the first alarm data, and classify the multiple first alarm data according to the system identifier to obtain the target alarm data corresponding to multiple cloud resources in the target system. The target alarm data includes the device identifier and historical alarm time identifier corresponding to the multiple cloud resources in the target system.

[0087] S203. Obtain alarm data from multiple targets within the target system.

[0088] Step S203 is similar to step S101 above, and will not be described again here.

[0089] S204. Based on the historical alarm time identifiers corresponding to the multiple target alarm data, classify the multiple target alarm data according to a preset time interval to obtain the horizontal alarm dataset corresponding to the target system.

[0090] The preset time interval refers to the range of intervals preset according to the required mining accuracy, such as hourly granularity, 15-minute granularity, 30-minute granularity, etc.

[0091] It is understandable that when the fine-grained unit of the historical alarm time identifier corresponding to the first alarm data is at the second level, the fine-grained unit of the historical alarm time identifier corresponding to the target alarm data is also at the second level.

[0092] For example, if the historical alarm time identifiers T1, T2, and T3 belong to the same preset time interval among multiple target alarm data, then the target alarm data corresponding to T1, T2, and T3 can be classified into a subset.

[0093] S205. Perform inverted indexing on the horizontal alarm dataset and perform bit storage on the inverted horizontal alarm dataset to obtain the vertical alarm dataset.

[0094] Step S205 is similar to step S102 above, and will not be described again here.

[0095] S206. Based on the number of device identifiers corresponding to multiple cloud resources and the preset support threshold, determine the preset support threshold corresponding to the target system, and determine the first support number within the historical alarm time identifier subset corresponding to each cloud resource.

[0096] Among them, the preset support threshold refers to the minimum support corresponding to the target system preset according to the mining accuracy requirements, and the first support number refers to the number of historical alarm time identifiers in the historical alarm time identifier subset corresponding to each cloud resource.

[0097] Understandably, the vertical alarm dataset contains device identifiers corresponding to all cloud resources within the target system. Based on the number of all device identifiers and the preset support threshold, the preset support threshold corresponding to the target system can be obtained.

[0098] For example, when there are multiple itemsets in the vertical alarm dataset, such as Service 1 (T1, T2, T4), Network 2 (T1, T3), Platform 7 (T2, T3, T4, T8), etc., with a total of 10 itemsets, the total number of device identifiers for Service 1, Network 2, Platform 7, etc. is 10, and the preset support threshold is 0.1, then the preset support threshold for the target system is 1, the first support number for cloud resource service 1 is 3, the first support number for network 2 is 2, and the first support number for platform 7 is 4.

[0099] S207. Based on the multiple first support numbers and the preset support number threshold, the multiple first frequent itemsets are obtained.

[0100] The first frequent itemset includes device identifiers corresponding to cloud resources whose first support number is greater than the preset support number threshold, and a subset of historical alarm time identifiers corresponding to the device identifiers.

[0101] For example, referring to the vertical alarm dataset in step S206, the first support number of cloud resource service 1 is 3, the first support number of network 2 is 2, and the first support number of platform 7 is 4. At this time, the preset support number threshold of the target system is 1, and the first frequent itemset of the target system includes service 1, network 2, and platform 7.

[0102] S208. Based on the prior principle, perform data mining processing on the multiple first frequent itemsets to obtain multiple target frequent itemsets.

[0103] Step S208 is similar to step S103 above, and will not be described again here.

[0104] S209. Based on the multiple device identifiers in the multiple target frequent item sets, determine the association relationship between multiple cloud resources in the target system, and store the association relationship in the configuration management database.

[0105] Step S209 is similar to step S104 above, and will not be described again here.

[0106] The cloud resource association mining method provided in this application embodiment obtains first alarm data from multiple systems, classifies the first alarm data according to the system identifiers corresponding to the multiple first alarm data, and obtains target alarm data corresponding to multiple cloud resources within the target system. It then obtains multiple target alarm data within the target system, classifies the target alarm data according to the historical alarm time identifiers corresponding to the multiple target alarm data according to a preset time interval, and obtains a horizontal alarm dataset corresponding to the target system. Finally, it performs inverted indexing on the horizontal alarm dataset and performs bit storage on the inverted horizontal alarm dataset to obtain vertical alarm data. The system determines the preset support threshold for the target system based on the number of device identifiers corresponding to multiple cloud resources and a preset support threshold. It also determines the first support number within the subset of historical alarm time identifiers corresponding to each cloud resource. Based on the multiple first support numbers and the preset support threshold, multiple first frequent itemsets are obtained. Data mining processing is performed on the multiple first frequent itemsets according to prior principles to obtain multiple target frequent itemsets. Based on the multiple device identifiers in the multiple target frequent itemsets, the association relationship between multiple cloud resources in the target system is determined, and the association relationship is stored in the configuration management database, thereby improving the accuracy and efficiency of mining the association relationship between cloud resources.

[0107] Figure 3 The process of the cloud resource association mining method provided in the embodiments of this application Figure 3 This embodiment is... Figure 2 Based on the embodiments, a detailed explanation is provided on how data mining processing is performed on the multiple first frequent itemsets according to prior principles to obtain multiple target frequent itemsets. For example... Figure 2 As shown, the cloud resource association mining method provided in this embodiment includes:

[0108] S301. Perform intersection processing on the historical alarm time identifier subsets corresponding to multiple first frequent itemsets to obtain multiple candidate second frequent itemsets and the second support number corresponding to each candidate second frequent itemset.

[0109] The candidate second frequent itemset includes the first intersection time identifier within the historical alarm time identifier subset corresponding to any two first frequent itemsets, and the second support number is used to indicate the number of the first intersection time identifiers within the corresponding candidate second frequent itemset.

[0110] Figure 4 This is a schematic diagram of the data mining process provided in this embodiment, such as... Figure 4As shown, the vertical alarm dataset within the target system contains 5 cloud resource subsets, with device identifiers A, B, C, D, and E corresponding to the cloud resources. Each device identifier corresponds to a subset of historical alarm time identifiers, and each subset of historical alarm time identifiers contains multiple historical alarm time identifiers. For example, the historical alarm time identifiers in the subset of historical alarm time identifiers corresponding to device identifier A are: T1, T4, T5, T7, T8, and T9. If the preset support number threshold for the target system is 2, the first support number for device identifier A is 6, the first support number for device identifier B is 7, the first support number for device identifier C is 6, the first support number for device identifier D is 2, and the first support number for device identifier E is 2, it is determined that all first support numbers are greater than the preset support number threshold. Therefore, the multiple first frequent itemsets corresponding to the target system are A, B, C, D, and E.

[0111] Intersection processing is performed on the historical alarm time identifier subsets corresponding to all first frequent itemsets: A, B, C, D, and E, to obtain multiple candidate second frequent itemsets: (A,B), (A,C), (A,D), (A,E), (B,C), (B,D), (B,E), (C,D), (C,E), (D,E), and the second support number corresponding to each candidate second frequent itemset. Among them, the first intersection time identifiers of the candidate second frequent itemset (A,B) are (T1, T4, T8, T9), therefore, the second support number of the candidate second frequent itemset (A,B) is 4; the first intersection time identifiers of the candidate second frequent itemset (A,E) are (T1, T8), therefore, the second support number of the candidate second frequent itemset (A,E) is 2; the first intersection time identifier set of the candidate second frequent itemset (D,E) is an empty set, therefore, the second support number of the candidate second frequent itemset (D,E) is 0.

[0112] Optionally, after obtaining multiple candidate second frequent itemsets, it is also possible to determine whether the second support number corresponding to each candidate second frequent itemset is zero. If so, the data mining process of the vertical alarm dataset is terminated.

[0113] S302. Based on the plurality of second support numbers and the preset support number threshold, at least one second frequent itemset is obtained.

[0114] The second frequent itemset is a candidate second frequent itemset whose second support number is greater than the preset support number threshold. The number of second frequent itemsets can be one or more.

[0115] Continue to refer to Figure 4 ,like Figure 4As shown, after obtaining multiple second support numbers based on the number of first intersection time identifiers corresponding to all candidate second frequent itemsets, the second support numbers need to be filtered according to a preset support number threshold. Specifically, the second support number of candidate second frequent itemset (A,D) is 1, the second support number of candidate second frequent itemset (B,E) is 1, the second support number of candidate second frequent itemset (C,E) is 1, and the second support number of candidate second frequent itemsets (C,D) and (D,E) is 0. If the preset support number threshold is 2, then the candidate second frequent itemsets (A,D), (B,E), (C,E), (C,D), and (D,E) with second support numbers less than 2 need to be deleted to obtain the second frequent itemsets (A,B), (A,C), (A,E), (B,C), and (B,D).

[0116] S303. Perform intersection processing on the first intersection time identifiers within the at least one second frequent itemset to obtain at least one candidate third frequent itemset and the third support number corresponding to each candidate third frequent itemset.

[0117] The candidate third frequent itemset includes the second intersection time identifiers within the first intersection time identifiers corresponding to any three different second frequent itemsets, and the third support number is used to indicate the number of second intersection time identifiers within the corresponding candidate third frequent itemset.

[0118] Continue to refer to Figure 4 After obtaining multiple second frequent itemsets (A,B), (A,C), (A,E), (B,C), and (B,D), the intersection of the first intersection time markers within the second frequent itemsets is calculated to obtain multiple candidate third frequent itemsets: (A,B,C), (A,B,D), (A,B,E), (A,C,D), (A,C,E), (A,D,E), (B,C,D), (B,C,E), (B,D,E), (C,D,E) and the third support number corresponding to each candidate third frequent itemset.

[0119] S304. Determine whether the number of third support items corresponding to each candidate third frequent itemset is zero. If yes, proceed to step S305; otherwise, proceed to step S306.

[0120] The termination condition for data mining is that no candidate frequent itemsets or frequent itemsets can be obtained. Therefore, after obtaining the candidate third frequent itemsets, it is necessary to determine whether the historical alarm time identifier set corresponding to each candidate third frequent itemsets is an empty set, that is, whether the third support number is zero. If so, the mining process of the vertical alarm dataset is terminated, and the second frequent itemset is taken as the target frequent itemset. If not, the mining of the candidate third frequent itemsets continues until no candidate frequent itemsets or frequent itemsets can be obtained.

[0121] Continue to refer to Figure 4 , Figure 4 The third support number corresponding to the candidate third frequent itemsets is not all 0. Specifically, the second intersection time marker of the candidate third frequent itemset (A,B,D) is (T4), therefore, the third support number of the candidate third frequent itemset (A,B,D) is 1; the second intersection time marker of the candidate third frequent itemset (A,C,E) is (T8), therefore, the third support number of the candidate third frequent itemset (A,C,E) is 1; the second intersection time marker of the candidate third frequent itemset (B,C,E) is (T8), therefore, the third support number of the candidate third frequent itemset (B,C,E) is 1; the candidate third frequent itemsets (A,C,D), (A,D,E), The second intersection time signature set of (B,C,D), (B,D,E), and (C,D,E) is empty. Therefore, the third support number of the candidate third frequent itemsets (A,C,D), (A,D,E), (B,C,D), (B,D,E), and (C,D,E) is 0. At this time, if the preset support number threshold is 2, the candidate third frequent itemsets (A,B,D), (A,C,E), (B,C,E), (A,C,D), (A,D,E), (B,C,D), (B,D,E), and (C,D,E) with a third support number less than 2 need to be deleted to obtain the third frequent itemsets (A,B,C) and (A,B,E).

[0122] S305. Terminate the mining process of the vertical alarm dataset and use the second frequent itemset as the target frequent itemset.

[0123] If the third support number corresponding to all candidate third frequent itemsets (A,B,C), (A,B,D), (A,B,E), (A,C,D), (A,C,E), (A,D,E), (B,C,D), (B,C,E), (B,D,E), (C,D,E) is zero, then the second frequent itemsets (A,B), (A,C), (A,E), (B,C), and (B,D) are taken as the target frequent itemsets of the target system.

[0124] Optionally, the confidence levels corresponding to the target frequent itemsets (A,B), (A,C), (A,E), (B,C), and (B,D) can be calculated to obtain multiple confidence levels. The accuracy of the target frequent itemsets can be verified based on the preset minimum confidence level, and then the association between cloud resources can be established.

[0125] S306. Perform data mining processing on the at least one candidate third frequent itemset according to the prior principle until the support number corresponding to the obtained frequent itemset is zero, and take the frequent itemsets with non-zero support number as the target frequent itemset.

[0126] Wherein, when the number of third support corresponding to each candidate third frequent itemset is not zero, data mining processing is continued for at least one candidate third frequent itemset until the number of support corresponding to the obtained frequent itemset is zero, and all frequent itemsets with a number of support not zero are taken as the target frequent itemset.

[0127] like Figure 4 As shown, the intersection of the second intersection time markers within the third frequent itemsets (A,B,C) and (A,B,E) is performed to obtain at least one candidate fourth frequent itemset and the fourth support number corresponding to each candidate fourth frequent itemset. Based on the multiple fourth support numbers and a preset support number threshold, if all candidate fourth frequent itemsets cannot be used as fourth frequent itemsets, then the data mining ends, and all frequent itemsets with a support number not equal to 0 are used as target frequent itemsets: (A,B), (A,C), (A,E), (B,C), (B,D), (A,B,C), and (A,B,E).

[0128] The cloud resource association mining method provided in this application improves the accuracy and efficiency of mining cloud resource associations by performing intersection processing on the historical alarm time identifier subsets corresponding to multiple first frequent itemsets to obtain multiple candidate second frequent itemsets and a second support number corresponding to each candidate second frequent itemset. Based on the multiple second support numbers and a preset support number threshold, at least one second frequent itemset is obtained. The first intersection time identifiers within the at least one second frequent itemset are then intersected to obtain at least one candidate third frequent itemset and a third support number corresponding to each candidate third frequent itemset. It is determined whether the third support number corresponding to each candidate third frequent itemset is zero. If it is, the mining process of the vertical alarm dataset is terminated, and the second frequent itemset is taken as the target frequent itemset. If not, data mining processing is performed on the at least one candidate third frequent itemset according to a priori principles until the support number corresponding to the obtained frequent itemset is zero. Frequent itemsets with non-zero support numbers are all taken as the target frequent itemsets.

[0129] Figure 5 A schematic diagram of the structure of the cloud resource association mining device provided in this application. (Example) Figure 5 As shown, the cloud resource association mining device 400 provided in this embodiment includes:

[0130] The acquisition module 401 is used to acquire alarm data from multiple targets within the target system.

[0131] Processing module 402 is used to classify and process the multiple target alarm data to obtain a horizontal alarm dataset. The target alarm data includes device identifiers and historical alarm time identifiers corresponding to multiple cloud resources in the target system.

[0132] The processing module 402 is further configured to perform inverted indexing on the horizontal alarm dataset and perform bit storage on the inverted horizontal alarm dataset to obtain a vertical alarm dataset, wherein the vertical alarm dataset includes a subset of historical alarm time identifiers corresponding to each cloud resource.

[0133] The processing module 402 is further configured to perform data mining processing based on the vertical alarm dataset to obtain multiple first frequent itemsets, and perform data mining processing on the multiple first frequent itemsets based on prior principles to obtain multiple target frequent itemsets.

[0134] The processing module 402 is further configured to determine the association relationship between multiple cloud resources in the target system based on multiple device identifiers in the multiple target frequent item sets, and store the association relationship in the configuration management database.

[0135] Optionally, the acquisition module 401 is further configured to acquire first alarm data of multiple systems, wherein the first alarm data includes system identifiers, device identifiers and historical alarm time identifiers corresponding to multiple cloud resources within the multiple systems;

[0136] The processing module 402 is further configured to classify the multiple first alarm data according to the system identifiers corresponding to the multiple first alarm data, so as to obtain target alarm data corresponding to multiple cloud resources in the target system.

[0137] Optionally, the processing module 402 is further configured to classify the multiple target alarm data according to the historical alarm time identifiers corresponding to the multiple target alarm data and according to a preset time interval to obtain a horizontal alarm dataset corresponding to the target system, wherein the horizontal alarm dataset includes a subset of device identifiers corresponding to each preset time interval.

[0138] Optionally, the processing module 402 is further configured to determine the preset support threshold corresponding to the target system based on the number of device identifiers corresponding to multiple cloud resources and the preset support threshold, and to determine the first support number within the historical alarm time identifier subset corresponding to each cloud resource.

[0139] The processing module 402 is further configured to obtain the plurality of first frequent itemsets based on the plurality of first support numbers and the preset support number threshold, wherein the first frequent itemsets include device identifiers corresponding to cloud resources whose first support numbers are greater than the preset support number threshold and a subset of historical alarm time identifiers corresponding to the device identifiers.

[0140] Optionally, the device further includes: a determination module 403;

[0141] The processing module 402 is further configured to perform intersection processing on the historical alarm time identifier subsets corresponding to multiple first frequent itemsets to obtain multiple candidate second frequent itemsets and a second support number corresponding to each candidate second frequent itemset. The candidate second frequent itemset includes the first intersection time identifiers within the historical alarm time identifier subsets corresponding to any two first frequent itemsets, and the second support number is used to indicate the number of first intersection time identifiers within the corresponding candidate second frequent itemset.

[0142] The processing module 402 is further configured to obtain at least one second frequent itemset based on the plurality of second support numbers and a preset support number threshold, wherein the second frequent itemset is a candidate second frequent itemset whose second support number is greater than the preset support number threshold;

[0143] The processing module 402 is further configured to perform intersection processing on the first intersection time identifiers in the at least one second frequent itemset to obtain at least one candidate third frequent itemset and a third support number corresponding to each candidate third frequent itemset. The candidate third frequent itemset includes the second intersection time identifiers in the first intersection time identifiers corresponding to any three different second frequent itemsets. The third support number is used to indicate the number of second intersection time identifiers in the corresponding candidate third frequent itemset.

[0144] The judgment module 403 is used to determine whether the third support number corresponding to each candidate third frequent itemset is zero;

[0145] The processing module 402 is further configured to terminate the mining process of the vertical alarm dataset when the third support number corresponding to each candidate third frequent itemset is zero, and to use the second frequent itemset as the target frequent itemset.

[0146] Optionally, the processing module 402 is further configured to perform data mining processing on the at least one candidate third frequent itemset according to a priori principles when the third support number corresponding to at least one candidate third frequent itemset is not zero, until the support number corresponding to the obtained frequent itemset is zero, and to take all frequent itemsets with non-zero support numbers as the target frequent itemset.

[0147] Figure 6 A schematic diagram of the structure of the cloud resource association mining equipment provided in this application. (For example...) Figure 6 As shown, this application provides a cloud resource association mining device 500, which includes: a receiver 501, a transmitter 502, a processor 503, and a memory 504.

[0148] Receiver 501 is used to receive instructions and data;

[0149] Transmitter 502 is used to send commands and data;

[0150] Memory 504 is used to store instructions executed by the computer;

[0151] Processor 503 is used to execute computer execution instructions stored in memory 504 to implement the various steps of the cloud resource association mining method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing embodiments of the cloud resource association mining method.

[0152] Alternatively, the memory 504 can be either standalone or integrated with the processor 503.

[0153] When the memory 504 is set up independently, the electronic device also includes a bus for connecting the memory 504 and the processor 503.

[0154] This application also provides a computer storage medium storing computer execution instructions, which, when executed by a processor, implement the cloud resource association mining method performed by the cloud resource association mining device described above.

[0155] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0156] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0157] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for mining cloud resource association relationships, characterized in that, The method includes: Acquire multiple target alarm data within the target system, and classify the multiple target alarm data to obtain a horizontal alarm dataset. The target alarm data includes device identifiers and historical alarm time identifiers corresponding to multiple cloud resources within the target system. The horizontal alarm dataset is inverted and then stored bitwise to obtain a vertical alarm dataset. The vertical alarm dataset includes a subset of historical alarm time identifiers corresponding to each cloud resource. The bitwise storage of the inverted horizontal alarm dataset includes: using a bit set to store the inverted horizontal alarm dataset in memory as a bit sequence, wherein the bit set wraps the bits in bytes. Data mining is performed on the vertical alarm dataset to obtain multiple first frequent itemsets, and data mining is performed on the multiple first frequent itemsets according to prior principles to obtain multiple target frequent itemsets. Based on multiple device identifiers in the multiple target frequent item sets, determine the association relationship between multiple cloud resources in the target system, and store the association relationship in the configuration management database; The step of performing data mining processing based on the vertical alarm dataset to obtain multiple first frequent itemsets includes: Based on the number of device identifiers corresponding to multiple cloud resources and the preset support threshold, determine the preset support threshold corresponding to the target system, and determine the first support number within the subset of historical alarm time identifiers corresponding to each cloud resource; Based on multiple first support numbers and the preset support number threshold, the multiple first frequent itemsets are obtained. The first frequent itemsets include device identifiers corresponding to cloud resources whose first support number is greater than the preset support number threshold, and a subset of historical alarm time identifiers corresponding to the device identifiers. The data mining process is performed on the multiple first frequent itemsets according to prior principles to obtain multiple target frequent itemsets, including: The intersection of the historical alarm time identifier subsets corresponding to multiple first frequent itemsets is performed to obtain multiple candidate second frequent itemsets and the second support number corresponding to each candidate second frequent itemset. The candidate second frequent itemset includes the first intersection time identifiers within the historical alarm time identifier subsets corresponding to any two first frequent itemsets, and the second support number is used to indicate the number of first intersection time identifiers within the corresponding candidate second frequent itemset. Based on the plurality of second support numbers and a preset support number threshold, at least one second frequent itemset is obtained, wherein the second frequent itemset is a candidate second frequent itemset whose second support number is greater than the preset support number threshold; The intersection of the first intersection time identifiers within the at least one second frequent itemset is performed to obtain at least one candidate third frequent itemset and a third support number corresponding to each candidate third frequent itemset. The candidate third frequent itemset includes the second intersection time identifiers within the first intersection time identifiers corresponding to any three different second frequent itemsets. The third support number is used to indicate the number of second intersection time identifiers within the corresponding candidate third frequent itemset. Determine whether the number of third support items for each candidate third frequent itemset is zero; If so, the mining process of the vertical alarm dataset is terminated, and the second frequent itemset is taken as the target frequent itemset; If there exists at least one candidate third frequent itemset with a non-zero third support number, then data mining processing is performed on the at least one candidate third frequent itemset according to the prior principle until the support number of the obtained frequent itemset is zero, and all frequent itemsets with non-zero support numbers are taken as the target frequent itemset.

2. The method according to claim 1, characterized in that, Before acquiring the target alarm data corresponding to multiple cloud resources within the target system, the method further includes: Acquire first alarm data from multiple systems, the first alarm data including system identifiers, device identifiers and historical alarm time identifiers corresponding to multiple cloud resources within the multiple systems; Based on the system identifiers corresponding to multiple first alarm data, the multiple first alarm data are classified and processed to obtain target alarm data corresponding to multiple cloud resources within the target system.

3. The method according to claim 1, characterized in that, The process of classifying and processing the multiple target alarm data to obtain a horizontal alarm dataset includes: Based on the historical alarm time identifiers corresponding to the multiple target alarm data, the multiple target alarm data are classified and processed according to a preset time interval to obtain the horizontal alarm dataset corresponding to the target system. The horizontal alarm dataset includes a subset of device identifiers corresponding to each preset time interval.

4. A device for mining cloud resource association relationships, characterized in that, include: The acquisition module is used to acquire alarm data from multiple targets within the target system. The processing module is used to classify and process the multiple target alarm data to obtain a horizontal alarm dataset. The target alarm data includes device identifiers and historical alarm time identifiers corresponding to multiple cloud resources in the target system. The processing module is further configured to perform inverted index processing on the horizontal alarm dataset and perform bit storage on the inverted horizontal alarm dataset to obtain a vertical alarm dataset, wherein the vertical alarm dataset includes a subset of historical alarm time identifiers corresponding to each cloud resource; wherein, performing bit storage on the inverted horizontal alarm dataset includes: using a bit set to store the inverted horizontal alarm dataset in the form of a bit sequence in memory space, wherein the bit set wraps the bits in bytes; The processing module is further configured to perform data mining processing based on the vertical alarm dataset to obtain multiple first frequent itemsets, and to perform data mining processing on the multiple first frequent itemsets according to prior principles to obtain multiple target frequent itemsets; wherein, the step of performing data mining processing based on the vertical alarm dataset to obtain multiple first frequent itemsets includes: determining a preset support number threshold corresponding to the target system based on the number of device identifiers corresponding to multiple cloud resources and a preset support threshold, and determining the first support number within the historical alarm time identifier subset corresponding to each cloud resource; and obtaining multiple first support numbers and the preset support number threshold based on the multiple first support numbers and the preset support number threshold. The plurality of first frequent itemsets, wherein each first frequent itemset includes a device identifier corresponding to a cloud resource whose first support number is greater than the preset support number threshold, and a subset of historical alarm time identifiers corresponding to the device identifier; the step of performing data mining processing on the plurality of first frequent itemsets according to a priori principles to obtain a plurality of target frequent itemsets includes: performing intersection processing on the subsets of historical alarm time identifiers corresponding to the plurality of first frequent itemsets to obtain a plurality of candidate second frequent itemsets and a second support number corresponding to each candidate second frequent itemset, wherein the candidate second frequent itemsets includes a subset of historical alarm time identifiers corresponding to any two first frequent itemsets. The first intersection time identifier within the set is used to indicate the number of first intersection time identifiers within the corresponding candidate second frequent itemset. Based on the plurality of second support numbers and a preset support number threshold, at least one second frequent itemset is obtained, where the second frequent itemset is a candidate second frequent itemset whose second support number is greater than the preset support number threshold. Intersection processing is performed on the first intersection time identifiers within the at least one second frequent itemset to obtain at least one candidate third frequent itemset and a third support number corresponding to each candidate third frequent itemset. The candidate third frequent itemset includes the first intersection times corresponding to any three different second frequent itemsets. The second intersection time identifier within the identifier and the third support number are used to indicate the number of second intersection time identifiers in the corresponding candidate third frequent itemset; it is determined whether the third support number corresponding to each candidate third frequent itemset is zero; if so, the mining process of the vertical alarm dataset is terminated, and the second frequent itemset is taken as the target frequent itemset; if there is at least one candidate third frequent itemset with a non-zero third support number, data mining processing is performed on the at least one candidate third frequent itemset according to the prior principle until the support number corresponding to the obtained frequent itemset is zero, and all frequent itemsets with non-zero support numbers are taken as the target frequent itemset; The processing module is further configured to determine the association relationship between multiple cloud resources in the target system based on multiple device identifiers in the multiple target frequent item sets, and store the association relationship in the configuration management database.

5. The apparatus according to claim 4, characterized in that, The acquisition module is also used to acquire first alarm data of multiple systems, the first alarm data including system identifiers, device identifiers and historical alarm time identifiers corresponding to multiple cloud resources in the multiple systems; The processing module is further configured to classify the multiple first alarm data according to the system identifiers corresponding to the multiple first alarm data, so as to obtain target alarm data corresponding to multiple cloud resources in the target system.

6. A device for mining cloud resource association relationships, characterized in that, include: Memory; processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the cloud resource association mining method as described in any one of claims 1-3.

7. A computer storage medium, characterized in that, The computer storage medium stores computer execution instructions, which, when executed by a processor, are used to implement the cloud resource association mining method as described in any one of claims 1-3.

Images