A backup and disaster recovery method and system combining AI technology

Abstract

This application provides a backup and disaster recovery method and system that combines AI technology. The method involves acquiring text data from each data node in an enterprise database, determining the redundancy between data nodes based on all text data, and filtering out all data nodes to be backed up based on these redundancies. Sensitive features of each data node to be backed up are extracted, and their privacy priorities are determined based on these features. Access logs of each data node to be backed up are acquired, and access anomaly factors are determined based on these logs. The backup urgency of each data node is determined using its privacy priority and access anomaly factors, and a backup and disaster recovery strategy is generated based on this urgency. This approach can determine the urgency of data backup and generate a disaster recovery strategy, thereby improving the efficiency of the data backup process.

Description

Technical Field

[0001] This application relates to the field of backup and disaster recovery technology, and more specifically, to a backup and disaster recovery method and system that incorporates AI technology. Background Technology

[0002] With the advancement of digital transformation, the importance and complexity of enterprise data are constantly increasing, and traditional backup and disaster recovery systems can no longer meet the high requirements for data security, recovery efficiency, and business continuity. Backup and disaster recovery solutions that combine AI technology are gradually becoming an important means for enterprises to cope with data disasters and improve data protection levels. Through AI technology, backup and disaster recovery systems can automate processes such as data classification, priority determination, and backup strategy optimization, thereby achieving more efficient and intelligent disaster recovery.

[0003] However, in practical applications, the diversity and complexity of data pose significant challenges to the training and inference of AI algorithms. Enterprise data typically includes both structured and unstructured data, with varying formats, sizes, and update frequencies. AI models need to process these complex data sources, extract valuable information, and perform effective classification. Furthermore, the introduction of AI technology requires backup systems with powerful computing capabilities and storage capacity. The training and inference processes of AI algorithms often require substantial computing resources, especially when dealing with large-scale datasets, where training AI models can consume significant time and computing power. In addition, real-time data backup and disaster recovery demand high availability and low latency from backup systems, and the computational processes of AI systems may become performance bottlenecks. Therefore, determining the urgency of data backup and generating disaster recovery strategies to improve the efficiency of the data backup process remains a challenging problem for the industry. Summary of the Invention

[0004] This application provides a backup and disaster recovery method and system that combines AI technology, which can determine the urgency of data backup and generate a disaster recovery strategy to improve the efficiency of the data backup process.

[0005] Firstly, this application provides a backup and disaster recovery method that combines AI technology, the disaster recovery method comprising the following steps:

[0006] Obtain the text data of each data node in the enterprise database, determine the correlation redundancy between each data node in the enterprise database based on all the text data, and filter out all data nodes in the enterprise database to be backed up based on each correlation redundancy.

[0007] The sensitive features of each data node to be backed up are extracted, and the privacy priority of each data node to be backed up is determined based on the corresponding sensitive features.

[0008] Obtain the access logs of each data node to be backed up, and determine the access anomaly factors of each data node to be backed up based on the corresponding access logs.

[0009] The backup urgency of each data node to be backed up in the enterprise database is determined by the corresponding privacy priority and the corresponding access anomaly factor, and a backup disaster recovery strategy for each data node to be backed up is generated based on the corresponding backup urgency.

[0010] Preferably, determining the correlation redundancy between various data nodes in the enterprise database based on all text data specifically includes:

[0011] Generate hash signature sets for each data node in the enterprise database based on the corresponding text data;

[0012] Select a data node from all data nodes in the enterprise database and obtain the hash signature set corresponding to the selected data node;

[0013] The data redundancy between the selected data node and other data nodes in the enterprise database is determined based on the hash signature set corresponding to the selected data node.

[0014] The correlation redundancy between the selected data node and other data nodes in the enterprise database is determined by the corresponding data redundancy, thereby obtaining the correlation redundancy between each data node in the enterprise database.

[0015] Preferably, based on various relevant redundancy levels, all data nodes in the enterprise database to be backed up are specifically selected, including:

[0016] Obtain a pre-set redundancy threshold, and then filter out all data node pairs whose redundancy exceeds the redundancy threshold;

[0017] Each selected data node pair is removed, and the remaining data nodes and the data nodes in each data node pair with a redundancy not exceeding the redundancy threshold are taken as the data nodes to be backed up in the enterprise database, thus obtaining all the data nodes to be backed up in the enterprise database.

[0018] Preferably, natural language processing techniques and machine learning-based classification models are used to extract the sensitive features of each data node to be backed up.

[0019] Preferably, determining the privacy priority of each data node to be backed up based on the corresponding sensitive features involves inputting the sensitive features of each data node to be backed up into the evaluation model for evaluation, thereby obtaining the privacy priority of each data node to be backed up.

[0020] Preferably, determining the access anomaly factors for each data node to be backed up based on the corresponding access logs specifically includes:

[0021] For each data node to be backed up, extract multiple access features of the data node to be backed up from the access log of the data node to be backed up.

[0022] The access anomaly factor of the data node to be backed up is determined by all access characteristics, and then the access anomaly factor of each data node to be backed up is obtained.

[0023] Preferably, the backup and disaster recovery strategy for each data node to be backed up, generated according to the corresponding backup urgency, specifically includes:

[0024] Determine the backup priority group for each data node to be backed up based on its corresponding backup urgency.

[0025] The backup and disaster recovery strategy for each data node to be backed up is determined by the corresponding backup priority group.

[0026] Secondly, this application provides a backup and disaster recovery system incorporating AI technology, used to execute a backup and disaster recovery method incorporating AI technology. The backup and disaster recovery system incorporating AI technology includes a disaster recovery strategy generation unit, which includes:

[0027] The data acquisition module is used to acquire the text data of each data node in the enterprise database, determine the correlation redundancy between each data node in the enterprise database based on all the text data, and filter out all data nodes in the enterprise database to be backed up based on each correlation redundancy.

[0028] The priority determination module is used to extract the sensitive features of each data node to be backed up, and then determine the privacy priority of each data node to be backed up based on the corresponding sensitive features.

[0029] The abnormal access determination module is used to obtain the access logs of each data node to be backed up and determine the access abnormal factors of each data node to be backed up based on the corresponding access logs.

[0030] The policy generation module is used to determine the backup urgency of each data node to be backed up in the enterprise database by using the corresponding privacy priority and the corresponding access anomaly factor, and to generate a backup disaster recovery policy for each data node to be backed up based on the corresponding backup urgency.

[0031] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory for storing computer programs, and the processor for calling and running the computer programs from the memory, so that the computer device performs the above-described backup and disaster recovery method combined with AI technology.

[0032] Fourthly, this application provides a computer-readable storage medium storing instructions or code that, when executed on a computer, cause the computer to implement the aforementioned backup and disaster recovery method incorporating AI technology.

[0033] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0034] This application provides a backup and disaster recovery method and system that combines AI technology. The method involves acquiring text data from each data node in an enterprise database, determining the redundancy between data nodes based on all text data, and filtering out all data nodes to be backed up based on these redundancies. Sensitive features of each data node to be backed up are extracted, and the privacy priority of each node is determined based on these features. Access logs of each data node to be backed up are obtained, and access anomaly factors are determined based on these logs. The backup urgency of each data node to be backed up is determined using the corresponding privacy priority and access anomaly factors, and a backup and disaster recovery strategy is generated based on the corresponding backup urgency.

[0035] Therefore, this application demonstrates the following: First, by acquiring the text data of each data node in the enterprise database and calculating the redundancy between them, redundant backups can be significantly reduced, storage resources optimized, and costs lowered. Filtering data nodes to be backed up based on redundancy avoids unnecessary duplicate backups, allowing for more efficient resource allocation and faster data recovery and disaster recovery response. Second, by extracting the sensitive features of each data node to be backed up and determining its privacy priority based on these features, it effectively identifies which data requires priority protection, thus enabling differentiated backup and disaster recovery strategies. Third, acquiring the access logs of each data node to be backed up and identifying access anomaly factors based on these logs effectively identifies potential security risks and abnormal access behavior, helping to assess which data nodes may face higher security threats, allowing for priority backup of higher-risk data nodes. Finally, by dividing data nodes into backup priority groups based on backup urgency and developing corresponding backup and disaster recovery strategies for each priority group, the efficiency of the data backup process is improved.

[0036] In summary, the technical solution adopted in this application can determine the urgency of data backup and generate a disaster recovery strategy, thereby improving the efficiency of the data backup process. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is an exemplary flowchart illustrating a backup and disaster recovery method incorporating AI technology, according to some embodiments of this application.

[0039] Figure 2 This is an exemplary flowchart illustrating the determination of the correlation redundancy between various data nodes in an enterprise database, according to some embodiments of this application.

[0040] Figure 3 This is an exemplary flowchart illustrating the determination of access anomaly factors for each data node to be backed up, according to some embodiments of this application.

[0041] Figure 4 These are schematic diagrams of exemplary hardware and / or software of a disaster recovery strategy generation unit according to some embodiments of this application;

[0042] Figure 5 This is a schematic diagram of the structure of a computer device that implements a backup and disaster recovery method incorporating AI technology, according to some embodiments of this application. Detailed Implementation

[0043] The technical solutions of the embodiments 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, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0044] This application provides a backup and disaster recovery method and system combining AI technology. The core of the method is to acquire text data of each data node in an enterprise database, determine the redundancy between data nodes based on all text data, and filter all data nodes to be backed up based on the redundancy. Sensitive features of each data node to be backed up are extracted, and the privacy priority of each data node is determined based on these features. Access logs of each data node to be backed up are acquired, and access anomaly factors are determined based on these logs. The backup urgency of each data node to be backed up is determined using the corresponding privacy priority and access anomaly factors, and a backup and disaster recovery strategy is generated based on the backup urgency. This approach can determine the urgency of data backup and generate a disaster recovery strategy, thereby improving the efficiency of the data backup process.

[0045] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific implementation methods. (Reference) Figure 1 The figure is an exemplary flowchart of a backup and disaster recovery method incorporating AI technology, according to some embodiments of this application. The disaster recovery method 100 mainly includes the following steps:

[0046] In step 101, text data of each data node in the enterprise database is obtained, the correlation redundancy between each data node in the enterprise database is determined based on all the text data, and all data nodes to be backed up in the enterprise database are selected based on the correlation redundancy.

[0047] In practice, text data for each data node in the enterprise database can be obtained through data query languages ​​and traversal. In actual implementation, all fields containing text data can be queried. For each data node, the text data it contains can be extracted, and structured data (such as tabular data) can be transformed into a form suitable for text processing. This allows for the cleaning and standardization of text data, and the removal of noise, such as irrelevant symbols and repeated words.

[0048] Preferably, in some embodiments, reference is made to Figure 2 As shown, this diagram is an exemplary flowchart of determining the correlation redundancy between various data nodes in an enterprise database in some embodiments of this application. In this embodiment, determining the correlation redundancy between various data nodes in an enterprise database based on all text data can be achieved using the following steps:

[0049] In step 1011, a hash signature set corresponding to each data node in the enterprise database is generated based on the corresponding text data;

[0050] In step 1012, a data node is selected from all data nodes in the enterprise database, and the hash signature set corresponding to the selected data node is obtained;

[0051] In step 1013, the data redundancy between the selected data node and other data nodes in the enterprise database is determined based on the hash signature set corresponding to the selected data node;

[0052] In step 1014, the correlation redundancy between the selected data node and other data nodes in the enterprise database is determined by the corresponding data redundancy, thereby obtaining the correlation redundancy between each data node in the enterprise database.

[0053] In practical implementation, firstly, a hash signature set corresponding to each data node in the enterprise database can be generated based on the corresponding text data. This is achieved by using the SimHash method to hash the text data of the data node, thus obtaining the hash signature set corresponding to that data node. This method yields the hash signature sets for each data node in the enterprise database. Then, a data node can be selected from all data nodes in the enterprise database, and its corresponding hash signature set can be obtained. Furthermore, based on the hash signature set of the selected data node, the data redundancy between the selected data node and other data nodes in the enterprise database can be determined. This data redundancy represents the degree of content similarity and overlap between data nodes. In actual implementation, this data redundancy can be determined in the following way:

[0054]

[0055] Wherein, λ(D) i D j ) represents data node D i With data node D j Data redundancy between D i D represents the i-th data node in the enterprise database. j h(D) represents the j-th data node in the enterprise database. i ) represents data node D i The corresponding hash signature set, h(D) j ) represents data node D j The corresponding hash signature set can be used to obtain the data redundancy between the selected data node and other data nodes in the enterprise database through the above method.

[0056] In addition, in practical implementation, the correlation redundancy between the selected data node and other data nodes in the enterprise database can be determined by the corresponding data redundancy. Correlation redundancy represents the degree of interrelation between the content of data nodes in the enterprise database. In actual implementation, correlation redundancy can be determined in the following way:

[0057]

[0058] Wherein, γ(D) i D j ) represents data node D i With data node D j The correlation redundancy between them, D i D represents the i-th data node in the enterprise database. j Let λ(D) represent the j-th data node in the enterprise database. i D j ) represents data node D i With data node D j Data redundancy between them, min(D) i D j ) and max(D i D j ) respectively represent the selection of data node D i With data node D j For both small and large amounts of text data, the redundancy between various data nodes in the enterprise database can be obtained using the methods described above.

[0059] In some embodiments, the selection of all data nodes to be backed up in the enterprise database based on various relevant redundancy levels can be achieved in the following manner:

[0060] Obtain a pre-set redundancy threshold, and then filter out all data node pairs whose redundancy exceeds the redundancy threshold;

[0061] Each selected data node pair is removed, and the remaining data nodes and the data nodes in each data node pair with a redundancy not exceeding the redundancy threshold are taken as the data nodes to be backed up in the enterprise database, thus obtaining all the data nodes to be backed up in the enterprise database.

[0062] In practice, firstly, a redundancy threshold can be set based on historical experiments and data analysis, which will not be elaborated here. Secondly, all relevant redundancies can be compared with the redundancy threshold to filter out all data node pairs with relevant redundancy exceeding the threshold. Finally, one data node from each of the filtered data node pairs can be removed. This can be done randomly or in other ways, without limitation. The remaining data nodes and the data nodes from each data node pair with relevant redundancy not exceeding the redundancy threshold can then be used as the data nodes to be backed up in the enterprise database. By using the above method, all the data nodes to be backed up in the enterprise database can be obtained.

[0063] It should be noted that by obtaining the text data of each data node in the enterprise database and calculating the correlation redundancy between each data node, redundant backups can be significantly reduced, storage resources can be optimized, and costs can be reduced. By selecting data nodes to be backed up based on correlation redundancy, unnecessary duplicate backups can be avoided, thereby allocating resources more efficiently and achieving faster data recovery and disaster recovery response.

[0064] In step 102, the sensitive features of each data node to be backed up are extracted, and then the privacy priority of each data node to be backed up is determined based on the corresponding sensitive features.

[0065] In some embodiments, natural language processing (NLP) techniques and machine learning-based classification models are used to extract sensitive features from each data node to be backed up. Specifically, firstly, NLP techniques can be used to automatically extract sensitive features from the text data of each data node. Then, a machine learning feature extraction model can be used to further classify the text data of the data nodes to determine which data in the text data belong to the sensitive features. The sensitive features of each data node to be backed up can be extracted in this way.

[0066] In some embodiments, determining the privacy priority of each data node to be backed up based on its corresponding sensitive features involves inputting the sensitive features of each data node to be backed up into an evaluation model for evaluation, thereby obtaining the privacy priority of each data node to be backed up.

[0067] It should be noted that in this application, privacy priority refers to the importance of the sensitive features of the corresponding data node to be backed up in terms of privacy protection. In practice, the sensitive features of each data node to be backed up can be input into an evaluation model for assessment. The evaluation model can assess the sensitivity of the sensitive features of the data node. Different types of sensitive features will have different priorities; for example, personal identification information and financial information have higher privacy priorities, while business logs or technical documents have lower privacy priorities. The privacy priority of each data node to be backed up can be obtained through the above method.

[0068] It's important to note that by extracting the sensitive features of each data node to be backed up and determining their privacy priority based on these features, it's possible to effectively identify which data requires priority protection, thereby enabling the development of differentiated backup and disaster recovery strategies. Sensitive data typically implies higher risks and compliance requirements; therefore, high-privacy-priority data nodes are given more stringent protection measures, such as more frequent backups, strong encryption, and distributed storage. Low-privacy-priority data, on the other hand, can be backed up using conventional strategies, thus saving resources and costs.

[0069] In step 103, the access logs of each data node to be backed up are obtained, and the access anomaly factors of each data node to be backed up are determined based on the corresponding access logs.

[0070] In practice, access logs for each data node to be backed up can be obtained from the enterprise database.

[0071] Preferably, in some embodiments, reference is made to Figure 3 As shown, this diagram is an exemplary flowchart of determining the access anomaly factor of each data node to be backed up in some embodiments of this application. In this embodiment, determining the access anomaly factor of each data node to be backed up based on the corresponding access log can be achieved by the following steps:

[0072] In step 1031, for each data node to be backed up, multiple access features of the data node to be backed up are extracted from the access log of the data node to be backed up.

[0073] In step 1032, the access anomaly factor of the data node to be backed up is determined by all access characteristics, thereby obtaining the access anomaly factor of each data node to be backed up.

[0074] In specific implementation, firstly, for each data node to be backed up, multiple access characteristics of the data node to be backed up can be extracted from the access logs of the data node to be backed up through statistical methods. The access characteristics in this application include: access frequency of text data, access interval, file modification frequency, and file size change rate. Then, the access anomaly factor of the data node to be backed up can be determined through all access characteristics. The access anomaly factor represents the degree of access anomaly of the text data of the data node to be backed up. In actual implementation, for each access characteristic, the mean and standard deviation of the access characteristic can be calculated, and the current value of the access characteristic can be obtained. The difference between the current value and the mean of the access characteristic is divided by the standard deviation, and the result is used as the anomaly factor of the access characteristic. The anomaly factor represents the degree of anomaly of the corresponding access characteristic. The anomaly factor of each access characteristic can be obtained in the above way. The anomaly factors of each access characteristic are weighted and summed. The corresponding weights can be set according to historical experience. The final result can be used as the access anomaly factor of the corresponding data node to be backed up. The access anomaly factor of each data node to be backed up can be obtained in the above way.

[0075] It should be noted that obtaining access logs for each data node to be backed up and identifying abnormal access factors based on these logs can effectively identify potential security risks and abnormal access behaviors. This helps assess which data nodes to be backed up may face higher security threats, allowing for priority backup of higher-risk data nodes. This ensures that critical data can be quickly recovered in the event of an attack or other disaster, reducing the risk of data loss.

[0076] In step 104, the backup urgency of each data node to be backed up in the enterprise database is determined by the corresponding privacy priority and the corresponding access anomaly factor, and a backup disaster recovery strategy for each data node to be backed up is generated based on the corresponding backup urgency.

[0077] In some embodiments, the backup urgency of each data node to be backed up in the enterprise database is determined by the corresponding privacy priority and the corresponding access anomaly factor. It should be noted that, in this application, the backup urgency represents the degree of urgency of backing up the corresponding data node. In specific implementation, for each data node to be backed up in the enterprise database, the privacy priority and the access anomaly factor of the data node to be backed up can be weighted and summed. The corresponding weights can be set through data analysis and historical experiments, and the calculation result is used as the backup urgency of the data node to be backed up. The backup urgency of each data node to be backed up in the enterprise database can be obtained in the above way.

[0078] In some embodiments, generating backup and disaster recovery strategies for each data node to be backed up based on its corresponding backup urgency can be done in the following manner:

[0079] Determine the backup priority group for each data node to be backed up based on its corresponding backup urgency.

[0080] The backup and disaster recovery strategy for each data node to be backed up is determined by the corresponding backup priority group.

[0081] In practical implementation, firstly, the backup priority group of each data node to be backed up can be determined based on the corresponding backup urgency. The backup priority group represents the priority range where the backup urgency of the corresponding data node falls. The backup urgency of the data node to be backed up can be mapped to a pre-defined priority table. In this application, this priority table includes an urgent priority group, a normal priority group, and a low priority group. Then, the backup disaster recovery strategy for each data node to be backed up can be determined based on the corresponding backup priority group. That is, once the data nodes to be backed up are divided into different backup priority groups according to their backup urgency, the backup disaster recovery strategy for each data node to be backed up can be determined based on the characteristics of each backup priority group. Different priority groups will adopt different backup frequencies, storage methods, encryption levels, and other strategies, for example:

[0082] The backup and disaster recovery strategy for the emergency priority group is as follows: perform real-time backups or hourly incremental backups to ensure data timeliness and avoid data loss; adopt off-site backups (such as cloud data centers in different geographical locations) and multi-cloud backups (distributing backup data across multiple cloud platforms) to ensure data recoverability even in the event of a disaster; perform high-strength encryption (such as AES-256 encryption) on backup data to ensure data security; set up real-time or near-real-time recovery mechanisms to quickly restore important data in the event of a disaster and avoid prolonged business interruption; and regularly verify the integrity of backup data to ensure its accuracy and recoverability.

[0083] The backup and disaster recovery strategy for the ordinary priority group is as follows: regular backups (such as daily incremental backups) are used to ensure data integrity and reliability; data backups are combined with local data centers and cloud storage to balance cost and security; data is encrypted using conventional methods (such as AES-128 encryption) to ensure that data backups are not leaked; a longer recovery time is set so that data can be recovered in the event of a disaster without affecting normal operations; and backup integrity checks are performed regularly, but not as frequently as for the high priority group.

[0084] The backup and disaster recovery strategy for low-priority groups is as follows: use infrequent backups (such as a full backup once a week or backups on demand); store backup data in local data centers or cloud storage, without the need for off-site backups; perform regular encryption on backup data, but without excessive encryption; in the event of a disaster, the time to recover these low-priority data can be appropriately extended, but it is still necessary to ensure that the data can be recovered; perform occasional verification of backup data, but it is not necessary to perform it frequently.

[0085] It's important to note that by prioritizing data nodes based on backup urgency and developing corresponding backup and disaster recovery strategies for each group, enterprises can achieve more efficient and targeted backup management. This ensures that the most critical and sensitive data receives priority protection, while less important data is backed up appropriately when resources are limited. Through flexible disaster recovery strategies and backup prioritization, enterprises can not only optimize the use of backup resources but also quickly recover critical data in the event of a disaster, reducing the risk of business interruption and improving the efficiency of data backup and recovery.

[0086] Therefore, this application demonstrates the following: First, by acquiring the text data of each data node in the enterprise database and calculating the redundancy between them, redundant backups can be significantly reduced, storage resources optimized, and costs lowered. Filtering data nodes to be backed up based on redundancy avoids unnecessary duplicate backups, allowing for more efficient resource allocation and faster data recovery and disaster recovery response. Second, by extracting the sensitive features of each data node to be backed up and determining its privacy priority based on these features, it effectively identifies which data requires priority protection, thus enabling differentiated backup and disaster recovery strategies. Third, acquiring the access logs of each data node to be backed up and identifying access anomaly factors based on these logs effectively identifies potential security risks and abnormal access behavior, helping to assess which data nodes may face higher security threats, allowing for priority backup of higher-risk data nodes. Finally, by dividing data nodes into backup priority groups based on backup urgency and developing corresponding backup and disaster recovery strategies for each priority group, the efficiency of the data backup process is improved.

[0087] In summary, the technical solution adopted in this application can determine the urgency of data backup and generate a disaster recovery strategy, thereby improving the efficiency of the data backup process.

[0088] Furthermore, in another aspect of this application, in some embodiments, this application provides a backup and disaster recovery system incorporating AI technology. This AI-integrated backup and disaster recovery system includes a disaster recovery strategy generation unit, as referenced... Figure 4 The figure is a schematic diagram of exemplary hardware and / or software of a disaster recovery strategy generation unit according to some embodiments of this application. The disaster recovery strategy generation unit 400 includes: a data acquisition module 401, a priority determination module 402, an abnormal access determination module 403, and a strategy generation module 404, which are described below:

[0089] The data acquisition module 401 in this application is mainly used to acquire the text data of each data node in the enterprise database, determine the correlation redundancy between each data node in the enterprise database based on all the text data, and filter out all data nodes to be backed up in the enterprise database based on each correlation redundancy.

[0090] Priority determination module 402: In this application, priority determination module 402 is mainly used to extract the sensitive features of each data node to be backed up, and then determine the privacy priority of each data node to be backed up based on the corresponding sensitive features.

[0091] Abnormal access determination module 403: In this application, abnormal access determination module 403 is mainly used to obtain the access logs of each data node to be backed up, and determine the access abnormal factors of each data node to be backed up based on the corresponding access logs.

[0092] The policy generation module 404 in this application is mainly used to determine the backup urgency of each data node to be backed up in the enterprise database by the corresponding privacy priority and the corresponding access anomaly factor, and to generate a backup disaster recovery policy for each data node to be backed up according to the corresponding backup urgency.

[0093] The foregoing has detailed an example of a backup and disaster recovery method and system incorporating AI technology, as provided in the embodiments of this application. It is understood that the corresponding apparatus, in order to achieve the above functions, includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0094] In some embodiments, this application also provides a computer device, the computer device including a memory and a processor, the memory for storing a computer program, and the processor for calling and running the computer program from the memory, so that the computer device performs the above-described backup and disaster recovery method incorporating AI technology.

[0095] In some embodiments, reference Figure 5 The dashed lines in the figure indicate that the unit or module is optional. This figure is a schematic diagram of the structure of a computer device according to an embodiment of this application providing a backup and disaster recovery method incorporating AI technology. The backup and disaster recovery method incorporating AI technology described in the above embodiment can... Figure 5 The computer device 500 shown is used to implement this, and the computer device 500 includes at least one processor 501, a memory 502 and at least one communication unit 505. The computer device 500 may be a terminal device, a server or a chip.

[0096] The processor 501 can be a general-purpose processor or a special-purpose processor. For example, the processor 501 can be a central processing unit (CPU). The CPU can be used to control the computer device 500, execute software programs, and process data from the software programs. The computer device 500 may also include a communication unit 505 for inputting (receiving) and outputting (transmitting) signals.

[0097] For example, computer device 500 may be a chip, communication unit 505 may be the input and / or output circuit of the chip, or communication unit 505 may be the communication interface of the chip, and the chip may be a component of terminal device, network device or other device.

[0098] For example, computer device 500 may be a terminal device or a server, and communication unit 505 may be a transceiver of the terminal device or the server, or communication unit 505 may be a transceiver circuit of the terminal device or the server.

[0099] The computer device 500 may include one or more memories 502 storing a program 504. The program 504 can be executed by a processor 501 to generate instructions 503, causing the processor 501 to perform the methods described in the above method embodiments according to the instructions 503. Optionally, the memory 502 may also store data (such as a target audit model). Optionally, the processor 501 may also read data stored in the memory 502, which may be stored at the same storage address as the program 504, or the data may be stored at a different storage address than the program 504.

[0100] The processor 501 and memory 502 can be configured separately or integrated together, for example, integrated on the system-on-chip (SOC) of the terminal device.

[0101] It should be understood that each step of the above method embodiment can be completed by hardware logic circuits or software instructions in the processor 501. The processor 501 can be a central processing unit, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, such as discrete gates, transistor logic devices, or discrete hardware components.

[0102] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] For example, in some embodiments, this application also provides a computer-readable storage medium storing instructions or code that, when executed on a computer, cause the computer to implement the aforementioned backup and disaster recovery method incorporating AI technology.

[0104] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0105] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A backup and disaster recovery method combining AI technology, characterized in that, The disaster recovery method includes the following steps: Obtain the text data of each data node in the enterprise database, determine the correlation redundancy between each data node in the enterprise database based on all the text data, and filter out all data nodes in the enterprise database to be backed up based on each correlation redundancy. The sensitive features of each data node to be backed up are extracted, and the privacy priority of each data node to be backed up is determined based on the corresponding sensitive features. Obtain the access logs of each data node to be backed up, and determine the access anomaly factors of each data node to be backed up based on the corresponding access logs. The backup urgency of each data node to be backed up in the enterprise database is determined by the corresponding privacy priority and the corresponding access anomaly factor. Specifically, the privacy priority and access anomaly factor of the data node to be backed up are weighted and summed, and the calculation result is used as the backup urgency of the data node to be backed up. The backup urgency indicates the degree of urgency of the backup of the corresponding data node to be backed up. The backup disaster recovery strategy of each data node to be backed up is generated according to the corresponding backup urgency. Specifically, determining the access anomaly factors for each data node to be backed up based on the corresponding access logs includes: For each data node to be backed up, extract multiple access features of the data node to be backed up from the access log of the data node to be backed up. The access anomaly factor of the data node to be backed up is determined by all access characteristics, thereby obtaining the access anomaly factor of each data node to be backed up, wherein the access anomaly factor represents the degree of anomaly of the corresponding access characteristic.

2. The backup and disaster recovery method combining AI technology as described in claim 1, characterized in that, Determining the correlation redundancy between various data nodes in the enterprise database based on all text data specifically includes: Generate hash signature sets for each data node in the enterprise database based on the corresponding text data; Select a data node from all data nodes in the enterprise database and obtain the hash signature set corresponding to the selected data node; The data redundancy between the selected data node and other data nodes in the enterprise database is determined based on the hash signature set corresponding to the selected data node. The correlation redundancy between the selected data node and other data nodes in the enterprise database is determined by the corresponding data redundancy, thereby obtaining the correlation redundancy between each data node in the enterprise database.

3. The backup and disaster recovery method combining AI technology as described in claim 1, characterized in that, Based on various relevant redundancy levels, all data nodes in the enterprise database to be backed up are specifically selected, including: Obtain a pre-set redundancy threshold, and then filter out all data node pairs whose redundancy exceeds the redundancy threshold; Each selected data node pair is removed, and the remaining data nodes and the data nodes in each data node pair with a redundancy not exceeding the redundancy threshold are taken as the data nodes to be backed up in the enterprise database, thus obtaining all the data nodes to be backed up in the enterprise database.

4. The backup and disaster recovery method combining AI technology as described in claim 1, characterized in that, The sensitive features of each data node to be backed up are extracted using natural language processing techniques and machine learning-based classification models.

5. The backup and disaster recovery method combining AI technology as described in claim 1, characterized in that, Determining the privacy priority of each data node to be backed up based on its corresponding sensitive features involves inputting the sensitive features of each data node to be backed up into the evaluation model for evaluation, thereby obtaining the privacy priority of each data node to be backed up.

6. The backup and disaster recovery method combining AI technology as described in claim 1, characterized in that, The backup and disaster recovery strategy for each data node to be backed up is generated based on the corresponding backup urgency, specifically including: Determine the backup priority group for each data node to be backed up based on its corresponding backup urgency. The backup and disaster recovery strategy for each data node to be backed up is determined by the corresponding backup priority group.

7. A backup and disaster recovery system incorporating AI technology, used to execute a backup and disaster recovery method incorporating AI technology as described in any one of claims 1 to 6, wherein the backup and disaster recovery system incorporating AI technology includes a disaster recovery strategy generation unit, characterized in that, The disaster recovery strategy generation unit includes: The data acquisition module is used to acquire the text data of each data node in the enterprise database, determine the correlation redundancy between each data node in the enterprise database based on all the text data, and filter out all data nodes in the enterprise database to be backed up based on each correlation redundancy. The priority determination module is used to extract the sensitive features of each data node to be backed up, and then determine the privacy priority of each data node to be backed up based on the corresponding sensitive features. The abnormal access determination module is used to obtain the access logs of each data node to be backed up, and determine the access abnormal factors of each data node to be backed up based on the corresponding access logs. The policy generation module is used to determine the backup urgency of each data node to be backed up in the enterprise database by using the corresponding privacy priority and the corresponding access anomaly factor, and to generate a backup disaster recovery policy for each data node to be backed up based on the corresponding backup urgency.

8. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to call and run the computer programs from the memory, so that the computer device performs a backup and disaster recovery method incorporating AI technology as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions or code that, when executed on a computer, cause the computer to implement a backup and disaster recovery method incorporating AI technology as described in any one of claims 1 to 6.

Images