Backup method and device, storage medium and electronic equipment

Abstract

The invention discloses a backup method and device, a storage medium and electronic equipment. Relates to the field of big data, and comprises the following steps: obtaining a current operation state of a storage pool, the storage pool comprising N storage addresses, the N storage addresses respectively corresponding to different storage performances, and N being a positive integer; based on the current running state, a target model is adopted to determine a target address from the N storage addresses, the target model is obtained through training on the basis of historical backup feedback information of the storage pool, and the historical backup feedback information is obtained on the basis of the historical storage success rate and the historical backup rate of the storage pool; and storing the business data to be backed up to the target address. Through the data backup control method and device, the problem that the accuracy of data backup control is not ideal in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of big data, and in particular, to a backup method, apparatus, storage medium, and electronic device. Background Art

[0002] With the rapid development of technologies such as big data and cloud computing, the amount of data has grown exponentially, and the importance of data has become increasingly prominent. Data security and integrity are important aspects to ensure business continuity. Therefore, in order to prevent data loss or damage, data integrity is usually ensured through backup. In related technologies, backup managers manually create backup policies based on actual usage information such as application requests and database usage. At the same time, operation and maintenance personnel establish backup policies in batches according to the planned usage scenarios of application backups and import them uniformly. Manually creating backup policies not only consumes huge human resources, but also cannot solve the problem of sudden increase in storage pool pressure because humans are insensitive to storage pool pressure. At the same time, under the batch plan, backup window times often concentrate on fixed storage pools, which will exacerbate the imbalance of storage pool storage capacity, resulting in unsatisfactory accuracy of data backup control and the problem of insufficient accuracy of data backup control.

[0003] In response to the problem of unsatisfactory accuracy of data backup control in related technologies, no effective solution has been proposed yet. Summary of the Invention

[0004] The main purpose of the present application is to provide a backup method, apparatus, storage medium, and electronic device to solve the problem of unsatisfactory accuracy of data backup control in related technologies.

[0005] To achieve the above object, according to one aspect of the present application, a backup method is provided. The method includes: obtaining the current operating state of a storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; based on the current operating state, determining a target address from the N storage addresses by using a target model, where the target model is trained based on historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical storage success rate and historical backup rate of the storage pool; storing the service data to be backed up at the target address.

[0006] Optionally, obtaining the current operating state of the storage pool includes: obtaining the storage hardware health status corresponding to each of the N storage addresses, and the historical occupied storage capacity of the storage pool; determining the transmission bandwidth between the storage pool and the service server that sends service data; based on the historical occupied storage capacity, the transmission bandwidth, and the storage hardware health status corresponding to each of the N storage addresses, determining the current operating state.

[0007] Optionally, after storing the service data to be backed up at the target address, the method further includes: obtaining the current storage success rate and the current backup rate of the stored service data; generating current backup feedback information based on the storage success rate and the backup rate; and performing iterative processing on the target model using the current backup feedback information.

[0008] Optionally, performing iterative processing on the target model using the current backup feedback information includes: determining a backup interval period based on the historical usage data of the storage pool; determining the estimated iteration duration of the target model; determining the iteration start time based on the estimated iteration duration and the backup interval period; and performing iterative processing on the target model starting from the iteration start time using the current backup feedback information.

[0009] Optionally, the method further includes: labeling with a historical address based on the historical operating state of the storage pool to obtain training data, where the historical address is the storage address allocated by the storage pool in the historical operating state; inputting the training data into an initial model for processing to obtain a predicted address; determining a prediction loss value of the initial model based on first performance quantization information of the predicted address and second performance quantization information of the historical address, where the prediction loss value is used to adjust iterative processing of the initial model; and training the initial model according to the prediction loss value until, when the new prediction loss value is less than a predetermined threshold, using the trained initial model as the target model.

[0010] Optionally, determining a target address from N storage addresses using the target model based on the current operating state includes: performing processing using the target model based on the current operating state to determine M candidate addresses from the N storage addresses, where M is a positive integer less than or equal to N; determining the security protection level corresponding to each of the M candidate addresses based on the hardware facility information corresponding to each of the M candidate addresses; and determining the target address from the M candidate addresses according to the data risk level corresponding to the service data and the security protection level corresponding to each of the M candidate addresses.

[0011] Optionally, the target model includes a first network for generating a target address and a second network for generating a backup policy, where there is a shared network structure between the first network and the second network, and the first network and the second network respectively correspond to different output layers. Storing the service data to be backed up at the target address includes: performing processing using the second network based on the current operating state to obtain a backup policy, where the backup policy includes full backup or incremental backup processing; and storing the service data at the target address using the backup policy.

[0012] To achieve the above object, according to another aspect of the present application, a backup device is provided. The device includes: an operating state acquisition module, configured to acquire the current operating state of a storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different respectively, and N is a positive integer; a target address determination module, configured to determine a target address from the N storage addresses based on the current operating state by using a target model, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical backup process of the storage pool; and a storage module, configured to store the service data to be backed up at the target address.

[0013] Optionally, the operating state acquisition module includes: a first determination module, configured to acquire the storage hardware health states corresponding to the N storage addresses respectively, and the historical occupied storage capacity of the storage pool; a transmission bandwidth determination module, configured to determine the transmission bandwidth between the storage pool and the service server that sends the service data; and a current state determination module, configured to determine the current operating state based on the historical occupied storage capacity, the transmission bandwidth, and the storage hardware health states corresponding to the N storage addresses respectively.

[0014] Optionally, the device further includes: a second determination module, configured to acquire the current storage success rate and the current backup rate of storing the service data; a feedback information determination module, configured to generate the current backup feedback information based on the storage success rate and the backup rate; and a model iteration module, configured to perform iterative processing on the target model by using the current backup feedback information.

[0015] Optionally, the model iteration module includes: a third determination module, configured to determine the backup interval period based on the historical usage data of the storage pool; an iteration duration determination module, configured to determine the expected iteration duration of the target model; an iteration start time determination module, configured to determine the iteration start time based on the expected iteration duration and the backup interval period; and an iteration processing module, configured to perform iterative processing on the target model from the iteration start time by using the current backup feedback information.

[0016] Optionally, the device further includes: a training data acquisition module, configured to label the historical operating state of the storage pool by using the historical addresses to obtain training data, where the historical addresses are the storage addresses allocated by the storage pool in the historical operating state; a predicted address determination module, configured to input the training data into an initial model for processing to obtain a predicted address; a loss value determination module, configured to determine the prediction loss value of the initial model based on the first performance quantization information of the predicted address and the second performance quantization information of the historical address, where the prediction loss value is used to adjust the iterative processing of the initial model; and a target model determination module, configured to train the initial model according to the prediction loss value until the new prediction loss value is less than a predetermined threshold, and then use the trained initial model as the target model.

[0017] Optionally, the target address determination module includes: a candidate address determination module, configured to process based on the current operating state using a target model to determine M candidate addresses from N storage addresses, where M is a positive integer less than or equal to N; a security level determination module, configured to determine the security protection levels corresponding to the M candidate addresses respectively based on the hardware facility information corresponding to the M candidate addresses; and a fourth determination module, configured to determine a target address from the M candidate addresses according to the data risk level corresponding to the service data and the security protection levels corresponding to the M candidate addresses respectively.

[0018] Optionally, the storage module includes: a backup policy determination module, configured to process based on the current operating state using a second network to obtain a backup policy, where the backup policy includes full backup or incremental backup processing; and a data storage module, configured to store the service data to the target address using the backup policy.

[0019] To achieve the above object, according to one aspect of the present application, there is provided a computer-readable storage medium, where the computer-readable storage medium includes an executable program stored therein, and when the executable program runs, it controls the device where the computer-readable storage medium is located to execute any one of the backup methods.

[0020] To achieve the above object, according to one aspect of the present application, there is provided an electronic device, including: a memory storing an executable program; and a processor configured to run the program, where when the program runs, it executes any one of the backup methods.

[0021] In the embodiment of the present application, by adopting the method of adaptive backup based on the storage pool pressure feedback, by obtaining the current operating state of the storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; based on the current operating state, using a target model to determine a target address from the N storage addresses, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical storage success rate and historical backup rate of the storage pool; storing the service data to be backed up to the target address. The purpose of adaptively adjusting the backup policy and backup address according to the historical backup result of the storage pool, the current operating state of the storage pool, and the risk level of the backup service data is achieved, thereby realizing the technical effects of improving the storage pool pressure load balance and improving the accuracy of data backup control, and further solving the technical problem that the accuracy of data backup control is not ideal. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings, which form a part of this application, are used to provide a further understanding of this application. The schematic embodiments and descriptions thereof of this application are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0023] Figure 1 A hardware structure block diagram of a computer terminal for implementing a backup method is shown;

[0024] Figure 2 It is a flowchart of a backup method provided according to an embodiment of this application;

[0025] Figure 3 It is a schematic diagram of a backup method provided according to an embodiment of this application;

[0026] Figure 4 It is a schematic diagram of a backup device provided according to an embodiment of this application;

[0027] Figure 5 It is a structure block diagram of an electronic device according to an embodiment of this application. Detailed implementation manners

[0028] In order to enable those skilled in the art of this technology to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the scope of protection of this application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above accompanying drawings are used to distinguish similar objects and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these process, method, product or device.

[0030] First, some nouns or terms that appear in the process of describing the embodiments of this application are applicable to the following explanations:

[0031] Full backup means that during the backup process, all data at a specified point in time is copied, regardless of whether the data has changed during this period. Full backup has the advantages of simple and fast data recovery and high data consistency.

[0032] Incremental backup means that after the first full backup, only the data that has changed since the last backup is backed up. Incremental backup has the advantages of saving storage space and reducing backup time.

[0033] It should be noted that the information collected in this application (including but not limited to storage hardware device information, storage pool running status information, and prior backup data, etc.) and data (including but not limited to data for display, analysis, backup feedback data, business risk level data, etc.) are information and data authorized by the user or fully authorized by all parties. And the processing of relevant data, such as collection, storage, use, processing, transmission, provision, disclosure, and application, all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse. For example, there are interfaces set between this system and relevant users or institutions to provide corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making; if the user chooses to refuse, it will enter the expert decision-making process.

[0034] Embodiment 1

[0035] According to the embodiments of the present application, a method embodiment of a backup method is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0036] The method embodiment provided by the first embodiment of the present application can be executed on a mobile terminal, a computer terminal, or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal (or mobile device) for implementing the backup method is shown. As Figure 1 shown, the computer terminal 10 (or mobile device) may include one or more (shown as 102a, 102b,..., 102n in the figure) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may further include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the BUS bus), a network interface, a power supply, and / or a camera. Those of ordinary skill in the art can understand, Figure 1The structure shown is only schematic and does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 may also include more or fewer components than those shown in Figure 1 and may have a configuration different from that shown in Figure 1 .

[0037] It should be noted that one or more of the above-mentioned processors 102 and / or other data processing circuits can generally be referred to as "data processing circuits" herein. The data processing circuit can be embodied in software, hardware, firmware, or any combination thereof, in whole or in part. In addition, the data processing circuit can be a single independent processing module, or be incorporated in whole or in part into any one of the other elements in the computer terminal 10 (or mobile device). As involved in the embodiments of the present application, the data processing circuit is a kind of processor control (such as the selection of a variable resistance terminal path connected to an interface).

[0038] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage devices corresponding to the backup method in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby implementing the above-mentioned backup method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely disposed relative to the processor 102, and these remote memories can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, enterprise intranets, local area networks, mobile communication networks, and combinations thereof.

[0039] The transmission device 106 is used to receive or send data via a network. Specific examples of the above-mentioned network may include the wireless network provided by the communication provider of the computer terminal 10. In one instance, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station and thus can communicate with the Internet. In one instance, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0040] The display can be, for example, a touch-screen liquid crystal display (LCD), which enables the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0041] Under the above operating environment, the present application provides a backup method as shown in Figure 2 . Figure 2It is a flowchart of a backup method provided according to an embodiment of the present application.

[0042] Step S201, obtain the current running state of the storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer;

[0043] It can be understood that the current running state data of the storage pool is obtained. The above storage pool running state data includes storage pool water level, storage pool read and write performance, storage pool CPU (Central Processing Unit) usage rate, backup rate, etc. The storage pool contains N storage addresses, and each storage address corresponds to different storage hardware. The storage hardware is used to store the business data to be backed up, and each storage hardware has different storage performances. By obtaining the running state of the storage pool in real time, it provides key information input for the adjustment of the adaptive backup strategy, enabling it to make intelligent decisions, achieve optimal allocation of resources, thereby improving the backup efficiency, reducing the pressure on the storage pool, and enhancing the stability and reliability of the system.

[0044] Optionally, in the backup method provided by the embodiment of the present application, obtaining the current running state of the storage pool includes: obtaining the health status of the storage hardware corresponding to the N storage addresses respectively, and the historical occupied storage capacity of the storage pool; determining the transmission bandwidth between the storage pool and the business server that sends the business data; and determining the current running state based on the historical occupied storage capacity, the transmission bandwidth, and the health status of the storage hardware corresponding to the N storage addresses respectively.

[0045] It can be understood that the health status of the storage hardware corresponding to the N storage addresses in the storage pool and the historical occupied storage capacity of the storage pool are obtained. The health status of the storage hardware is used to reflect the health level of the storage hardware, including whether there are hardware fault warnings, signs of performance degradation, etc.; the historical occupied storage capacity of the storage pool is used to determine the remaining storage capacity of the storage pool. Determine the transmission bandwidth between the storage pool and the business server corresponding to the business data to be stored. The above transmission bandwidth can be used to determine the speed and efficiency of backing up the business data. Based on the historical occupied storage capacity, the transmission bandwidth, and the health status of the storage hardware corresponding to the N storage addresses in the storage pool, determine the current running state of the storage pool. By obtaining detailed information on the current running state of the storage pool, the adaptive backup adjustment method can better understand the real-time situation and historical performance of the storage pool, thereby making more accurate decisions, optimizing the backup strategy, improving the efficiency of data backup and recovery, effectively controlling the cost of storage and network resources, and enhancing the security and stability of the system.

[0046] Step S202: Based on the current running state, determine the target address from N storage addresses using the target model, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical storage success rate and historical backup rate of the storage pool;

[0047] It can be understood that a machine learning model is trained with a training data set composed of the historical storage success rate and historical backup rate of the historical backup results as historical feedback information to obtain the target model. Using the target model, based on the current running state of the storage pool, determine the storage address corresponding to the storage hardware for storing business data from N storage addresses, that is, the target address. The adaptive backup adjustment method based on the storage pool pressure feedback can manage the data backup process more intelligently and efficiently, ensuring that the requirements for data backup and recovery are met under the background of big data, while reasonably controlling the pressure on storage capacity and performance.

[0048] Optionally, in the backup method provided in the embodiments of the present application, the method further includes: Based on the historical running state of the storage pool, label with historical addresses to obtain training data, where the historical addresses are the storage addresses allocated by the storage pool in the historical running state; input the training data into the initial model for processing to obtain predicted addresses; based on the first performance quantization information of the predicted addresses and the second performance quantization information of the historical addresses, determine the prediction loss value of the initial model, where the prediction loss value is used to adjust the iterative processing of the initial model; train the initial model according to the prediction loss value until the new prediction loss value is less than a predetermined threshold, and then use the trained initial model as the target model.

[0049] It can be understood that according to the historical backup process of the storage pool, the historical operating status information of the storage pool and the corresponding historical addresses are obtained, where the historical addresses are the storage addresses allocated by the storage pool in the historical operating status. The historical operating status information corresponding to the historical addresses is used for annotation respectively to obtain the annotated historical operating status information data set, which is used as the training data set of the target model. The above training data set is input into the initial model to obtain the predicted addresses, where the initial model can adopt a supervised machine learning model. According to the predicted addresses obtained from the training data, the storage performance of the predicted addresses is quantitatively processed to obtain the first performance quantization information corresponding to the predicted addresses. Similarly, the storage performance of the historical addresses corresponding to the predicted addresses is quantitatively processed to obtain the second performance quantization information. Based on the first performance quantization information and the second performance quantization information, the prediction loss value of the initial model is obtained. The prediction loss value is compared with a predetermined threshold: if the prediction loss value is less than the predetermined threshold, the initial model trained at this time is used as the target model; if the prediction loss value is greater than or equal to the predetermined threshold, the model parameters are adjusted to optimize the model, and the model is continuously trained using the training data. Training the model based on the historical operating status data of the storage pool realizes the function of intelligent prediction of storage addresses, improves the efficiency and reliability of data backup, and at the same time enhances the stability and performance of the system through resource balancing and fault prevention.

[0050] Optionally, first, according to the historical backup process of the storage pool, the historical operating status information of the storage pool and the corresponding historical addresses are collected, and the historical operating status information corresponding to the historical addresses is used for annotation, that is, the storage pool addresses actually allocated by past backup tasks are used as the labels of the training data. The collected data is preprocessed and feature extracted. For example, time series data is converted into fixed-length feature vectors, missing values are filled, outliers are processed, and the data is normalized or standardized, etc. The annotated historical operating status information is formed into a training set, where each data point contains a set of feature vectors (i.e., the historical operating status information after feature extraction) and a label (i.e., the storage pool address). The training set data is used to train a machine learning model. The goal of model training is to learn a function or model that can predict the most suitable storage pool address for backup according to the current operating status of the storage pool. After the initial training, the model may need to go through multiple iterations and optimizations to improve the prediction accuracy. Ways such as adjusting the model parameters, selecting a better feature subset, and using a more advanced algorithm architecture can be adopted to improve the prediction accuracy.

[0051] Optionally, the predicted address and the corresponding historical address can be quantified according to metrics such as the read / write speed of data, response latency, storage capacity usage, CPU load, etc., to obtain the first performance quantization information and the second performance quantization information. According to the first performance quantization information and the second performance quantization information, using a loss function, the predicted loss value of the predicted address is obtained. The loss function is used to measure the gap between the predicted value and the actual value of the model. The formula of the loss function is:

[0052]

[0053] where L represents the predicted loss value; y exp respectively represent the first performance quantization information and the second performance quantization information; w(θ) represents the set of input features; y r represents the execution time of the actual production feedback; λ is a parameter constant.

[0054] Optionally, in the backup method provided in the embodiments of the present application, determining a target address from N storage addresses using a target model based on the current running state includes: based on the current running state, using the target model for processing to determine M candidate addresses from the N storage addresses, where M is a positive integer less than or equal to N; determining the security protection levels corresponding to the M candidate addresses respectively based on the hardware facility information corresponding to the M candidate addresses; and determining the target address from the M candidate addresses according to the data risk level corresponding to the service data and the security protection levels corresponding to the M candidate addresses respectively.

[0055] It can be understood that according to the current running state of the storage pool, using the trained target model, M addresses are determined from the N storage addresses as candidate addresses. Obtain the hardware facility information corresponding to the M candidate addresses respectively, and based on the hardware facility information and the security level evaluation mechanism of the storage pool management system, determine the security protection levels corresponding to the M candidate addresses respectively. Determine the risk level of the data according to the characteristics of the service data. Based on the risk level of the service data and the security protection levels of the M candidate addresses, determine the target address from the M candidate addresses. According to the current running state, M candidate storage addresses are intelligently screened, avoiding the blind screening of all addresses, improving the efficiency and accuracy of backup strategy formulation. At the same time, by evaluating the hardware facility information and the security protection level, the security of the service data during storage is ensured, preventing data leakage, tampering, and loss.

[0056] Optionally, the above-mentioned hardware facility information includes the type, model, redundancy configuration, security features, etc. of the storage device; the characteristics of the above-mentioned business data include data sensitivity, data value, data usage frequency, etc. Match the risk level of the business data with the security protection levels of the M candidate addresses, and select the storage address that can provide an appropriate security protection level as the target address. If multiple candidate addresses meet the requirements simultaneously, the target address can be further selected according to other factors such as storage capacity and read / write performance, so as to achieve the best backup effect. By matching the data risk level with the security protection level of the storage address, differential security management of different business data is realized, and the overall risk level is reduced.

[0057] Step S203: Store the business data to be backed up at the target address.

[0058] It can be understood that the target address corresponding to the storage hardware for backing up the business data to be backed up is obtained by using the target model, and the business data to be backed up is stored at the target address according to the determined backup strategy. By intelligently storing the business data at the target address, the optimal allocation of backup resources is realized, the security and efficiency of data backup are improved, and at the same time, the load balancing and fault recovery capabilities of the storage pool are enhanced, which has a significant effect on improving the stability of the entire system and the data protection level.

[0059] Optionally, in the backup method provided in the embodiment of the present application, after storing the business data to be backed up at the target address, the method further includes: obtaining the current storage success rate and the current backup rate of storing the business data; generating the current backup feedback information based on the storage success rate and the backup rate; and performing iterative processing on the target model by using the current backup feedback information.

[0060] It can be understood that after storing the business data to be backed up at the target address, the current storage success rate and the current backup rate of storing the business data are obtained, and the current backup feedback information is obtained based on the storage success rate and the backup rate. The target model is iteratively processed by using the current backup feedback information to update the parameters and weights of the model, and the updated model is verified and tested. By collecting real-time feedback information and performing model iteration, the prediction ability of the target model can be continuously optimized to make it more adaptable to the dynamic changes of the storage pool and the backup requirements of business data. At the same time, the model iteration processing can enhance the adaptability of the backup system, enabling it to quickly respond to real-time situations such as changes in storage pool pressure and fluctuations in business data volume, and effectively avoiding problems such as storage capacity alarms and backup failures.

[0061] Optionally, in the backup method provided by the embodiments of the present application, the target model is iteratively processed using the current backup feedback information, including: determining a backup interval period based on the historical usage data of the storage pool; determining the estimated iteration duration of the target model; determining the iteration start time based on the estimated iteration duration and the backup interval period; and performing iterative processing on the target model from the iteration start time using the current backup feedback information.

[0062] It can be understood that based on the historical usage data of the storage pool, the periodicity and volatility of data backup requirements are analyzed to determine the backup interval period. According to the training requirements of the model and the change speed of business data, the estimated iteration duration of the target model is determined. Based on the estimated iteration duration and the backup interval period, the iteration start time is determined. At the iteration start time, the backup feedback information is input into the target model for model optimization and update. The iterative processing of the target model enables the adaptive backup adjustment method to be dynamically adjusted to cope with changes in the business environment, such as sudden growth in data volume, performance decline of storage hardware, etc., improving the flexibility and adaptability of the backup strategy, optimizing the allocation of storage resources, avoiding overloading of the storage pool, increasing the utilization rate of storage space, and extending the service life of storage devices.

[0063] Optionally, in the backup method provided by the embodiments of the present application, the target model includes a first network for generating a target address and a second network for generating a backup strategy, where there is a shared network structure between the first network and the second network, and the first network and the second network respectively correspond to different output layers. Storing the business data to be backed up at the target address includes: using the second network to process based on the current operating state to obtain a backup strategy, where the backup strategy includes full backup or incremental backup processing; and storing the business data at the target address using the backup strategy.

[0064] It can be understood that the target model consists of two networks, and the two networks respectively correspond to two output layers. The output layer of the first network is used to output the target address of the business data to be backed up, and the output layer of the second network is used to output the backup strategy of the business data to be backed up, and there is a shared network structure between the first network and the second network. Based on the current operating state of the storage pool, the target model is used to obtain the target address and backup strategy of the business data to be backed up. According to the backup strategy, it is determined whether to perform full backup or incremental backup on the business data to be backed up, and it is backed up to the storage hardware corresponding to the target address. By designing a dual-network structure, the system can optimize both the backup strategy and the target address selection simultaneously, ensuring the maximization of backup efficiency and storage resource utilization, and the shared network structure enables the model to understand data from the perspectives of cross-strategy and address, improving the coordination and intelligence of overall decision-making.

[0065] Optionally, in the dual-network structure target model with a shared network structure, the two networks may share weights in some hidden layers or feature extraction layers to utilize the same data preprocessing and feature representation. At the same time, there are two different output layers, which are respectively used to output the target address and the backup strategy. First, the data is input into the shared preprocessing layer for feature extraction, and the extracted features are shared by the two networks. Based on the extracted features, the first network is used to output the target address, and the second network is used to output the backup strategy. For the update and optimization of the model, the outputs of the two networks can be used as feedback signals to adjust the weights of the shared layer. The loss functions of the two networks can also be jointly optimized to achieve the balance of multiple tasks. The use of the shared layer in the dual-network structure reduces redundant learning, enables the model to converge faster, and at the same time, can better handle problems containing multiple interdependent subtasks, providing a more comprehensive solution for complex problems.

[0066] Through the above step S201, the current running state of the storage pool is obtained, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; in step S202, based on the current running state, the target model is used to determine the target address from the N storage addresses, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical storage success rate and historical backup rate of the storage pool; in step S203, the business data to be backed up is stored at the target address. It can achieve the purpose of adaptively adjusting the backup strategy and backup address according to the historical backup result of the storage pool, the current running state of the storage pool, and the risk level of the business data to be backed up, and achieve the technical effects of improving the pressure load balance of the storage pool and the accuracy of data backup control, thereby solving the technical problem of unsatisfactory accuracy of data backup control.

[0067] Based on the above embodiments and optional embodiments, the present application proposes an optional implementation manner for the backup processing of business data. According to the backup method provided by the embodiments of the present application, a backup adaptive adjustment method is provided. This method takes various indicators such as the storage pool water level and read / write performance (i.e., the running state information of the storage pool) as inputs, and based on a machine learning model, performs intelligent collection and management of backup tasks, and issues them to the backup management server by the backup collection system for automatic backup. The machine learning model initially learns using the backup strategy set by the prior knowledge of the backup management personnel, and is optimized and iterated according to the backup execution situation feedback by the backup management server. Figure 3 is a schematic diagram of the backup method provided by the embodiments of the present application, as Figure 3As shown in the figure, the backup method provided by the embodiment of the present application mainly includes three parts: a storage pool management part, a machine learning model, and a backup management part. Among them, the storage pool management part is used to detect and manage the health status and pressure of the storage pool, and collect storage pool water level, read / write performance capacity, and CPU usage metrics. The machine learning model is used to train the model by using the experience strategies of backup management personnel, the feedback information after backup execution, and the storage pool operation status information, and at the same time use the prediction function provided by the model to generate new backup strategies. The backup management part is used to receive the backup strategy predicted and issued by the machine learning model device, execute the backup task within the specified time, and feedback the backup execution status (success or failure, backup rate) after the backup task ends.

[0068] As Figure 3 shown, the backup method provided by the embodiment of the present application includes the following steps.

[0069] Step 1, the storage pool management device actively collects the current operation status information of the storage pool, including health status, pressure situation information, etc.;

[0070] Obtain the current operation status data of the storage pool, including health status and pressure situation information. The health status and pressure situation of the storage pool can be measured by information such as the storage pool water level, the read / write performance of the storage pool, the CPU (Central Processing Unit) usage rate of the storage pool, and the backup rate. The storage pool contains N storage addresses, and each storage address corresponds to different storage hardware. The storage hardware is used to store the business data of the backup, and each storage hardware has different storage performance. By obtaining the operation status of the storage pool in real time, it provides key information input for the adjustment of the adaptive backup strategy, enabling it to make intelligent decisions, achieve the optimal allocation of resources, thereby improving the backup efficiency, reducing the pressure on the storage pool, and enhancing the stability and reliability of the system.

[0071] Step 2, train the model based on the predicted values provided by prior knowledge (i.e., expert experience);

[0072] According to the historical backup process of the storage pool, obtain the historical operating status information of the storage pool and the corresponding historical addresses, where the historical addresses are the storage addresses allocated by the storage pool in the historical operating status. The above historical operating status information and the corresponding historical addresses are the predicted values provided based on prior knowledge (i.e., expert experience). Use the historical operating status information corresponding to the historical addresses respectively for annotation to obtain the annotated historical operating status information data set, and use it as the training data set for the target model. Input the above training data set into the initial model to obtain the predicted addresses, where the initial model can adopt a supervised machine learning model. According to the predicted addresses obtained from the training data, perform quantization processing on the storage performance of the predicted addresses to obtain the first performance quantization information corresponding to the predicted addresses. Similarly, perform quantization processing on the predicted addresses and the corresponding historical addresses to obtain the first performance quantization information and the second performance quantization information. Based on the first performance quantization information and the second performance quantization information, obtain the prediction loss value of the initial model. Compare the prediction loss value with a predetermined threshold: if the prediction loss value is less than the predetermined threshold, then use the initial model trained at this time as the target model; if the prediction loss value is greater than or equal to the predetermined threshold, then adjust the model parameters to optimize the model, and continue to train the model using the training data. Training the model based on the historical operating status data of the storage pool realizes the function of intelligent prediction of storage addresses, improves the efficiency and reliability of data backup, and at the same time enhances the stability and performance of the system through resource balancing and fault prevention.

[0073] First, according to the historical backup process of the storage pool, collect the historical operating status information of the storage pool and the corresponding historical addresses, and use the historical operating status information corresponding to the historical addresses for annotation, that is, use the storage pool addresses actually allocated by past backup tasks as the labels of the training data. Preprocess and extract features from the collected data. For example, convert time series data into fixed-length feature vectors, fill in missing values, process outliers, and normalize or standardize the data, etc. Compose the annotated historical operating status information into a training set, where each data point contains a set of feature vectors (i.e., the historical operating status information after feature extraction) and a label (i.e., the storage pool address). Use the training set data to train a machine learning model. The goal of model training is to learn a function or model that can predict the most suitable storage pool address for backup according to the current operating status of the storage pool. After the initial training, the model may need to go through multiple iterations and optimizations to improve the prediction accuracy. Ways such as adjusting the model parameters, selecting a better feature subset, and using a more advanced algorithm architecture can be adopted to improve the prediction accuracy.

[0074] The predicted address and the corresponding historical address can be quantified according to indicators such as the read / write speed of data, response latency, storage capacity usage, and CPU load to obtain the first performance quantization information and the second performance quantization information. Based on the first performance quantization information and the second performance quantization information, using a loss function, the prediction loss value of the predicted address is obtained. The loss function is used to measure the gap between the model prediction value and the actual value. The formula of the loss function is:

[0075]

[0076] Among them, L represents the prediction loss value; y exp respectively represent the first performance quantization information and the second performance quantization information; w(θ) represents the input feature set; y r represents the execution time of the actual production feedback; λ is a parameter constant.

[0077] Step 3, according to the running status information obtained by the storage pool management and the prediction value information provided by the expert experience, use a machine learning prediction model to obtain the storage pool allocation address (i.e., the target address) and the backup strategy;

[0078] The target model (i.e., the machine learning prediction model) consists of two networks. The two networks respectively correspond to two output layers. The output layer of the first network is used to output the target address of the service data to be backed up (i.e., the backup target), and the output layer of the second network is used to output the backup strategy of the service data to be backed up. And there is a shared network structure between the first network and the second network. Use the prediction value information provided by the expert experience to train the initial model to obtain the target model. Based on the current running status of the storage pool, use the target model to obtain the target address of the service data to be backed up and the backup strategy. According to the backup strategy, determine whether to perform a full backup or an incremental backup to back up the service data to be backed up, and back it up to the storage hardware corresponding to the target address. By designing a dual-network structure, the system can optimize the backup strategy and the target address selection simultaneously, ensuring the maximization of backup efficiency and storage resource utilization. And the shared network structure enables the model to understand the data from the perspectives of cross-strategies and addresses, improving the coordination and intelligence of the overall decision-making.

[0079] Step 4, send the backup strategy to the backup management part;

[0080] Use the target model to obtain the target address corresponding to the storage hardware for backing up the service data to be backed up, and send the backup strategy and the target address information to the backup management part.

[0081] Step 5, the backup management part initiates a backup task at a specified time;

[0082] The backup management part executes the backup task at the specified time to complete the backup of the service data to be backed up.

[0083] Step 6, after the backup task is completed, the backup management part feeds back the backup success rate and backup rate (i.e., backup feedback information) according to the backup result, and optimizes the model according to the backup success rate and backup rate.

[0084] After storing the business data to be backed up at the target address, obtain the current storage success rate and current backup rate of the stored business data, and obtain the current backup feedback information based on the storage success rate and backup rate. Use the current backup feedback information to perform iterative processing on the target model, update the parameters and weights of the model, and verify and test the updated model. By collecting real-time feedback information and performing model iteration, the prediction ability of the target model can be continuously optimized to make it more adaptable to the dynamic changes of the storage pool and the backup requirements of business data. At the same time, the model iteration processing can enhance the adaptability of the backup system, enabling it to quickly respond to real-time situations such as changes in storage pool pressure and fluctuations in business data volume, effectively avoiding problems such as storage capacity alarms and backup failures.

[0085] Based on the historical usage data of the storage pool, analyze the periodicity and volatility of data backup requirements, and determine the backup interval period. According to the training requirements of the model and the change speed of business data, determine the expected iteration duration of the target model. Based on the expected iteration duration and backup interval period, determine the iteration start time. At the iteration start time, input the backup feedback information into the target model for model optimization and update. The iterative processing of the target model enables the adaptive backup adjustment method to be dynamically adjusted to cope with changes in the business environment, such as sudden increases in data volume and performance degradation of storage hardware, improving the flexibility and adaptability of the backup strategy, optimizing the allocation of storage resources, avoiding overloading of the storage pool, increasing the utilization rate of storage space, and extending the service life of storage devices.

[0086] The backup method provided by the embodiments of this application overcomes the problem that traditional backup methods do not consider the health of the storage pool, improves the problem of uneven storage pool load easily caused by traditional timed backups, and is more adaptable to the backup requirements of massive data in the big data era.

[0087] The above backup method achieves the following effects: adaptively adjusting the backup strategy and backup address of the storage pool based on the running state information of the storage pool, improving the flexibility and adaptability of the backup strategy, optimizing the allocation of storage resources, avoiding overloading of the storage pool, increasing the utilization rate of storage space, and extending the service life of storage devices; using the labeled prior knowledge as training data to obtain a target machine learning model, and using the target machine learning model to obtain the current backup strategy and backup address, realizing the function of intelligently predicting the backup address and backup strategy, and improving the efficiency and reliability of data backup. Iteratively optimizing the model based on the backup feedback information can continuously optimize the prediction ability of the target model to make it more adaptable to the dynamic changes of the storage pool and the backup requirements of business data.

[0088] The backup method provided by the embodiment of the present application obtains the current running state of the storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; based on the current running state, a target model is used to determine a target address from the N storage addresses, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical storage success rate and historical backup rate of the storage pool; the business data to be backed up is stored in the target address. This solves the problem that the accuracy of data backup control in the related art is not ideal. Furthermore, it achieves the purpose of adaptively adjusting the backup strategy and backup address according to the historical backup result of the storage pool, the current running state of the storage pool, and the risk level of the backup business data, thereby realizing the technical effects of improving the pressure load balance of the storage pool and the accuracy of data backup control.

[0089] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0090] Embodiment 2

[0091] The embodiment of the present application also provides a backup device. It should be noted that the backup device of the embodiment of the present application can be used to execute the backup method provided by the embodiment of the present application. The following introduces the Z device provided by the embodiment of the present application.

[0092] According to the embodiment of the present application, there is also provided a device for implementing the above backup method. Figure 4 is a schematic diagram of the backup device provided by the embodiment of the present application, as Figure 4As shown in the figure, the device includes: an operating status acquisition module 401, configured to acquire the current operating status of the storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; a target address determination module 402, connected to the operating status acquisition module 401, configured to determine a target address from the N storage addresses based on the current operating status by using a target model, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical backup process of the storage pool; a storage module 403, connected to the target address determination module 402, configured to store the service data to be backed up at the target address.

[0093] The backup device provided by the embodiment of the present application includes an operating status acquisition module 401, configured to acquire the current operating status of the storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; a target address determination module 402, connected to the operating status acquisition module 401, configured to determine a target address from the N storage addresses based on the current operating status by using a target model, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical backup process of the storage pool; a storage module 403, connected to the target address determination module 402, configured to store the service data to be backed up at the target address. This solves the problem that the accuracy of data backup control in the related art is not ideal. Furthermore, it achieves the purpose of adaptively adjusting the backup strategy and backup address according to the historical backup result of the storage pool, the current operating status of the storage pool, and the risk level of the service data to be backed up, thereby realizing the technical effects of improving the pressure load balance of the storage pool and improving the accuracy of data backup control.

[0094] Optionally, in the backup device provided by the embodiment of the present application, the operating status acquisition module includes: a first determination module, configured to acquire the storage hardware health status corresponding to each of the N storage addresses, and the historical occupied storage capacity of the storage pool; a transmission bandwidth determination module, configured to determine the transmission bandwidth between the storage pool and the service server that sends the service data; a current status determination module, configured to determine the current operating status based on the historical occupied storage capacity, the transmission bandwidth, and the storage hardware health status corresponding to each of the N storage addresses.

[0095] Optionally, in the backup device provided by the embodiment of the present application, the device further includes: a second determination module, configured to acquire the current storage success rate and the current backup rate of the stored service data; a feedback information determination module, configured to generate current backup feedback information based on the storage success rate and the backup rate; a model iteration module, configured to perform iterative processing on the target model by using the current backup feedback information.

[0096] Optionally, in the backup device provided in the embodiments of the present application, the model iteration module includes: a third determination module, configured to determine a backup interval period based on the historical usage data of the storage pool; an iteration duration determination module, configured to determine the estimated iteration duration of the target model; an iteration start time determination module, configured to determine the iteration start time based on the estimated iteration duration and the backup interval period; and an iteration processing module, configured to perform iteration processing on the target model from the iteration start time by using the current backup feedback information.

[0097] Optionally, in the backup device provided in the embodiments of the present application, the device further includes: a training data acquisition module, configured to obtain training data by using historical addresses for annotation based on the historical operating state of the storage pool, where the historical address is the storage address allocated by the storage pool in the historical operating state; a predicted address determination module, configured to input the training data into an initial model for processing to obtain a predicted address; a loss value determination module, configured to determine a prediction loss value of the initial model based on first performance quantization information of the predicted address and second performance quantization information of the historical address, where the prediction loss value is used to adjust the iteration processing of the initial model; and a target model determination module, configured to train the initial model according to the prediction loss value until the new prediction loss value is less than a predetermined threshold, and then use the trained initial model as the target model.

[0098] Optionally, in the backup device provided in the embodiments of the present application, the target address determination module includes: a candidate address determination module, configured to process based on the current operating state by using the target model to determine M candidate addresses from N storage addresses, where M is a positive integer less than or equal to N; a security level determination module, configured to determine the security protection levels corresponding to the M candidate addresses respectively based on the hardware facility information corresponding to the M candidate addresses; and a fourth determination module, configured to determine a target address from the M candidate addresses according to the data risk level corresponding to the service data and the security protection levels corresponding to the M candidate addresses respectively.

[0099] Optionally, in the backup device provided in the embodiments of the present application, the storage module includes: a backup policy determination module, configured to process based on the current operating state by using a second network to obtain a backup policy, where the backup policy includes full backup or incremental backup processing; and a data storage module, configured to store the service data to the target address by using the backup policy.

[0100] It should be noted here that the above-mentioned running state acquisition module 401, target address determination module 402, and storage module 403 correspond to steps S201 to S203 in Embodiment 1. The functions and application scenarios implemented by the two modules and the corresponding steps are the same, but are not limited to the content disclosed in the above-mentioned Embodiment 1. It should be noted that the above-mentioned module or unit can be a hardware component or software component stored in a memory (for example, memory 104) and processed by one or more processors (for example, processors 102a, 102b,..., 102n). The above-mentioned module can also be part of a device and can run in the computer terminal 10 provided in Embodiment 1.

[0101] Embodiment 3

[0102] An embodiment of the present application can provide an electronic device. Figure 5 It is a structural block diagram of an electronic device according to an embodiment of the present application. As Figure 5 shown, the electronic device may include: one or more ( Figure 5 only one is shown in the figure) processors 502, a memory 504, a storage controller, and a peripheral interface. Among them, the peripheral interface is connected to a radio frequency module, an audio module, and a display.

[0103] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implements the above-mentioned methods. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, a flash memory, or other non-volatile solid-state memories. In some instances, the memory may further include a memory remotely provided relative to the processor, and these remote memories can be connected to the terminal through a network. Examples of the above-mentioned network include but are not limited to the Internet, an enterprise internal network, a local area network, a mobile communication network, and combinations thereof.

[0104] The processor can call the information and application programs stored in the memory through a transmission device to execute the following steps: obtaining the current running state of the storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; based on the current running state, determining a target address from the N storage addresses by using a target model, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical storage success rate and historical backup rate of the storage pool; storing the service data to be backed up at the target address.

[0105] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the storage hardware health status corresponding to N storage addresses respectively, and the historical occupied storage capacity of the storage pool; determine the transmission bandwidth between the storage pool and the service server that sends service data; based on the historical occupied storage capacity, the transmission bandwidth, and the storage hardware health status corresponding to N storage addresses respectively, determine the current operating state.

[0106] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the current storage success rate and the current backup rate of the stored service data; generate the current backup feedback information based on the storage success rate and the backup rate; use the current backup feedback information to perform iterative processing on the target model.

[0107] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: determine the backup interval period based on the historical usage data of the storage pool; determine the estimated iteration duration of the target model; based on the estimated iteration duration and the backup interval period, determine the iteration start time; use the current backup feedback information to perform iterative processing on the target model starting from the iteration start time.

[0108] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: based on the historical operating state of the storage pool, label with historical addresses to obtain training data, where the historical address is the storage address allocated by the storage pool in the historical operating state; input the training data into the initial model for processing to obtain predicted addresses; based on the first performance quantization information of the predicted addresses and the second performance quantization information of the historical addresses, determine the prediction loss value of the initial model, where the prediction loss value is used to adjust the iterative processing of the initial model; train the initial model according to the prediction loss value until the new prediction loss value is less than a predetermined threshold, and then use the trained initial model as the target model.

[0109] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: based on the current operating state, use the target model for processing to determine M candidate addresses among the N storage addresses, where M is a positive integer less than or equal to N; based on the hardware facility information corresponding to the M candidate addresses respectively, determine the security protection levels corresponding to the M candidate addresses respectively; according to the data risk level corresponding to the service data and the security protection levels corresponding to the M candidate addresses respectively, determine the target address among the M candidate addresses.

[0110] The processor can also call the information and application programs stored in the memory through the transmission device to execute the following steps: adopt a second network to process based on the current operating state to obtain a backup policy, where the backup policy includes full backup or incremental backup processing; adopt the backup policy to store the service data to the target address.

[0111] The embodiment of the present application provides a solution for a backup method. By obtaining the current operating state of the storage pool, where the storage pool includes N storage addresses, and the storage performances corresponding to the N storage addresses are different, and N is a positive integer; based on the current operating state, determine the target address from the N storage addresses by using a target model, where the target model is trained based on the historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on the historical storage success rate and historical backup rate of the storage pool; store the service data to be backed up to the target address. Thus, the purpose of adaptively adjusting the backup policy and backup address according to the historical backup result of the storage pool, the current operating state of the storage pool, and the risk level of the service data to be backed up is achieved, and further the technical problem of unsatisfactory pressure load balancing of the storage pool during backup caused by responding to a large number of backup requirements is solved.

[0112] Those of ordinary skill in the art can understand that Figure 5 The structure shown is only for illustration, and the electronic device can also be a terminal device such as a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a handheld computer, and a Mobile Internet Device (MID), a PAD, etc. Figure 5 It does not limit the structure of the above-mentioned electronic device. For example, the electronic device may further include more or fewer components (such as a network interface, a display device, etc.) than those shown Figure 5 in, or have a different configuration from that shown Figure 5 .

[0113] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the relevant hardware of the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disc, etc.

[0114] Embodiment 4

[0115] The embodiment of the present application also provides a storage medium. Optionally, in this embodiment, the above storage medium can be used to save the program code executed by the backup method provided in the first embodiment above.

[0116] Optionally, in this embodiment, the above storage medium may be located in any one of the computer terminals in the computer terminal group in the computer network, or in any one of the mobile terminals in the mobile terminal group.

[0117] The present application also provides a computer program product, which is adapted to execute a program for the backup method steps when executed on a data processing device.

[0118] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0119] In the above embodiments of the present application, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0120] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only illustrative. For example, the above division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of units or modules can be in electrical or other forms.

[0121] The units described as separate components above may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, the functional units in the respective embodiments of the present application may be integrated in a processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0123] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the above methods in various embodiments of this application. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks, or optical discs.

[0124] The above are only the preferred embodiments of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this application.

Claims

1. A backup method, characterized in that: include: Obtaining a current running state of a storage pool, wherein the storage pool includes N storage addresses, the N storage addresses respectively correspond to different storage performances, and N is a positive integer; Based on the current running state, a target address is determined from the N storage addresses using a target model, wherein the target model is trained based on historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on a historical storage success rate and a historical backup rate of the storage pool; The business data to be backed up is stored in the target address.

2. The method according to claim 1, characterized in that The obtaining of the current running status of the storage pool includes: Obtaining the storage hardware health status corresponding to the N storage addresses, and the historical occupied storage capacity of the storage pool; Determining a transmission bandwidth between the storage pool and a service server that sends the service data; The current operation status is determined based on the historical occupied storage capacity, the transmission bandwidth, and the health status of the storage hardware corresponding to the N storage addresses.

3. The method according to claim 1, characterized in that After storing the business data to be backed up at the target address, the method further includes: Obtaining a current storage success rate and a current backup rate for storing the business data; Based on the storage success rate and the backup rate, generating current backup feedback information; The target model is iteratively processed using the current backup feedback information.

4. The method according to claim 3, characterized in that The adopting the current backup feedback information to iteratively process the target model includes: Determining a backup interval time period based on historical usage data of the storage pool; Determining an estimated iteration duration of the target model; Determine an iteration start time based on the estimated iteration duration and the backup interval time period; The current backup feedback information is used to perform iterative processing on the target model from the iteration start time.

5. The method according to claim 1, characterized in that The method further comprises: Based on the historical operation status of the storage pool, the historical address is used for marking to obtain training data, wherein the historical address is the storage address allocated by the storage pool under the historical operation status; Input the training data into the initial model for processing to obtain a predicted address; Determining a prediction loss value of the initial model based on the first performance quantification information of the predicted address and the second performance quantification information of the historical address, wherein the prediction loss value is used to adjust the iterative processing of the initial model; The initial model is trained according to the predicted loss value until a new predicted loss value is less than a predetermined threshold, and the trained initial model is used as the target model.

6. The method according to claim 1, characterized in that The determining a target address from the N storage addresses using a target model based on the current running state includes: Based on the current running state, the target model is used for processing to determine M candidate addresses from the N storage addresses, where M is a positive integer less than or equal to N; Determine the security protection levels corresponding to the M candidate addresses based on the hardware facility information corresponding to the M candidate addresses respectively; The target address is determined from the M candidate addresses according to the data risk level corresponding to the business data and the security protection levels respectively corresponding to the M candidate addresses.

7. The method according to any one of claims 1 to 6, characterized in that: The target model includes a first network for generating a target address, and a second network for generating a backup strategy, wherein the first network and the second network have a shared network structure, the first network and the second network correspond to different output layers, respectively, and storing the service data to be backed up to the target address includes: Using the second network, processing based on the current running state to obtain the backup strategy, wherein the backup strategy includes full backup or incremental backup processing; The backup strategy is adopted to store the service data to the target address.

8. A backup device, characterized in that: include: An operation status acquisition module, used to acquire the current operation status of the storage pool, wherein the storage pool includes N storage addresses, the N storage addresses respectively correspond to different storage performances, and N is a positive integer; a target address determination module, configured to determine a target address from the N storage addresses using a target model based on the current running state, wherein the target model is obtained by training based on historical backup feedback information of the storage pool, and the historical backup feedback information is obtained based on a historical backup process of the storage pool; The storage module is used to store the business data to be backed up to the target address.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the computer-readable storage medium is located is controlled to execute the backup method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 7 when running.

Images