Remote copying method and device, computer equipment, medium and product

Abstract

The invention relates to a remote copying method and device, computer equipment, a medium and a product. The method comprises the following steps: acquiring a remote replication comprehensive index value of each logical volume in a slave end cluster, selecting a target logical volume from each logical volume according to the remote replication comprehensive index value of each logical volume, and fully synchronizing data in a source logical volume of a master end cluster to the target logical volume to complete remote replication. By adopting the method, the target logical volume corresponding to the source logical volume can be reasonably selected from all the logical volumes of the slave-end cluster based on the remote replication comprehensive index value of each logical volume in the slave-end cluster, and remote replication between the source logical volume in the master-end cluster and the target logical volume in the slave-end cluster is realized on the basis of the target logical volume. The remote copying performance can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and particularly to a remote replication method, apparatus, computer device, medium and product. Background Art

[0002] With the development of business, enterprises pay more and more attention to the demand for business continuity. Among them, remote replication, as an important technology to ensure business continuity, is widely used.

[0003] In related technologies, after the master cluster enables the remote replication function, all the data in the source logical volume is synchronously replicated to the target logical volume in the slave cluster in full volume, or the differential data in the source logical volume is synchronously replicated to the target logical volume in the slave cluster.

[0004] However, the unreasonable selection of the target logical volume in the slave cluster in related technologies will result in low performance of remote replication. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a remote replication method, apparatus, computer device, medium and product, which can reasonably select the target logical volume in the slave cluster when remotely replicating the data in the source logical volume in the master cluster, so as to improve the performance of remote replication.

[0006] In a first aspect, an embodiment of the present application provides a remote replication method, and the method includes:

[0007] Obtain the remote replication comprehensive index values of each logical volume in the slave cluster;

[0008] Select a target logical volume from each logical volume according to the remote replication comprehensive index values of each logical volume;

[0009] Synchronously replicate all the data in the source logical volume of the master cluster to the target logical volume to complete the remote replication.

[0010] In the technical solution of the embodiment of the present application, the remote replication comprehensive index values of each logical volume in the slave cluster are obtained. According to the remote replication comprehensive index values of each logical volume, a target logical volume is selected from each logical volume, and the data in the source logical volume of the master cluster is fully synchronized to the target logical volume to complete remote replication. The above method can reasonably select a target logical volume corresponding to the source logical volume from all the logical volumes of the slave cluster based on the remote replication comprehensive index values of each logical volume in the slave cluster. Based on this, the remote replication between the source logical volume in the master cluster and the target logical volume in the slave cluster is realized, which can improve the performance of remote replication. At the same time, when selecting the target logical volume of the slave cluster, the above method does not require manual participation, thereby reducing the manual participation error, greatly improving the rationality of the selection of the target logical volume and the availability of the selected target logical volume, reducing the abnormal risk of remote replication, and maximizing the utilization of the resources of the existing logical volumes in the slave cluster.

[0011] In one embodiment, obtaining the remote replication comprehensive index values of each logical volume in the slave cluster includes:

[0012] Obtain the configuration attributes of each logical volume and the target configuration attributes of the target logical volume;

[0013] Process the configuration attributes of each logical volume and the target configuration attributes through a pre-constructed logical volume recommendation model to obtain the remote replication comprehensive index values of each logical volume.

[0014] In the technical solution of the embodiment of the present application, the configuration attributes of each logical volume and the target configuration attributes of the target logical volume are obtained, and the configuration attributes of each logical volume and the target configuration attributes are processed through a pre-constructed logical volume recommendation model to obtain the remote replication comprehensive index values of each logical volume. The above method can obtain the remote replication comprehensive index values of each logical volume in the slave cluster through the logical volume recommendation model, which can improve the acquisition speed of the remote replication comprehensive index values. And this process does not require manual participation, can also reduce the manual participation error, can also improve the accuracy of the obtained remote replication comprehensive index values, and can prepare for further selecting a reasonable target logical volume.

[0015] In one embodiment, the logical volume recommendation model includes a performance prediction model and an evaluation model. Processing the configuration attributes of each logical volume and the target configuration attributes through a pre-constructed logical volume recommendation model to obtain the remote replication comprehensive index values of each logical volume includes:

[0016] For any logical volume, input the configuration attributes of the logical volume and the target configuration attributes into the performance prediction model, and output the performance index vector of the logical volume. The performance index vector includes the values of multiple remote replication performance indicators;

[0017] Input the performance metric vector into the evaluation model to output the comprehensive remote replication metric value of the logical volume.

[0018] In the technical solution of the embodiments of the present application, for any logical volume, input the configuration attributes and target configuration attributes of the logical volume into the performance prediction model to output the values of multiple remote replication performance metrics of the logical volume, and then input the performance metric vector into the evaluation model to output the comprehensive remote replication metric value of the logical volume; the above method can predict the values of the multi-dimensional remote replication performance metrics of the logical volume in the slave cluster, and then convert the values of the multi-dimensional remote replication performance metrics of the logical volume into the one-dimensional comprehensive remote replication metric value of the logical volume, so as to provide a reference basis for further determining whether the logical volume can be used as the target logical volume to be recommended; at the same time, the above method can be implemented through the performance prediction model and the evaluation model, which can improve the acquisition speed of the comprehensive remote replication metric value, and this process does not require manual participation, can also reduce the manual participation error, and can also improve the accuracy of the obtained comprehensive remote replication metric value.

[0019] In one embodiment, selecting the target logical volume from each logical volume according to the comprehensive remote replication metric value of each logical volume includes:

[0020] Determine the logical volume whose comprehensive remote replication metric value in each logical volume exceeds the preset threshold as the target logical volume; or,

[0021] If the comprehensive remote replication metric values of all logical volumes are less than the preset threshold, then create a logical volume corresponding to the source logical volume and determine it as the target logical volume.

[0022] In the technical solution of the embodiments of the present application, determine the logical volume whose comprehensive remote replication metric value in each logical volume exceeds the preset threshold as the target logical volume. If the comprehensive remote replication metric values of all logical volumes are less than the preset threshold, then create a logical volume corresponding to the source logical volume and determine it as the target logical volume; the above method can flexibly determine the target logical volume through different target logical volume determination strategies, thereby improving the wide applicability of selecting the target logical volume and also improving the success rate of determining the target logical volume.

[0023] In one embodiment, fully synchronizing the data in the source logical volume of the master cluster to the target logical volume includes:

[0024] Establish a remote replication relationship according to the source logical volume and the target logical volume in the master cluster;

[0025] Based on the remote replication relationship, fully synchronize the data in the source logical volume to the target logical volume.

[0026] In the technical solution of the embodiment of the present application, a remote replication relationship is established based on the source logical volume and the target logical volume in the master-end cluster, and based on the remote replication relationship, the data in the source logical volume is synchronously replicated in full to the target logical volume; the above method can first establish a remote replication relationship, and then synchronize the data in the source logical volume to the target logical volume based on the remote replication relationship, which can avoid the problem of incorrect data synchronization location caused by synchronizing the data in the source logical volume to other logical volumes, and can improve the accuracy of data synchronization; at the same time, the above method can also synchronize the data in the source logical volume to the target logical volume based on the remote replication relationship, so that the success rate of data synchronization can be improved through the remote replication relationship.

[0027] In one embodiment, the above method further includes:

[0028] Detect the status of the target logical volume;

[0029] If the status of the target logical volume is abnormal and in the data not transmitted state, perform remote replication according to the replication mode and abnormal recovery duration of the target logical volume;

[0030] If the status of the target logical volume is abnormal and in the data transmission state, record the remote replication failure event and cancel the remote replication process of the target logical volume to make the target logical volume in the data not transmitted state.

[0031] In the technical solution of the embodiment of the present application, the status of the target logical volume is detected. If the status of the target logical volume is abnormal and in the data not transmitted state, perform remote replication according to the replication mode and abnormal recovery duration of the target logical volume. If the status of the target logical volume is abnormal and in the data transmission state, record the remote replication failure event and cancel the remote replication process of the target logical volume to make the target logical volume in the data not transmitted state; when the target logical volume is abnormal during the remote replication process, the above method can adopt different exception handling methods according to different transmission states of the target logical volume to achieve remote replication, thereby improving the success rate of remote replication; at the same time, the above method does not require complex algorithms to participate, thereby reducing the amount of data processed during the remote replication process and improving the speed and efficiency of remote replication.

[0032] In one embodiment, performing remote replication according to the replication mode and abnormal recovery duration of the target logical volume includes:

[0033] If the replication mode is the asynchronous replication mode, perform remote replication based on the abnormal recovery duration of the target logical volume;

[0034] If the replication mode is the synchronous replication mode, change the synchronous replication mode to make the target logical volume in the asynchronous replication mode.

[0035] In the technical solution of the embodiment of the present application, if the replication mode is an asynchronous replication mode, remote replication is performed based on the exception recovery duration of the target logical volume. If the replication mode is a synchronous replication mode, the synchronous replication mode is changed so that the target logical volume is in the asynchronous replication mode. The above method can adopt corresponding exception handling methods to implement remote replication according to different replication modes of the target logical volume, thereby improving the success rate of remote replication of the target logical volume.

[0036] In one embodiment, performing remote replication based on the exception recovery duration of the target logical volume includes:

[0037] In the case where the exception type of the target logical volume is an offline failure and the exception recovery duration is less than the duration threshold, perform consistency verification on the data in the source logical volume and the target logical volume;

[0038] If the consistency verification fails, based on the remote replication relationship, synchronize all the data in the source logical volume to the target logical volume in full.

[0039] In the technical solution of the embodiment of the present application, in the case where the exception type of the target logical volume is an offline failure and the exception recovery duration is less than the duration threshold, perform consistency verification on the data in the source logical volume and the target logical volume. When the consistency verification fails, based on the remote replication relationship, synchronize all the data in the source logical volume to the target logical volume in full. The above method can ensure that after the exception recovery of the target logical volume, all the data in the source logical volume is accurately synchronized to the target logical volume through the consistency verification method, ensuring the consistency of the remote replication data after restoring the remote replication relationship, and avoiding the problem that some data in the source logical volume is not synchronized to the target logical volume.

[0040] In one embodiment, performing remote replication based on the exception recovery duration of the target logical volume includes:

[0041] In the case where the exception recovery duration is greater than or equal to the duration threshold, split the remote replication relationship, and establish a new remote replication relationship according to the target logical volume reselected from each logical volume;

[0042] Based on the new remote replication relationship, synchronize all the data in the source logical volume to the target logical volume in full.

[0043] In the technical solution of the embodiment of the present application, when the abnormal recovery duration of the target logical volume is greater than or equal to the duration threshold, the remote replication relationship is split, a new remote replication relationship is established based on the target logical volume reselected from each logical volume, and based on the new remote replication relationship, the data in the source logical volume is fully synchronized to the target logical volume; when the abnormal recovery duration of the target logical volume times out, the above method establishes a new remote replication relationship and further performs remote replication based on the new remote replication relationship, which can improve the success rate of remote replication and ensure that the data in the source logical volume of the master cluster can be successfully replicated to the slave cluster.

[0044] In one embodiment, performing remote replication based on the abnormal recovery duration of the target logical volume includes:

[0045] When the abnormal type of the target logical volume is insufficient storage pool capacity and the abnormal recovery duration is less than the duration threshold, the unsynchronized data in the source logical volume is synchronized to the target logical volume;

[0046] When the synchronization of the unsynchronized data is completed, if the data consistency check between the source logical volume and the target logical volume fails, the differential data between the source logical volume and the target logical volume is synchronized to the target logical volume to complete the remote replication.

[0047] In the technical solution of the embodiment of the present application, when the abnormal type of the target logical volume is insufficient storage pool capacity and the abnormal recovery duration is less than the duration threshold, the unsynchronized data in the source logical volume is synchronized to the target logical volume. When the synchronization of the unsynchronized data is completed, if the data consistency check between the source logical volume and the target logical volume fails, the differential data between the source logical volume and the target logical volume is synchronized to the target logical volume to complete the remote replication; the above method can continue to perform data synchronization to complete remote replication after the abnormal recovery of the target logical volume, and uses the consistency check method after data synchronization to ensure that all the data in the source logical volume is synchronized to the target logical volume, thereby improving the accuracy of the remote replication result and ensuring the consistency of the remote replication data.

[0048] In a second aspect, an embodiment of the present application provides a remote replication device, which includes:

[0049] An acquisition module, configured to acquire the remote replication comprehensive index values of each logical volume in the slave cluster;

[0050] A selection module, configured to select a target logical volume from each logical volume according to the remote replication comprehensive index values of each logical volume;

[0051] A synchronization module, configured to fully synchronize the data in the source logical volume of the master cluster to the target logical volume to complete the remote replication.

[0052] In a third aspect, an embodiment of the present application further provides a computer device, which includes a controller and a memory. The controller includes a processor, and the memory stores a computer program. When the controller executes the computer program, the steps of the method according to any one of the embodiments in the first aspect are implemented.

[0053] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a controller, the steps of the method according to any one of the embodiments in the first aspect are implemented.

[0054] In a fifth aspect, an embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a controller, the steps of the method according to any one of the embodiments in the first aspect are implemented.

[0055] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically describes the embodiments of the present application. Description of the Drawings

[0056] Figure 1 It is an application environment diagram of the remote replication method in an embodiment;

[0057] Figure 2 It is a schematic flowchart of the remote replication method in an embodiment;

[0058] Figure 3 It is a schematic flowchart of the remote replication method in another embodiment;

[0059] Figure 4 It is a schematic flowchart of the remote replication method in another embodiment;

[0060] Figure 5 It is a schematic diagram of the input data and output data of the performance prediction model and the evaluation model in an embodiment;

[0061] Figure 6 It is a schematic flowchart of the remote replication method in another embodiment;

[0062] Figure 7 It is a schematic flowchart of the remote replication method in another embodiment;

[0063] Figure 8 It is a schematic flowchart of the remote replication method in another embodiment;

[0064] Figure 9 It is a schematic flowchart of the remote replication method in another embodiment;

[0065] Figure 10 It is a schematic flowchart of the remote replication method in another embodiment;

[0066] Figure 11 It is a structural block diagram of the remote replication device in one embodiment;

[0067] Figure 12 It is an internal structure diagram of a computer device in one embodiment. Detailed implementation manners

[0068] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0069] In the field of data storage, as the core technology of disaster recovery and backup, remote replication can achieve remote data backup and disaster recovery. Taking the remote replication of data in the source logical volume in the master cluster to the target logical volume in the slave cluster as an example, after the master cluster enables the remote replication function, the master cluster synchronizes all the data in the source logical volume to the target logical volume in the slave cluster in full amount, or synchronizes the incremental data newly added in the source logical volume to the target logical volume in the slave cluster to achieve remote replication, so as to prepare for preventing the data loss in the source logical volume and disaster recovery. However, in the related technology, only the factors that the capacity of the target logical volume is not less than that of the source logical volume, the target logical volume cannot be mapped, and the read and write of the target logical volume are restricted are considered, so that the target logical volume selected in the slave cluster in the related technology is unreasonable, which will lead to low performance of remote replication. Based on this, the embodiments of the present application provide a remote replication method to reasonably select the target logical volume in the slave cluster to improve the performance of remote replication.

[0070] The remote replication method provided by the embodiments of the present application can be applicable to Figure 1 The remote replication system shown in the figure. The remote replication system includes a master cluster and a slave cluster. Among them, the master cluster may include multiple master nodes, and multiple logical volumes may be set on each master node. The source logical volume may be any logical volume in the master cluster; the slave cluster may include multiple slave nodes, and multiple logical volumes may be set on each slave node. The target logical volume may be any logical volume in the slave cluster. Optionally, the attributes of each logical volume in the above master cluster and slave cluster may be the same or different. In practical applications, both the master cluster and the slave cluster include controllers, and the two controllers may be communicatively connected, and the communication method may be Ethernet, Fibre Channel (FC) network or Remote Direct Memory Access (RDMA) network, etc.

[0071] Here, the controller in the master cluster is called the master controller, and the controller in the slave cluster is called the slave controller. Among them, the structures of the master controller and the slave controller are basically similar. Figure 1 The basic structures of the master controller and the slave controller are shown. In the embodiments of the present application, the master controller includes a management module, a logical volume management module, and a logical volume service management module. Among them, the management module includes a remote replication unit, a target volume selection unit, and a snapshot unit. The logical volume management module includes a replication engine unit and a replication link management unit. The logical volume service management module includes a replication data unit and a replication link management unit.

[0072] In practical applications, the management module in the master controller is connected to the management module in the slave controller through a replication management network. The logical volume service management module in the master controller is connected to the logical volume service management module in the slave controller through a replication data network. Optionally, the remote replication unit is used to provide the remote replication function. The target volume selection unit is used to implement the target volume selection function. The snapshot unit is used to implement the data snapshot function. The replication engine unit is used to manage the running start and stop status of the remote replication task and the status transition of the remote replication relationship (that is, to switch the states of the remote replication relationship such as unsynchronized, normal, in-sync, paused, sync interrupted, pending recovery, failed, and split). The replication link management unit in the logical volume management module is used to configure the routing information of the replication management network between the master cluster and the slave cluster (that is, the link information required for data synchronization). The replication data unit is used to read data from the source logical volume of the master cluster. The replication link management unit in the logical volume service management module is used to configure the routing information of the replication data network, execute the replication data network connection, and execute data synchronization.

[0073] Among them, the replication management network is maintained by the management module and is used to establish the connection between the master cluster and the slave cluster, manage the permissions of the master cluster and the slave cluster to have the remote replication function, as well as the link status of the replication management network, the status of the remote replication relationship, and the status of the replication channel. The replication data network is maintained by the replication link management unit in the logical volume service management module and is responsible for the actual transmission of data between the source logical volume and the target logical volume. Correspondingly, the structure of the slave controller is similar to that of the master controller, except that the replication data unit in the master controller is used to write the data received from the source logical volume into the target logical volume. The following embodiments introduce the specific process of the remote replication method with the execution subject of the remote replication method being the master controller in the master cluster.

[0074] As Figure 2 shown, it is a schematic flowchart of the remote replication method provided by the embodiments of the present application. This method can be implemented through the following steps:

[0075] S100. Obtain the remote replication comprehensive index values of each logical volume in the slave cluster.

[0076] Specifically, the master controller of the master cluster can obtain the attribute information of each logical volume in the slave cluster from locations such as local, cloud, and disk, or communicate with the slave controller based on the replication management network to obtain the attribute information of each logical volume in the slave cluster sent by the slave controller. Then, for any logical volume, determine the remote replication comprehensive index value (i.e., evaluation) of the logical volume according to the attribute information of the logical volume.

[0077] In one embodiment, the method for determining the remote replication comprehensive index value of a logical volume according to the attribute information of the logical volume can be to perform processing such as information conversion, information comparison, and logical operation on the attribute information of the logical volume to obtain the remote replication comprehensive index value of the logical volume.

[0078] In another embodiment, the method for determining the remote replication comprehensive index value of a logical volume according to the attribute information of the logical volume can also be to look up the attribute information of the logical volume in the mapping relationship table, and then determine the remote replication comprehensive index value corresponding to the attribute information of the logical volume in the mapping relationship table as the remote replication comprehensive index value of the logical volume.

[0079] Optionally, the above-mentioned attribute information of the logical volume can be information such as the size of the logical volume, write mode, and expandable capacity.

[0080] In practical applications, if there is a list of available logical volumes in the slave cluster, each logical volume in the list of available logical volumes can be determined as each logical volume in the slave cluster. In this case, the selection range of the target logical volume can be narrowed, the accuracy of the target logical volume selection can be improved, the amount of data processed in the target logical volume selection process can be reduced, and the remote replication speed can be further improved; if there is no list of available logical volumes in the slave cluster, each logical volume in the slave cluster can be determined as each logical volume in the slave cluster.

[0081] S200. Select a target logical volume from each logical volume according to the remote replication comprehensive index values of each logical volume.

[0082] In one implementation manner, for any logical volume in the slave cluster, the master controller can perform arithmetic operation processing, comparison processing, or judgment processing on the remote replication comprehensive index value of the logical volume to obtain a processing result. Further, select a target logical volume from each logical volume according to the processing results of each logical volume.

[0083] Optionally, the above arithmetic operations can be implemented by at least one of addition operation, subtraction operation, multiplication operation, division operation, exponential operation, and logarithmic operation, etc.

[0084] In another implementation, the master controller can pre-train an algorithm model, and then sequentially input the remote replication comprehensive index values of each logical volume into the algorithm model. After the algorithm model selects the target logical volume from each logical volume and outputs the target logical volume identifier, the corresponding target logical volume is determined according to the target logical volume identifier.

[0085] It should be noted here that the attribute information of the target logical volume and the source logical volume in the master cluster may be the same or different, and the embodiments of the present application do not make any limitations in this regard.

[0086] S300. Fully synchronize the data in the source logical volume of the master cluster to the target logical volume to complete the remote replication.

[0087] Based on the target logical volume selected in the previous steps, it can be controlled to fully synchronize the data in the source logical volume of the master cluster to the target logical volume, so as to complete the remote replication between the source logical volume in the master cluster and the target logical volume in the slave cluster.

[0088] In the embodiments of the present application, the source logical volume in the master cluster and the target logical volume in the slave cluster can be connected through an authentication method, that is, a replication channel is established between the source logical volume and the target logical volume.

[0089] The technical solutions in the embodiments of the present application obtain the remote replication comprehensive index values of each logical volume in the slave cluster, select the target logical volume from each logical volume according to the remote replication comprehensive index values of each logical volume, and fully synchronize the data in the source logical volume of the master cluster to the target logical volume to complete the remote replication; the above method can reasonably select the target logical volume corresponding to the source logical volume from all the logical volumes in the slave cluster based on the remote replication comprehensive index values of each logical volume in the slave cluster, and on this basis, realize the remote replication between the source logical volume in the master cluster and the target logical volume in the slave cluster, which can improve the performance of remote replication; at the same time, when selecting the target logical volume of the slave cluster, the above method does not require manual participation, thereby reducing the manual participation error, greatly improving the rationality of the selection of the target logical volume and the availability of the selected target logical volume, reducing the abnormal risk of remote replication, and at the same time maximizing the utilization of the resources of the existing logical volumes in the slave cluster.

[0090] The process of obtaining the remote replication comprehensive index values of each logical volume in the slave cluster will be described below. In one embodiment, as Figure 3 shown, the steps in S100 above can be implemented in the following manner:

[0091] S110. Obtain the configuration attributes of each logical volume and the target configuration attributes of the target logical volume.

[0092] Specifically, the master controller can obtain the configuration attributes of each logical volume in the slave cluster, as well as the target configuration attributes of the pre-determined target logical volume, from locations such as the cloud, local, and disk.

[0093] Here, it should be noted that since logical volumes are created based on storage pools, the redundancy ratio of logical volumes is inherited from the storage pools. Storage pools are composed of physical disks, and performance metrics such as the average latency of disks, the number of read and write operations per second, and the aggregated bandwidth directly affect the block bandwidth, logical bandwidth, and synchronization duration of remote replication. Therefore, the storage pool where the logical volume is located can be used as one of the configuration attributes of the logical volume.

[0094] The thin provisioning types of logical volumes include non-thin and thin. Non-thin logical volumes are reserved according to the configured capacity at the time of creation, while thin logical volumes allocate the actual occupied capacity as data is written. Designating a thin logical volume as the target logical volume may cause an exception of storage pool capacity exhaustion during data synchronization, resulting in the interruption of remote replication. Therefore, the thin provisioning type of the logical volume can also be used as one of the configuration attributes of the logical volume.

[0095] After the data deduplication function of the logical volume is enabled, the logical volume can occupy less capacity in the cluster, and the corresponding remote replication synchronization duration is shorter, but the impact on the front-end business performance of the source logical volume is greater. Therefore, the data deduplication function of the logical volume can also be used as one of the configuration attributes of the logical volume.

[0096] The application type of the logical volume corresponds to different CELL sizes. The CELL size is equivalent to the deduplication granularity. An appropriate application type can control the impact of deduplication on the front-end business performance of the source logical volume. Therefore, the application type of the logical volume can also be used as one of the configuration attributes of the logical volume.

[0097] Data compression can reduce the storage space occupied by files in the logical volume. The remote replication synchronization duration corresponding to the logical volume with data compression enabled is shorter. Therefore, the data compression function of the logical volume can also be used as one of the configuration attributes of the logical volume.

[0098] The quality of service policy of the logical volume specifies the bandwidth, the number of read and write operations per second, and the normalized latency within the period when the policy takes effect. If the source logical volume and the target logical volume are set with quality of service policies with overlapping effective periods, the remote replication block bandwidth and logical bandwidth will be significantly reduced during this period. Therefore, the quality of service policy of the logical volume can also be used as one of the configuration attributes of the logical volume.

[0099] The health status of the logical volume includes the normal state, the degraded state, and the faulty state. The health status only shows the change in redundancy under possible media failures. Designating a logical volume in the degraded state as the target logical volume may lead to a reduction in the reliability of remote replication. Therefore, the health status of the logical volume can also be used as one of the configuration attributes of the logical volume.

[0100] In an embodiment of the present application, the configuration attributes of the above logical volume may include the storage pool where the logical volume is located (src_storpool), the thin type of the logical volume (src_type), the enabled status of the data deduplication function of the logical volume (src_deduplic), the application type of the logical volume (src_apptype), the enabled status of the compression function of the logical volume (src_compress), the quality of service policy of the logical volume (src_qos), and the health status of the logical volume (src_health).

[0101] Optionally, the target configuration attributes of the above target logical volume may be the storage pool where the target logical volume is located, the thin type of the target logical volume (dst_type), the enabled status of the data deduplication function of the target logical volume (dst_deduplic), the application type of the target logical volume (dst_apptype), the enabled status of the compression function of the target logical volume (dst_compress), the quality of service policy of the target logical volume (dst_qos), and the health status of the target logical volume (dst_health).

[0102] S120. Process the configuration attributes and target configuration attributes of each logical volume through a pre-constructed logical volume recommendation model to obtain the remote replication comprehensive index value of each logical volume.

[0103] In practical applications, for any logical volume, the main controller can input both the configuration attributes and target configuration attributes of the logical volume into the pre-constructed logical volume recommendation model. After processing the configuration attributes and target configuration attributes of the logical volume, the logical volume recommendation model outputs the remote replication comprehensive index value of the logical volume. Optionally, the above processing can be achieved through processes such as arithmetic operations, information conversion, and information comparison.

[0104] Among them, the above logical volume recommendation model can be obtained by training an initial logical volume recommendation model with a training set. Optionally, the above training set may include the configuration attributes of multiple logical volumes of different types and the configuration attributes of the logical volume to be selected, and the logical volume to be selected may be any one of the multiple logical volumes of different types.

[0105] Specifically, the training set can be input into the initial logical volume recommendation model to obtain a predicted logical volume selected from multiple logical volumes of different types. Calculate the prediction error value between the predicted logical volume and the standard logical volume through a loss function, and update the initial network parameters in the initial logical volume recommendation model according to the prediction error value. Continuously iterate the above training steps until the prediction error value meets the preset error threshold or the number of iterations reaches the preset iteration number threshold, to obtain the logical volume recommendation model. The above standard logical volume can be an idealized logical volume to be selected.

[0106] It should be noted here that the above comprehensive index of remote replication of logical volumes refers to the comprehensive index for measuring the remote replication performance of logical volumes.

[0107] In the technical solution of the embodiment of the present application, the configuration attributes of each logical volume and the target configuration attributes of the target logical volume are obtained, and through a pre-constructed logical volume recommendation model, the configuration attributes of each logical volume and the target configuration attributes are processed to obtain the comprehensive index values of remote replication of each logical volume; the above method can obtain the comprehensive index values of remote replication of each logical volume in the slave cluster through the logical volume recommendation model, which can improve the acquisition speed of the comprehensive index values of remote replication. Moreover, this process does not require manual participation, can also reduce the error caused by manual participation, can improve the accuracy of the obtained comprehensive index values of remote replication, and can prepare for further selection of reasonable target logical volumes.

[0108] In one embodiment, the above logical volume recommendation model includes a performance prediction model and an evaluation model; as Figure 4 shown, the step of obtaining the comprehensive index values of remote replication of each logical volume by processing the configuration attributes of each logical volume and the target configuration attributes through a pre-constructed logical volume recommendation model in S120 can be implemented in the following manner:

[0109] S210. For any logical volume, input the configuration attributes and target configuration attributes of the logical volume into the performance prediction model, and output the performance index vector of the logical volume. Among them, the performance index vector includes the values of multiple remote replication performance indicators.

[0110] In the embodiment of the present application, the logical volume recommendation model includes a performance prediction model (i.e., an SVR prediction model) and an evaluation model (i.e., a Lagrangian evaluation model).

[0111] In practical applications, for any logical volume in the slave cluster, the configuration attributes and target configuration attributes of the logical volume can be input into the performance prediction model in the logical volume recommendation model, and the performance prediction model outputs the performance index vector of the logical volume.

[0112] Optionally, the performance index vector of the logical volume may include a vector composed of 6-dimensional remote replication performance indicators of the logical volume. The 6-dimensional remote replication performance indicators here can be the block bandwidth (b_bandwith), logical bandwidth (l_bandwith), business impact degree (biz_decline), the time when the data in the source logical volume lags behind the target logical volume, that is, the data lag time (rpo), the duration of synchronizing the data in the source logical volume (sync_duration), and reliability.

[0113] Among them, the above-mentioned block bandwidth refers to the amount of data read and written per second on average after data transmission compression in remote replication; the logical bandwidth refers to the amount of data read and written per second on average before data transmission compression in remote replication; the service impact degree refers to the impact degree on the front-end service performance of the source logical volume; the time when the data in the source logical volume lags behind the target logical volume refers to the time when the data in the source logical volume lags behind the target logical volume in remote replication, that is, RPO; the duration of synchronizing the data in the source logical volume refers to the longest synchronization time within the adjacent two sampling intervals; the reliability refers to the number of interruptions in data synchronization within the adjacent two sampling intervals.

[0114] In practical applications, the process of determining the penalty factor and kernel function width of the performance prediction model can be as follows: Use the supervised learning linear discriminant dimensionality reduction algorithm to extract features from the training set to eliminate the interference of redundant features and linearly correlated components, and then use the particle swarm algorithm to optimize the parameters of the feature extraction results to determine the penalty factor and kernel function width of the performance prediction model.

[0115] S220. Input the performance index vector into the evaluation model, and output the comprehensive remote replication index value of the logical volume.

[0116] Furthermore, the main controller can input the performance index vector of the logical volume output by the performance prediction model into the evaluation model, and the evaluation model outputs the comprehensive remote replication index value of the logical volume. In the embodiment of the present application, the output end of the performance prediction model is connected to the input end of the evaluation model.

[0117] Such as Figure 5 shown is a schematic diagram of the input data and output data of the performance prediction model and the evaluation model; among them, the performance prediction model can obtain a regression equation , represents a 14-dimensional vector corresponding to the configuration attributes of the logical volume and the target configuration attributes of the target logical volume, represents a 6-dimensional vector of remote replication performance indicators (i.e., the performance index vector), and are the regression equation parameters; the evaluation model quantifies and weights the performance index vector and then outputs the comprehensive remote replication index value.

[0118] At the same time, the Lagrangian function in the evaluation model is , is the Lagrange multiplier of the inequality constraint . Specifically, the minimum time when the data in the source logical volume lags behind the target logical volume can be used as the objective function, and inequality constraints are imposed on the block bandwidth, logical bandwidth, service impact degree, duration of synchronizing the data in the source logical volume, and reliability (i.e., the inequality constraint ), solve the non - linear programming to obtain the Lagrange multipliers corresponding to the inequality constraints , substitute and the performance index vector into the Lagrangian function to obtain the comprehensive remote replication index value.

[0119] In the technical solution of the embodiment of the present application, for any logical volume, input the configuration attributes and target configuration attributes of the logical volume into the performance prediction model, output the values of multiple remote replication performance indexes of the logical volume, and then input the performance index vector into the evaluation model to output the comprehensive remote replication index value of the logical volume; the above - mentioned method can predict the values of multi - dimensional remote replication performance indexes of the logical volume in the slave cluster, and then convert the values of multi - dimensional remote replication performance indexes of the logical volume into the one - dimensional comprehensive remote replication index value of the logical volume, so as to provide a reference basis for further determining whether the logical volume can be used as the target logical volume to be recommended; at the same time, the above - mentioned method can be realized through the performance prediction model and the evaluation model, which can improve the acquisition speed of the comprehensive remote replication index value, and this process does not require manual participation, can also reduce the manual participation error, and can also improve the accuracy of the obtained comprehensive remote replication index value.

[0120] The process of selecting the target logical volume from each logical volume according to the comprehensive remote replication index value of each logical volume is described below. In one embodiment, the steps in S200 may include:

[0121] Determine the logical volumes whose comprehensive remote replication index values exceed the preset threshold in each logical volume as the target logical volumes; or, if the comprehensive remote replication index values of all logical volumes are less than the preset threshold, create a logical volume corresponding to the source logical volume and determine it as the target logical volume.

[0122] In the embodiment of the present application, for any logical volume, the master controller can determine whether the comprehensive remote replication index value of the logical volume exceeds the preset threshold. If so, determine the logical volume as the target logical volume.

[0123] Among them, if there are multiple logical volumes whose comprehensive remote replication index values exceed the preset threshold, the logical volume corresponding to the largest comprehensive remote replication index value among these multiple logical volumes can be determined as the target logical volume. Optionally, if there is one logical volume whose comprehensive remote replication index value exceeds the preset threshold, the logical volume can be determined as the target logical volume.

[0124] Optionally, the above - mentioned preset threshold can be user - defined or determined according to historical experience values, and the embodiments of the present application do not make any limitations in this regard.

[0125] In addition, when the comprehensive remote replication index values of all logical volumes are less than a preset threshold, it indicates that none of the logical volumes can be recommended as the target logical volume. In this case, a corresponding logical volume with configuration attributes approximately equal to those of the source logical volume can be created as the target logical volume.

[0126] Meanwhile, when the comprehensive remote replication index values of all logical volumes are less than a preset threshold, the master controller can also output an alarm message to remind the user that there are risks in using each logical volume in the slave cluster as the target logical volume, and it is necessary to create a logical volume corresponding to the source logical volume and determine it as the target logical volume.

[0127] In the technical solution of this embodiment of the present application, the logical volume with a comprehensive remote replication index value exceeding the preset threshold in each logical volume is determined as the target logical volume. If the comprehensive remote replication index values of all logical volumes are less than the preset threshold, a logical volume corresponding to the source logical volume is created and determined as the target logical volume; the above method can flexibly determine the target logical volume through different target logical volume determination strategies, thereby improving the wide applicability of selecting the target logical volume and also improving the success rate of determining the target logical volume.

[0128] In practical applications, it is necessary to first establish a remote replication relationship, and then implement the remote replication from the source logical volume in the master cluster to the target logical volume in the slave cluster according to the remote replication relationship. The process of how to implement remote replication will be described below. In one embodiment, as Figure 6 shown, the step of fully synchronizing the data in the source logical volume of the master cluster to the target logical volume in S300 above can be implemented in the following manner:

[0129] S310. Establish a remote replication relationship according to the source logical volume and the target logical volume in the master cluster.

[0130] Among them, the above remote replication relationship can be understood as the data replication relationship between the source logical volume and the target logical volume. The remote replication relationship may include the name of the source logical volume, the master-slave role of the source logical volume, the name of the target logical volume, the mutual trust device name between the slave cluster and the master cluster, the health status of the source logical volume and the target logical volume, the running status of the source logical volume and the target logical volume, the synchronization progress of the source logical volume and the target logical volume, the replication mode of the source logical volume and the target logical volume, the replication rate of the source logical volume and the target logical volume, the replication strategy of the source logical volume and the target logical volume, the time strategy of the source logical volume and the target logical volume, etc. In the embodiment of the present application, the above remote replication relationship can be understood as the remote replication Pair relationship.

[0131] In the embodiments of the present application, after determining the source logical volume and the target logical volume, the identifiers of the source logical volume and the target logical volume can be determined. Optionally, the identifier of the source logical volume can represent the location of the source logical volume, and the identifier of the target logical volume can represent the location of the target logical volume.

[0132] Furthermore, a remote replication relationship can be established according to the identifiers of the source logical volume and the target logical volume in the master cluster. Additionally, an algorithm model can be pre-trained, and then the identifiers of the source logical volume and the target logical volume are both input into the algorithm model, and the algorithm model outputs the remote replication relationship to complete the establishment of the remote replication relationship.

[0133] S320. Based on the remote replication relationship, synchronize all the data in the source logical volume to the target logical volume.

[0134] Based on the remote replication relationship established in the above steps, in the case where all the data in the source logical volume has not been synchronized to the target logical volume, that is, in the case of the first synchronization, all the data in the source logical volume can be synchronized to the target logical volume.

[0135] In practical applications, after the first synchronization is achieved between the source logical volume and the target logical volume, during each remote replication, only the incremental data newly added in the source logical volume needs to be synchronized to the target logical volume.

[0136] In addition, during the remote replication process, the master controller can also monitor the configuration attributes of the target logical volume and add the configuration attributes of the target logical volume to the training set for optimizing the logical volume recommendation model and improving the accuracy of the output result of the logical volume recommendation model.

[0137] The technical solution in the embodiments of the present application establishes a remote replication relationship according to the source logical volume and the target logical volume in the master cluster, and based on the remote replication relationship, synchronizes all the data in the source logical volume to the target logical volume; the above method can first establish a remote replication relationship and then synchronize the data in the source logical volume to the target logical volume based on the remote replication relationship, which can avoid the problem of incorrect data synchronization location caused by synchronizing the data in the source logical volume to other logical volumes and can improve the accuracy of data synchronization.

[0138] In practical applications, the target logical volume may be abnormal during the process of implementing remote replication. The following describes the processing process after the target logical volume is abnormal during the remote replication process. In one embodiment, during the process of implementing remote replication for the target logical volume, as Figure 7 shown, the above method may further include the following steps:

[0139] S400. Detect the status of the target logical volume.

[0140] Specifically, the master controller can detect the status of the target logical volume in real time or periodically. Optionally, the status of the target logical volume can be abnormal or normal.

[0141] S500. If the status of the target logical volume is abnormal and it is in a data non - transmission state, perform remote replication according to the replication mode of the target logical volume and the abnormal recovery duration.

[0142] It should be noted here that during the remote replication process, the target logical volume does not always perform data synchronization, that is, receive data transmitted from the source logical volume; correspondingly, during the remote replication process, the target logical volume can be in a data transmission state or a data non - transmission state.

[0143] In practical applications, if the status of the target logical volume is abnormal and the target logical volume is in a data non - transmission state, remote replication can be performed according to the replication mode of the target logical volume and the abnormal recovery duration.

[0144] Optionally, the replication mode of the target logical volume refers to the current replication mode of the target logical volume. This replication mode can be a synchronous replication mode or an asynchronous replication mode. In the embodiments of the present application, before data synchronization is achieved, the master controller can control to enable the remote replication function of the source logical volume and the target logical volume, and set the replication modes of the source logical volume and the target logical volume.

[0145] Among them, the above - mentioned synchronous replication mode means that the host writes data to the source logical volume in the master - side cluster, and the master - side cluster synchronizes the data in the source logical volume to the target logical volume in the slave - side cluster, and then the master - side cluster returns a successful write request to the host; the above - mentioned asynchronous replication mode means that after the host writes data to the source logical volume in the master - side cluster, the master - side cluster does not need to wait for the process of synchronizing the data in the source logical volume to the target logical volume in the slave - side cluster, and the master - side cluster directly returns a successful write request to the host.

[0146] In addition, in scenarios such as the failure, shutdown for maintenance, or disaster recovery drill of the master - side cluster where the source logical volume in the master - side cluster is located, the established remote replication relationship can be split, and the slave - side cluster where the target logical volume is located takes over the corresponding services.

[0147] In the embodiments of the present application, the abnormal recovery duration of the above - mentioned target logical volume can be understood as the interval duration from the moment when the abnormality of the target logical volume is detected until the moment when the target logical volume is detected to be normal.

[0148] For example, the method of performing remote replication according to the replication mode of the target logical volume and the abnormal recovery duration can be that when the replication mode of the target logical volume is the synchronous replication mode and the abnormal recovery duration is less than or equal to a preset duration threshold, all or differential data in the source logical volume is synchronized to the target logical volume.

[0149] Optionally, the above duration threshold can be user-defined or determined according to historical experience values, and the embodiments of the present application do not limit this.

[0150] S600. If the status of the target logical volume is abnormal and it is in the data transmission state, record the remote replication failure event and cancel the remote replication process of the target logical volume, so that the target logical volume is in the data non-transmission state.

[0151] Among them, when the status of the target logical volume is abnormal and the target logical volume is in the data transmission state, the master controller can record the remote replication failure event between the source logical volume and the target logical volume in the system log, cancel the remote replication process of the target logical volume, so that the target logical volume is in the data non-transmission state, and continue to execute the steps in S500 above.

[0152] In the technical solution of the embodiments of the present application, the status of the target logical volume is detected. If the status of the target logical volume is abnormal and it is in the data non-transmission state, remote replication is performed according to the replication mode and abnormal recovery duration of the target logical volume. If the status of the target logical volume is abnormal and it is in the data transmission state, record the remote replication failure event and cancel the remote replication process of the target logical volume, so that the target logical volume is in the data non-transmission state; when the target logical volume appears abnormal during the remote replication process, the above method can adopt different exception handling methods according to different transmission states of the target logical volume to achieve remote replication, thereby improving the success rate of remote replication; at the same time, the above method does not require complex algorithms to participate, thereby reducing the amount of data processed during the remote replication process and improving the speed and efficiency of remote replication.

[0153] The following describes the process of performing remote replication according to the replication mode and abnormal recovery duration of the target logical volume. In one embodiment, the step of performing remote replication according to the replication mode and abnormal recovery duration of the target logical volume in S500 above can be implemented in the following manner:

[0154] If the replication mode is the asynchronous replication mode, perform remote replication based on the abnormal recovery duration of the target logical volume; or, if the replication mode is the synchronous replication mode, change the synchronous replication mode so that the target logical volume is in the asynchronous replication mode.

[0155] In the embodiments of the present application, when the replication mode of the target logical volume is the abnormal replication mode, it can be determined whether the abnormal recovery duration of the target logical volume is less than the preset duration threshold. If so, synchronize all or differential data in the source logical volume to the target logical volume.

[0156] In addition, when the replication mode of the target logical volume is the synchronous replication mode, the synchronous replication mode of the source logical volume in the master cluster and the target logical volume in the slave cluster can be changed so that the source logical volume and the target logical volume are in the asynchronous replication mode, and then remote replication is performed based on the exception recovery duration of the target logical volume.

[0157] In the technical solution of the embodiment of the present application, if the replication mode is the asynchronous replication mode, remote replication is performed based on the exception recovery duration of the target logical volume. If the replication mode is the synchronous replication mode, the synchronous replication mode is changed so that the target logical volume is in the asynchronous replication mode; the above method can adopt corresponding exception handling methods to implement remote replication according to different replication modes of the target logical volume, thereby improving the success rate of remote replication of the target logical volume.

[0158] In one embodiment, the step of performing remote replication based on the exception recovery duration of the target logical volume, as Figure 8 shown, can be implemented in the following manner:

[0159] S510. When the exception type of the target logical volume is an offline failure and the exception recovery duration is less than the duration threshold, perform consistency verification on the data in the source logical volume and the target logical volume.

[0160] Specifically, the exception type of the target logical volume can be detected. When it is determined that the exception type of the target logical volume is an offline failure and the exception recovery duration of the target logical volume is less than the duration threshold, consistency verification can be performed on the data in the source logical volume and the target logical volume.

[0161] Among them, the method of performing consistency verification on the data in the source logical volume and the target logical volume can be to search for all the data in the source logical volume among all the data of the target logical volume. When all the data of the target logical volume contains all the data of the source logical volume, it is determined that the consistency verification of the data in the source logical volume and the target logical volume passes; otherwise, it is determined that the consistency verification of the data in the source logical volume and the target logical volume fails.

[0162] In addition, the method of performing consistency verification on the data in the source logical volume and the target logical volume can also be to compare each data in the source logical volume with all the data in the target logical volume one by one. If there is corresponding data in all the data of the target logical volume that is equal to each data in the source logical volume respectively, it is determined that the consistency verification of the data in the source logical volume and the target logical volume passes. If only some of the data in all the data of the target logical volume is equal to some of the data in the source logical volume respectively, it is determined that the consistency verification of the data in the source logical volume and the target logical volume fails.

[0163] S520. If the consistency check fails, based on the remote replication relationship, synchronize all the data in the source logical volume to the target logical volume.

[0164] Among them, when the data consistency check between the source logical volume and the target logical volume fails, the remote replication relationship established in the previous steps can be restored. Based on the remote replication relationship established previously, synchronize all the data in the source logical volume to the target logical volume according to the original data synchronization method.

[0165] In the technical solution of this application embodiment, when the abnormal type of the target logical volume is an offline fault and the abnormal recovery duration is less than the duration threshold, perform a consistency check on the data in the source logical volume and the target logical volume. When the consistency check fails, based on the remote replication relationship, synchronize all the data in the source logical volume to the target logical volume; the above method can ensure that after the abnormal recovery of the target logical volume, accurately synchronize all the data in the source logical volume to the target logical volume through the consistency check method, ensure the consistency of the remote replication data after restoring the remote replication relationship, and avoid the problem that some data in the source logical volume is not synchronized to the target logical volume.

[0166] In another embodiment, as Figure 9 shown, the above steps of performing remote replication based on the abnormal recovery duration of the target logical volume can be implemented in the following way:

[0167] S530. When the abnormal recovery duration is greater than or equal to the duration threshold, split the remote replication relationship and establish a new remote replication relationship according to the target logical volume reselected from each logical volume.

[0168] In practical applications, the main controller can determine whether the abnormal recovery duration of the target logical volume is greater than or equal to the duration threshold. When it is determined that the abnormal recovery duration is greater than or equal to the duration threshold, split the remote replication relationship established in the previous steps, and then the steps in the above S100 - S200 can be re - executed to reselect the target logical volume from each logical volume and establish a new remote replication relationship according to the source logical volume and the reselected target logical volume.

[0169] It should be noted here that splitting the remote replication relationship means disconnecting the Pair relationship between the source logical volume and the target logical volume, and no longer synchronizing the data in the source logical volume to the target logical volume, that is, no longer performing remote replication. At this time, a new target logical volume needs to be reselected and a new remote replication relationship needs to be established to complete the remote replication.

[0170] Optionally, the reselected target logical volume and the previously selected target logical volume can be the same logical volume or different logical volumes, and this application embodiment does not make any limitations in this regard.

[0171] S540. Based on the new remote replication relationship, synchronize all the data in the source logical volume to the target logical volume in full volume.

[0172] Based on the new remote replication relationship re-established in the previous steps, all the data in the source logical volume can be synchronously replicated in full volume to the reselected target logical volume.

[0173] In the technical solution of the embodiment of the present application, when the abnormal recovery duration of the target logical volume is greater than or equal to the duration threshold, split the remote replication relationship, establish a new remote replication relationship according to the reselected target logical volume from each logical volume, and based on the new remote replication relationship, synchronize all the data in the source logical volume to the target logical volume; when the abnormal recovery duration of the target logical volume times out, establish a new remote replication relationship and further perform remote replication based on the new remote replication relationship, which can improve the success rate of remote replication and ensure that the data in the source logical volume of the master cluster can be successfully replicated to the slave cluster.

[0174] In another embodiment, as Figure 10 shown, the above steps of remote replication based on the abnormal recovery duration of the target logical volume can be implemented in the following manner:

[0175] S550. When the abnormal type of the target logical volume is insufficient storage pool capacity and the abnormal recovery duration is less than the duration threshold, synchronize the unsynchronized data in the source logical volume to the target logical volume.

[0176] In practical applications, the master controller can detect the abnormal type of the target logical volume. When it is determined that the abnormal type of the target logical volume is insufficient storage pool capacity and the abnormal recovery duration of the target logical volume is less than the duration threshold, endpoint resumption of data in the source logical volume can be performed to synchronize the unsynchronized data in the source logical volume to the target logical volume.

[0177] In the embodiment of the present application, when the target logical volume is of the thin type, the problem of insufficient capacity of the storage pool where the target logical volume is located will occur.

[0178] S560. When the synchronization of the unsynchronized data is completed, if the data consistency check between the source logical volume and the target logical volume fails, synchronize the differential data between the source logical volume and the target logical volume to the target logical volume to complete the remote replication.

[0179] Among them, when it is determined to synchronize the unsynchronized data in the source logical volume to the target logical volume, a consistency check can be performed on the data in the source logical volume and the target logical volume. When the consistency check fails, obtain the differential data between the source logical volume and the target logical volume, and continue to synchronize the differential data in the source logical volume to the target logical volume to complete the remote replication.

[0180] In the technical solution of the embodiment of the present application, when the abnormal type of the target logical volume is insufficient storage pool capacity and the abnormal recovery duration is less than the duration threshold, the unsynchronized data in the source logical volume is synchronized to the target logical volume. After the unsynchronized data is synchronized, if the data consistency check between the source logical volume and the target logical volume fails, the differential data between the source logical volume and the target logical volume is synchronized to the target logical volume to complete remote replication. The above method can continue to perform data synchronization to complete remote replication after the abnormal recovery of the target logical volume, and use the consistency check method after data synchronization to ensure that all the data in the source logical volume is synchronized to the target logical volume, thereby improving the accuracy of the remote replication result and ensuring the consistency of the remote replication data.

[0181] In one embodiment, the embodiment of the present application further provides a remote replication method, and the method includes the following processes:

[0182] (1) Obtain the configuration attributes of each logical volume in the slave cluster and the target configuration attributes of the target logical volume;

[0183] (2) For any logical volume, input the configuration attributes of the logical volume and the target configuration attributes into the performance prediction model in the logical volume recommendation model, and output the performance index vector of the logical volume. The performance index vector includes the values of multiple remote replication performance indexes;

[0184] (3) Input the performance index vector into the evaluation model, and output the remote replication comprehensive index value of the logical volume;

[0185] (4) Determine the logical volumes whose remote replication comprehensive index values in each logical volume exceed the preset threshold as the target logical volumes; or, if the remote replication comprehensive index values of each logical volume are all less than the preset threshold, create a logical volume corresponding to the source logical volume and determine it as the target logical volume;

[0186] (5) Establish a remote replication relationship according to the source logical volume and the target logical volume in the master cluster;

[0187] (6) Based on the remote replication relationship, synchronize all the data in the source logical volume to the target logical volume;

[0188] Among them, in the process of executing the above step (6), the following steps are further included:

[0189] (61) Detect the status of the target logical volume;

[0190] If the status of the target logical volume is abnormal and it is in a data non - transfer state, then if the replication mode is asynchronous replication mode, remote replication is performed based on the abnormal recovery duration of the target logical volume; or, if the replication mode is synchronous replication mode, the synchronous replication mode is changed so that the target logical volume is in asynchronous replication mode.

[0191] The above step (62) can be implemented in the following three ways:

[0192] The first way includes:

[0193] (621) When the abnormal type of the target logical volume is an offline fault and the abnormal recovery duration is less than the duration threshold, data consistency verification is performed on the data in the source logical volume and the target logical volume.

[0194] (622) If the consistency verification fails, based on the remote replication relationship, all the data in the source logical volume is synchronously replicated to the target logical volume.

[0195] The second way includes:

[0196] (623) When the abnormal recovery duration is greater than or equal to the duration threshold, the remote replication relationship is split, and a new remote replication relationship is established based on the target logical volume re - selected from each logical volume.

[0197] (624) Based on the new remote replication relationship, all the data in the source logical volume is synchronously replicated to the target logical volume.

[0198] The third way includes:

[0199] (625) When the abnormal type of the target logical volume is insufficient storage pool capacity and the abnormal recovery duration is less than the duration threshold, the unsynchronized data in the source logical volume is synchronously replicated to the target logical volume.

[0200] (626) When the synchronization of the unsynchronized data is completed, if the data consistency verification between the source logical volume and the target logical volume fails, the differential data between the source logical volume and the target logical volume is synchronously replicated to the target logical volume to complete the remote replication.

[0201] (7) If the status of the target logical volume is abnormal and it is in a data transfer state, a remote replication failure event is recorded, and the remote replication process of the target logical volume is cancelled so that the target logical volume is in a data non - transfer state.

[0202] The execution processes of the above (1) to (7) can specifically refer to the descriptions of the above embodiments, and their implementation principles and technical effects are similar, so they will not be elaborated here.

[0203] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless specifically stated herein, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0204] Based on the same inventive concept, an embodiment of the present application also provides a remote replication device for implementing the remote replication method described above. The solution provided by this device for solving problems is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the remote replication device provided below can refer to the limitations on the remote replication method in the above text, and will not be repeated here.

[0205] In one embodiment, Figure 11 is a schematic structural diagram of a remote replication device in an embodiment of the present application. The remote replication device provided in the embodiment of the present application can be applied to the master controller in the master-end cluster. As Figure 11 shown, the remote replication device in the embodiment of the present application may include: an acquisition module 11, a selection module 12, and a synchronization module 13, where:

[0206] The acquisition module 11 is configured to acquire the remote replication comprehensive index values of each logical volume in the slave-end cluster;

[0207] The selection module 12 is configured to select a target logical volume from each logical volume according to the remote replication comprehensive index values of each logical volume;

[0208] The synchronization module 13 is configured to synchronize all the data in the source logical volume of the master-end cluster to the target logical volume to complete remote replication.

[0209] The remote replication device provided in the embodiment of the present application can be used to execute the technical solutions in the above embodiments of the remote replication method of the present application. The implementation principle and technical effects are similar, and will not be repeated here.

[0210] In one of the embodiments, the acquisition module 11 includes: a configuration attribute acquisition unit and a processing unit, where:

[0211] The configuration attribute acquisition unit is configured to acquire the configuration attributes of each logical volume and the target configuration attributes of the target logical volume;

[0212] A processing unit, configured to process the configuration attributes and target configuration attributes of each logical volume through a pre-constructed logical volume recommendation model, so as to obtain the remote replication comprehensive index value of each logical volume.

[0213] The remote replication device provided by the embodiments of the present application can be used to execute the technical solutions in the above-mentioned remote replication method embodiments of the present application. The implementation principles and technical effects are similar, and will not be elaborated here.

[0214] In one embodiment, the logical volume recommendation model includes a performance prediction model and an evaluation model; the processing unit includes: a first processing subunit and a second processing subunit, where:

[0215] The first processing subunit is configured to input the configuration attributes and target configuration attributes of any logical volume into the performance prediction model, and output a performance index vector of the logical volume; the performance index vector includes the values of multiple remote replication performance indexes;

[0216] The second processing subunit is configured to input the performance index vector into the evaluation model, and output the remote replication comprehensive index value of the logical volume.

[0217] The remote replication device provided by the embodiments of the present application can be used to execute the technical solutions in the above-mentioned remote replication method embodiments of the present application. The implementation principles and technical effects are similar, and will not be elaborated here.

[0218] In one embodiment, the selection module 12 is specifically configured to:

[0219] Determine the logical volumes whose remote replication comprehensive index values exceed a preset threshold among all logical volumes; or,

[0220] If the remote replication comprehensive index values of all logical volumes are less than the preset threshold, create a logical volume corresponding to the source logical volume and determine it as the target logical volume.

[0221] The remote replication device provided by the embodiments of the present application can be used to execute the technical solutions in the above-mentioned remote replication method embodiments of the present application. The implementation principles and technical effects are similar, and will not be elaborated here.

[0222] In one embodiment, the synchronization module 13 is specifically configured to:

[0223] Establish a remote replication relationship according to the source logical volume and the target logical volume in the master cluster;

[0224] Based on the remote replication relationship, synchronize all the data in the source logical volume to the target logical volume in full volume.

[0225] The remote replication device provided by the embodiment of the present application can be used to execute the technical solutions in the above-mentioned embodiment of the remote replication method of the present application. The implementation principle and technical effect are similar, and will not be elaborated here.

[0226] In one embodiment, the above-mentioned remote replication device further includes: a status detection module, an execution module, and a processing module, where:

[0227] The status detection module is used to detect the status of the target logical volume;

[0228] The execution module is used to perform remote replication according to the replication mode of the target logical volume and the exception recovery duration when the status of the target logical volume is abnormal and in the data untransmitted state;

[0229] The processing module is used to record the remote replication failure event and cancel the remote replication process of the target logical volume when the status of the target logical volume is abnormal and in the data transmission state, so that the target logical volume is in the data untransmitted state.

[0230] The remote replication device provided by the embodiment of the present application can be used to execute the technical solutions in the above-mentioned embodiment of the remote replication method of the present application. The implementation principle and technical effect are similar, and will not be elaborated here.

[0231] In one embodiment, the execution module includes: an execution unit and a mode change unit, where:

[0232] The execution unit is used to perform remote replication based on the exception recovery duration of the target logical volume when the replication mode is the asynchronous replication mode;

[0233] The mode change unit is used to change the synchronous replication mode when the replication mode is the synchronous replication mode, so that the target logical volume is in the asynchronous replication mode.

[0234] The remote replication device provided by the embodiment of the present application can be used to execute the technical solutions in the above-mentioned embodiment of the remote replication method of the present application. The implementation principle and technical effect are similar, and will not be elaborated here.

[0235] In one embodiment, the execution unit includes: a first execution subunit, where the first execution subunit is specifically used for:

[0236] In the case where the exception type of the target logical volume is an offline failure and the exception recovery duration is less than the duration threshold, perform consistency verification on the data in the source logical volume and the target logical volume;

[0237] If the consistency verification fails, synchronize all the data in the source logical volume to the target logical volume based on the remote replication relationship.

[0238] The remote replication device provided by the embodiment of the present application can be used to execute the technical solutions in the above-mentioned remote replication method embodiments of the present application. The implementation principles and technical effects are similar and will not be elaborated here.

[0239] In one embodiment, the execution unit includes: a second execution subunit, where the second execution subunit is specifically configured to:

[0240] In the case where the abnormal recovery duration is greater than or equal to the duration threshold, split the remote replication relationship, and establish a new remote replication relationship based on the target logical volumes reselected from each logical volume;

[0241] Based on the new remote replication relationship, synchronize all the data in the source logical volume to the target logical volume.

[0242] The remote replication device provided by the embodiment of the present application can be used to execute the technical solutions in the above-mentioned remote replication method embodiments of the present application. The implementation principles and technical effects are similar and will not be elaborated here.

[0243] In one embodiment, the execution unit includes: a third execution subunit, where the third execution subunit is specifically configured to:

[0244] In the case where the abnormal type of the target logical volume is insufficient storage pool capacity and the abnormal recovery duration is less than the duration threshold, synchronize the unsynchronized data in the source logical volume to the target logical volume;

[0245] In the case where the synchronization of the unsynchronized data is completed, if the data consistency check between the source logical volume and the target logical volume fails, synchronize the differential data between the source logical volume and the target logical volume to the target logical volume to complete the remote replication.

[0246] The remote replication device provided by the embodiment of the present application can be used to execute the technical solutions in the above-mentioned remote replication method embodiments of the present application. The implementation principles and technical effects are similar and will not be elaborated here.

[0247] For the specific limitations of the remote replication device, reference can be made to the limitations on the remote replication method in the foregoing text, which will not be elaborated here. Each module in the above-mentioned remote replication device can be implemented in whole or in part through software, hardware, and their combinations. The above-mentioned modules can be embedded in the controller in the computer device in hardware form or independent of it, or stored in the memory in the computer device in software form, so that the controller can call and execute the operations corresponding to the above-mentioned modules.

[0248] In one embodiment, a computer device is further provided. The computer device can be a server, and its internal structure diagram can be as Figure 12As shown. The computer device includes a controller, a memory, and an input / output interface connected via a system bus, as well as a communication interface, an input device, and a display unit connected via the input / output interface, and a disk array connected via the communication interface; the above-mentioned controller includes a processor. Among them, the controller of the computer device is used to provide processing capabilities, and the disk array is used to create logical volumes. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the remote replication comprehensive index values of each logical volume in the slave cluster. The network interface of the computer device is used to communicate with an external endpoint via a network connection. When the computer program is executed by the controller, it implements a remote replication method.

[0249] Those skilled in the art can understand that Figure 12 the structure shown in is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component layout.

[0250] In one embodiment, a computer device is further provided, including a controller and a memory. The controller includes a processor, and a computer program is stored in the memory. When the controller executes the computer program, it implements the technical solution in the above-mentioned remote replication method embodiment of this application. The implementation principle and technical effect are similar and will not be elaborated here.

[0251] In one embodiment, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by the controller, it implements the technical solution of the above-mentioned remote replication method of this application. The implementation principle and technical effect are similar and will not be elaborated here.

[0252] In one embodiment, a computer program product is further provided, including a computer program. When the computer program is executed by the controller, it implements the technical solution of the above-mentioned remote replication method of this application. The implementation principle and technical effect are similar and will not be elaborated here.

[0253] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above various methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0254] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0255] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A remote replication method, characterized in that, The method includes: Obtaining the remote replication comprehensive metric values of each logical volume in the slave cluster; Selecting target logical volumes from each of the logical volumes according to the remote replication comprehensive metric values of each logical volume; Fully synchronizing the data in the source logical volume of the master cluster to the target logical volume to complete remote replication.

2. The method according to claim 1, characterized in that, The obtaining the remote replication comprehensive metric values of each logical volume in the slave cluster includes: Obtaining the configuration attributes of each logical volume and the target configuration attributes of the target logical volume; Processing the configuration attributes of each logical volume and the target configuration attributes through a pre-constructed logical volume recommendation model to obtain the remote replication comprehensive metric values of each logical volume.

3. The method according to claim 2, characterized in that, The logical volume recommendation model includes a performance prediction model and an evaluation model; the processing the configuration attributes of each logical volume and the target configuration attributes through a pre-constructed logical volume recommendation model to obtain the remote replication comprehensive metric values of each logical volume includes: For any logical volume, inputting the configuration attributes of the logical volume and the target configuration attributes into the performance prediction model to output the performance metric vector of the logical volume; the performance metric vector includes the values of multiple remote replication performance metrics; Inputting the performance metric vector into the evaluation model to output the remote replication comprehensive metric value of the logical volume.

4. The method according to any one of claims 1-3, characterized in that, The selecting target logical volumes from each of the logical volumes according to the remote replication comprehensive metric values of each logical volume includes: Determining the logical volumes whose remote replication comprehensive metric values exceed a preset threshold among each of the logical volumes as the target logical volumes; or, If the remote replication comprehensive metric values of each of the logical volumes are all less than the preset threshold, creating a logical volume corresponding to the source logical volume and determining it as the target logical volume.

5. The method according to any one of claims 1-3, characterized in that, The fully synchronizing the data in the source logical volume of the master cluster to the target logical volume includes: Establishing a remote replication relationship based on the source logical volume and the target logical volume in the master cluster; Based on the remote replication relationship, fully synchronizing the data in the source logical volume to the target logical volume.

6. The method according to any one of claims 1-3, characterized in that, The method further includes: Detecting the status of the target logical volume; If the status of the target logical volume is abnormal and in a data not transmitted state, performing the remote replication according to the replication mode and the abnormal recovery duration of the target logical volume; If the status of the target logical volume is abnormal and in a data transmission state, recording a remote replication failure event and canceling the remote replication process of the target logical volume to make the target logical volume in a data not transmitted state.

7. The method according to claim 6, characterized in that, The performing the remote replication according to the replication mode and the abnormal recovery duration of the target logical volume includes: If the replication mode is an asynchronous replication mode, performing the remote replication based on the abnormal recovery duration of the target logical volume; If the replication mode is a synchronous replication mode, changing the synchronous replication mode to make the target logical volume in an asynchronous replication mode.

8. The method according to claim 7, characterized in that, The performing the remote replication based on the abnormal recovery duration of the target logical volume includes: When the exception type of the target logical volume is an offline fault and the exception recovery duration is less than the duration threshold, perform consistency check on the data in the source logical volume and the target logical volume; If the consistency check fails, based on the remote replication relationship, fully synchronize the data in the source logical volume to the target logical volume.

9. The method according to claim 7, characterized in that, The performing the remote replication based on the exception recovery duration of the target logical volume includes: When the exception recovery duration is greater than or equal to the duration threshold, split the remote replication relationship, and establish a new remote replication relationship according to the target logical volume reselected from each of the logical volumes; Based on the new remote replication relationship, fully synchronize the data in the source logical volume to the target logical volume.

10. The method according to claim 7, characterized in that, The performing the remote replication based on the exception recovery duration of the target logical volume includes: When the exception type of the target logical volume is insufficient storage pool capacity and the exception recovery duration is less than the duration threshold, synchronize the unsynchronized data in the source logical volume to the target logical volume; When the synchronization of the unsynchronized data is completed, if the consistency check of the data in the source logical volume and the target logical volume fails, synchronize the differential data between the source logical volume and the target logical volume to the target logical volume to complete the remote replication.

11. A remote replication device, characterized in that, The device includes: An acquisition module, configured to acquire the remote replication comprehensive index values of each logical volume in the slave cluster; A selection module, configured to select a target logical volume from each of the logical volumes according to the remote replication comprehensive index values of each of the logical volumes; A synchronization module, configured to fully synchronize the data in the source logical volume of the master cluster to the target logical volume to complete the remote replication.

12. A computer device, comprising a controller and a memory, the controller comprising a processor, the memory storing a computer program, characterized in that, When the controller executes the computer program, the steps of the method according to any one of claims 1-10 are implemented.

13. A computer-readable storage medium, having a computer program stored thereon, characterized in that, When the computer program is executed by the controller, the steps of the method according to any one of claims 1-10 are implemented.

14. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the controller, the steps of the method according to any one of claims 1-10 are implemented.

Images