Disaster recovery method, device and system

Abstract

A disaster recovery method, device and system can be applied to the technical field of storage. The method comprises the steps that a first data center synchronously copies backup data to a second data center through a first synchronous copying link; under the condition that a third data center is normal, the first data center asynchronously copies the backup data to the third data center through a first asynchronous copying link; under the condition that the third data center breaks down, the first data center asynchronously copies the backup data to a fourth data center through a second asynchronous copying link; wherein the first data center and the second data center are both production centers, the first data center and the second data center are located in a first area, the third data center and the fourth data center are both disaster recovery centers, and the third data center and the fourth data center are located in a second area. The disaster recovery performance of the disaster recovery system can be reduced in a stepped manner, so that user demands can be met to the greatest extent.

Description

Technical Field

[0001] The present application relates to the field of storage technology, and in particular to a disaster recovery method, device and system. Background Art

[0002] The disaster recovery system can provide the business system with an environment to cope with various disasters, thereby ensuring the availability and data integrity of the business system. In the existing disaster recovery system, the first data center (data center 1, DC1) is the production center that carries the business, and it forms a dual-active data center with the production center DC2 in the same area to keep the data of DC1 and DC2 synchronized in real time. Then, DC1 relies on the data transmission between DC2 and DC3 and the data transmission between DC3 and DC4 to achieve the consistency of the business data of DC1 and the other three DCs. Specifically, asynchronous replication technology is used between DC2 and the disaster recovery center DC3 in another city to keep the data cycle of DC2 and DC3 synchronized, so that the data cycle of DC1 and DC3 can also be kept synchronized. DC3 transmits data in real time with the disaster recovery center DC4 in the same area, so that the data cycle of DC1 and DC4 can also be kept synchronized. In summary, in the disaster recovery system composed of "series", when the four DCs are operating normally, the data of DC1 can be consistent with the data of the other three DCs.

[0003] However, when the data of DC2 is destroyed due to irresistible natural disasters such as fire, flood, earthquake, tsunami, or man-made disasters such as computer crime, computer virus, power failure, network / communication failure, hardware / software error, and human operation error, although DC1 can take over the business on DC2, DC1 becomes an "island" because it cannot communicate with DC3 and DC4, and loses the disaster recovery protection of DC3 and DC4. Therefore, when DC2 fails, the disaster recovery scale of the disaster recovery system will be reduced from four to one, causing the disaster recovery performance of the disaster recovery system to drop drastically. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a disaster recovery method, device and system, which can achieve a step-by-step reduction in the disaster recovery performance of the disaster recovery system, thereby meeting user demands to the greatest extent possible.

[0005] In the first aspect, a disaster recovery system is provided, including: a first data center, a second data center, a third data center and a fourth data center. Among them, the first data center and the second data center are both production centers, the third data center and the fourth data center are both disaster recovery centers, the first data center and the second data center are located in the first area, the third data center and the fourth data center are located in the second area different from the first area, there is a first synchronous replication link between the first data center and the second data center, there is a second synchronous replication link between the third data center and the fourth data center, and there is a first asynchronous replication link between the first data center and the third data center. Further, the disaster recovery system also includes one, two or three of the following backup asynchronous replication links: there is a second asynchronous replication link between the first data center and the fourth data center, there is a third asynchronous replication link between the second data center and the third data center, and there is a fourth asynchronous replication link between the second data center and the fourth data center. Among them, the synchronous replication link is used to synchronously replicate the backup data of the data centers at both ends of the synchronous replication link. The asynchronous replication link is used to asynchronously replicate the backup data of the data centers at both ends of the asynchronous replication link. The backup asynchronous replication link is used to asynchronously replicate the backup data between the first area and the second area in the event of a failure in the first data center or the third data center.

[0006] In the above scheme, after establishing the first synchronous replication link, the second synchronous replication link and the first asynchronous replication link, by establishing one, two or three backup asynchronous replication links, the data centers at both ends of the backup asynchronous replication link can directly transmit the backup data and synchronize the backup data, so that there is a disaster recovery protection relationship between the data centers at both ends of the backup asynchronous replication link. In the event of a failure in the first data center or the third data center, the remaining three data centers that are operating normally can still synchronize the backup data through the backup asynchronous replication link. Therefore, the disaster recovery scale of the disaster recovery system is reduced from four to three, and the disaster recovery performance is reduced in a step-by-step manner, which meets the user's demands to the greatest extent.

[0007] In some possible implementations, the multiple backup asynchronous replication links are backup links for each other.

[0008] In the above scheme, multiple backup asynchronous replication links are made to serve as backup links for each other. Then, in the case of failure of any two data centers, for example, two data centers fail at the same time, or one data center fails first and then the other data center fails after running normally for a period of time, the disaster recovery system can still use the backup link to realize asynchronous replication of backup data, thereby synchronizing the backup data of the remaining two normally running data centers, ensuring that the disaster recovery scale of the disaster recovery system is reduced from four to two, and the disaster recovery performance is reduced in a step-by-step manner. Alternatively, in the case of sudden failure of any two asynchronous replication links, the disaster recovery system can still use the backup link to realize asynchronous replication of backup data, thereby realizing synchronization of backup data of four data centers, ensuring that the disaster recovery scale of the disaster recovery system is still four.

[0009] In some possible implementations, the backup data is a snapshot or a log.

[0010] In some possible implementations, the first data center and the second data center are active-active data centers. The first data center and the second data center are used to provide the same services and the same logical unit number LUN to the outside world, so that the client can access the first data center or the second data center through the LUN, and in the event of a failure in the first data center or the second data center, the client's access is switched to the data center that has not failed through the LUN.

[0011] In the above solution, the first data center and the second data center form an active-active data center, so that the two data centers can carry services at the same time, with high compatibility and applicability, high resource utilization and good user experience.

[0012] In some possible implementations, the third data center and the fourth data center are active-active data centers. The third data center and the fourth data center are used to provide the same services and the same logical unit number LUN to the outside world, so that the client can access the third data center or the fourth data center through the LUN, and in the event of a failure in the third data center or the fourth data center, the client's access is switched to the data center that has not failed through the LUN.

[0013] In the above solution, the third data center and the fourth data center form an active-active data center, so that the two data centers can carry services at the same time, with high compatibility and applicability, high resource utilization and good user experience.

[0014] In a second aspect, a disaster recovery method is provided, including: a first data center synchronously replicates backup data to a second data center via a first synchronous replication link. Furthermore, when the third data center is normal, the first data center asynchronously replicates the backup data to the third data center via a first asynchronous replication link. In the event of a failure in the third data center, the first data center asynchronously replicates the backup data to a fourth data center via a second asynchronous replication link. The first data center and the second data center are both production centers, the first data center and the second data center are located in a first region, the third data center and the fourth data center are both disaster recovery centers, and the third data center and the fourth data center are located in a second region different from the first region.

[0015] In the above scheme, the first data center synchronously replicates the backup data to the second data center through the first synchronous replication link, so that the backup data between the first data center and the second data center are consistent in real time. When the third data center is normal, the first data center asynchronously replicates the backup data to the third data center through the first asynchronous replication link, so that the backup data between the first data center and the third data center are consistent. Therefore, the synchronization of the backup data of the first data center, the second data center and the third data center is achieved. In addition, in the case of a failure of the third data center, the first data center can still send the backup data to the fourth data center through the second asynchronous replication link, thereby achieving the synchronization of the backup data of the first data center and the backup data of the fourth data center, that is, a disaster recovery protection relationship is established between the first data center and the fourth data center. In the case of a disaster recovery protection relationship between the first data center and the fourth data center, if the second data center is operating normally, the disaster recovery scale of the disaster recovery system will be reduced from four to three, and the disaster recovery performance will decrease in a step-by-step manner; if the second data center also fails, the disaster recovery scale of the disaster recovery system will be reduced from four to two, and the disaster recovery performance will still decrease in a step-by-step manner, thereby maximizing the satisfaction of user demands.

[0016] In some possible implementations, the above method also includes: in the event of a failure of the first data center, the second data center asynchronously replicates the above backup data to the third data center via a third asynchronous replication link, or the second data center asynchronously replicates the above backup data to the fourth data center via a fourth asynchronous replication link.

[0017] In the above scheme, in the case of failure of the first data center, the third data center can still receive the backup data sent by the second data center through the third asynchronous replication link, thereby realizing synchronization of the backup data of the third data center with the backup data of the second data center, that is, establishing a disaster recovery protection relationship between the second data center and the third data center. In the case of a disaster recovery protection relationship between the second data center and the third data center, if the fourth data center operates normally, the disaster recovery scale of the disaster recovery system will be reduced from four to three, and the disaster recovery performance will decrease in a step-by-step manner; if the fourth data center also fails, the disaster recovery scale of the disaster recovery system will be reduced from four to two, and the disaster recovery performance will still decrease in a step-by-step manner, thereby maximizing the satisfaction of user demands.

[0018] In some possible implementations, the backup data is a snapshot or a log.

[0019] In some possible implementations, the first data center and the second data center are active-active data centers. The first data center and the second data center provide the same services and the same logical unit number LUN to the outside world, so that the client accesses the first data center or the second data center through the LUN, and in the event of a failure in the first data center or the second data center, the client's access is switched to the data center that has not failed through the LUN.

[0020] In some possible implementations, the first data center synchronously replicates the backup data to the second data center through the first synchronous replication link, including:

[0021] The first data center synchronously replicates the backup data and the data generated according to the client's access to the first data center to the second data center through the first synchronous replication link.

[0022] In the above solution, in addition to synchronously replicating the backup data, the first data center also synchronously replicates the data generated according to the client's access to the second data center, thereby being able to synchronize the data between the first data center and the second data center in real time.

[0023] In some possible implementations, the third data center and the fourth data center are active-active data centers. The third data center and the fourth data center provide the same services and the same logical unit number LUN to the outside, so that the client accesses the third data center or the fourth data center through the LUN, and in the event of a failure in the third data center or the fourth data center, the client's access is switched to the data center that has not failed through the LUN.

[0024] In a third aspect, a disaster recovery device is provided, which is used as a first disaster recovery device, and includes: a synchronous replication module and an asynchronous replication module. The synchronous replication module is used to synchronously replicate backup data to a second disaster recovery device through a first synchronous replication link. The asynchronous replication module is used to asynchronously replicate the backup data to a third disaster recovery device through a first asynchronous replication link when the third disaster recovery device is normal. The asynchronous replication module is also used to asynchronously replicate the backup data to a fourth disaster recovery device through a second asynchronous replication link when a failure occurs in the third disaster recovery device. The first disaster recovery device and the second disaster recovery device are both production centers, the first disaster recovery device and the second disaster recovery device are located in a first area, the third disaster recovery device and the fourth disaster recovery device are both disaster recovery centers, and the third disaster recovery device and the fourth disaster recovery device are located in a second area different from the first area.

[0025] In some possible implementations, the backup data is a snapshot or a log.

[0026] In some possible implementations, the first disaster recovery device and the second disaster recovery device provide the same services and the same logical unit number LUN to the outside, so that the client can access the first disaster recovery device or the second disaster recovery device through the LUN, and in the event of a failure of the first disaster recovery device or the second disaster recovery device, the client's access is switched to the disaster recovery device that has not failed through the LUN.

[0027] In some possible implementations, the synchronous replication module is specifically used to synchronously replicate the backup data and the data generated by the client accessing the first disaster recovery device to the second disaster recovery device through the first synchronous replication link.

[0028] In a fourth aspect, a storage server is provided, comprising a processor and a memory, wherein the memory is used to store instructions, and the processor is used to execute the instructions. When the processor executes the instructions, the method of any one of the second aspect is performed.

[0029] In a fifth aspect, a computer program product comprising instructions is provided, which, when executed by a computing device, causes the computing device to execute any method as described in the second aspect.

[0030] In a sixth aspect, a computer-readable storage medium is provided, comprising computer program instructions. When the computer program instructions are executed by a computing device, the computing device executes a method as described in any one of the second aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the background technology, the drawings required for use in the embodiments of the present application or the background technology will be described below.

[0032] Figure 1It is a structural diagram of a disaster recovery system provided by this application;

[0033] Figure 2A It is a structural diagram of a disaster recovery system under a normal operating mechanism provided by this application;

[0034] Figure 2B It is a structural diagram of a disaster recovery system under another normal operating mechanism provided by this application;

[0035] Figure 2C It is a structural diagram of a disaster recovery system under another normal operating mechanism provided by this application;

[0036] Figure 2D It is a structural diagram of a disaster recovery system under another normal operating mechanism provided by this application;

[0037] Figure 3 It is a flowchart of a data synchronization method provided by this application;

[0038] Figure 4A It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data center A;

[0039] Figure 4B It is a schematic diagram of the operation mechanism of another disaster recovery system provided by the present application under the failure of data center A;

[0040] Figure 4C It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data center B;

[0041] Figure 4D It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data center C;

[0042] Figure 4E It is a schematic diagram of the operation mechanism of another disaster recovery system provided by the present application under the failure of data center C;

[0043] Figure 4F It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data center D;

[0044] Figure 4G It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers A and B;

[0045] Figure 4H It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers A and C;

[0046] Fig. 4I It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers A and D;

[0047] Figure 4J It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers B and C;

[0048] Figure 4K It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers B and D;

[0049] Figure 4L It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers C and D;

[0050] Figure 4M It is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of an asynchronous replication link between data centers A and C;

[0051] Figure 4N It is a schematic diagram of the operation mechanism of another disaster recovery system provided by the present application under the failure of the asynchronous replication link between data centers A and C;

[0052] Fig.4O It is a schematic diagram of the operation mechanism of another disaster recovery system provided by the present application under the failure of the asynchronous replication link between data centers A and C;

[0053] Figure 5 It is a flowchart of a disaster recovery method provided by this application;

[0054] Figure 5A This is a flow chart of a preset disaster recovery solution adopted in case of a failure in data center A provided by the present application;

[0055] Figure 5B This is a flow chart of a preset disaster recovery solution adopted in the event of a failure in data center C provided by the present application;

[0056] Figure 5C This is a flowchart of a preset disaster recovery solution for failure of two data centers provided by the present application;

[0057] Figure 5D This is a flow chart of a preset disaster recovery solution adopted when any asynchronous replication link fails, provided by the present application;

[0058] Figure 6 It is a structural schematic diagram of a disaster recovery device provided by the present application;

[0059] Figure 7 It is a structural diagram of a storage server provided by this application. DETAILED DESCRIPTION

[0060] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0061] In order to solve the problem of a precipitous drop in the disaster recovery performance of the disaster recovery system due to a DC2 failure in the existing disaster recovery system, the present application provides a disaster recovery method, device and system, which can realize full interconnection of four data centers, so that there is a disaster recovery protection relationship between any two data centers, and can realize a step-by-step decrease in the disaster recovery performance of the disaster recovery system, thereby maximally satisfying user demands.

[0062] See also Figure 1 , Figure 1 Schematic diagram of the structure of a disaster recovery system provided by this application. The disaster recovery system is used to provide a business system with an environment for coping with various disasters, ensuring the availability of the business system and the integrity of business data. Figure 1 As shown, the disaster recovery system 10 includes four data centers, namely a first data center 11 (DC11), a second data center 12 (DC12), a third data center 13 (DC13) and a fourth data center 14 (DC14). Among them, DC11 and DC12 are both production centers, DC13 and DC14 are both disaster recovery centers, DC11 and DC12 are in a first region 151, DC13 and DC14 are in a second region 152, and the first region 151 and the second region 152 are different regions. Among them, the region is used to describe the geographical location relationship, which can be a country, province, city, etc. DC11 and DC12 are located in the same region, so their geographical location distance is closer, such as in the same city. DC11 and DC13 are located in different regions, so their geographical location distance is farther, such as in different cities.

[0063] In the disaster recovery system 10, DC11 and DC12 in the first area 151 form an active-active data center, so that the two data centers can carry services at the same time, with high compatibility and applicability, high resource utilization, and good user experience. Specifically, when the four data centers are operating normally, DC11 and DC12 jointly carry the services of the first area 151. DC11 and DC12 provide the same logical unit number (logical unit number, LUN) and the same services to the outside world, so that the client 16a can access DC11 or DC12 through LUN to handle the user's business. At this time, DC11 is responsible for generating, storing and managing the backup data 111 in DC11, and DC12 is responsible for generating, storing and managing the backup data 121 in DC12. DC13, as a disaster recovery center for storing backup data, is responsible for storing and managing the backup data 131 in DC13. DC14, as a disaster recovery center for storing backup data, is responsible for storing and managing the backup data 141 in DC14. In the event of a failure of DC11 or DC12, the access of the client 16a is switched to the data center that has not failed through LUN. In the case that both DC11 and DC12 fail, DC13 and DC14 form an active-active data center to jointly carry the services of the first area 151. Specifically, DC13 and DC14 provide the same LUN and the same services to the outside, so that the client 16a can access DC13 or DC14 through the LUN to handle the user's services. At this time, DC13 is responsible for generating, storing and managing backup data 131, and DC14 is responsible for generating, storing and managing backup data 141.

[0064] In the disaster recovery system 10, a synchronous replication link is established between two data centers in the same region, so that the two data centers in the same region synchronously replicate the backup data. The synchronous replication link is established between the two data centers in the same region, including:

[0065] (1) A first synchronous replication link 171 is established between DC11 and DC12. In a specific implementation, when the first synchronous replication link 171 is in a normal state, it is connected to DC11 and DC12 for communication, respectively, and is used to send the backup data 112 generated by DC11 to DC12, or to send the backup data 122 generated by DC12 to DC11, so that the backup data 111 in DC11 and the backup data 121 in DC12 are synchronized in real time to maintain consistency. In another specific implementation, when the first synchronous replication link 171 is in a normal state, it is also used to send the data 113 generated by the first change operation performed by DC11 according to the first business request 161 sent by the client 16a to DC12, or to send the data 123 generated by the second change operation performed by DC12 according to the second business request 162 sent by the client 16b to DC11. The first change operation or the second change operation includes an add operation, a write operation, a modify operation, a delete operation, and the like.

[0066] (2) A second synchronous replication link 172 is established between DC13 and DC14. In a specific implementation, when the second synchronous replication link 172 is in a normal state, it is connected to DC13 and DC14 for communication, and is used to send the backup data 132 in DC13 to DC14, or to send the backup data 142 in DC14 to DC13, so that the backup data 131 in DC13 and the backup data 141 in DC14 are synchronized in real time to maintain consistency. When both DC11 and DC12 fail and the second synchronous replication link 172 is in a normal state, the second synchronous replication link 172 is also used to send the data 133 generated by the third change operation performed by DC13 according to the service request sent by client 16a or client 16b to DC14, or to send the data 143 generated by the fourth change operation performed by DC14 according to the service request sent by client 16a or client 16b to DC14. The third change operation or the fourth change operation includes an add operation, a write operation, a modify operation, a delete operation, and the like.

[0067] In the disaster recovery system 10, an asynchronous replication link is established between two data centers in different regions, so that the two data centers in different regions asynchronously replicate backup data. The asynchronous replication link is established between two data centers in different regions, including:

[0068] (1) A first asynchronous replication link 173 is established between DC11 and DC13, with DC11 serving as the primary data center and DC13 serving as the secondary data center. When the first asynchronous replication link 173 is in normal state, it is connected to DC11 and DC13 for communication, and is used to send the backup data 112 generated by DC11 to DC13, so that DC13 stores the backup data 112, thereby making the backup data 131 in DC13 consistent with the backup data 111 in DC11. Usually, after DC11 generates the backup data 112, it sends it to DC12 in real time, and sends it to DC13 in non-real time. For example, DC11 sends the backup data 112 to DC13 only within the asynchronous replication cycle.

[0069] (2) A second asynchronous replication link 174 is established between DC11 and DC14, with DC11 serving as the primary data center and DC14 serving as the secondary data center. When the second asynchronous replication link 174 is in normal state, it is connected to DC11 and DC14 for communication, and is used to send the backup data 112 generated by DC11 to DC14, so that DC14 stores the backup data 112, thereby making the backup data 141 in DC14 and the backup data 111 in DC11 consistent. Usually, after DC11 generates the backup data 112, it sends it to DC12 in real time, and sends it to DC14 in non-real time. For example, DC11 sends the backup data 112 to DC14 only within the asynchronous replication cycle.

[0070] (3) A third asynchronous replication link 175 is established between DC12 and DC13, with DC12 serving as the primary data center and DC13 serving as the secondary data center. When the third asynchronous replication link 175 is in normal state, it is connected to DC12 and DC13 for communication, respectively, and is used to send the backup data 122 generated by DC12 to DC13, so that DC13 stores the backup data 122, thereby making the backup data 131 in DC13 consistent with the backup data 121 in DC12. Usually, after DC12 generates the backup data 122, it sends it to DC11 in real time, and sends it to DC13 in non-real time. For example, DC12 sends the backup data 122 to DC13 only within the asynchronous replication cycle.

[0071] (4) A fourth asynchronous replication link 176 is established between DC12 and DC14, with DC12 serving as the primary data center and DC14 serving as the secondary data center. When the fourth asynchronous replication link 176 is in a normal state, it is connected to DC12 and DC14 for communication, respectively, and is used to send the backup data 122 generated by DC12 to DC14, so that DC14 stores the backup data 122, thereby making the backup data 141 in DC14 consistent with the backup data 121 in DC12. Usually, after DC12 generates the backup data 122, it sends it to DC11 in real time, and sends it to DC14 in non-real time. For example, DC12 sends the backup data 122 to DC14 only within the asynchronous replication cycle.

[0072] In summary, a synchronous replication link is established between two data centers in the same region, and an asynchronous replication link is established between two data centers in different regions, thereby enabling full interconnection of all data centers in the disaster recovery system.

[0073] In some possible implementations, the backup data may be snapshots, logs, data files, etc. The following will take snapshots as an example of backup data and describe the operation mechanism of the disaster recovery system in combination with the structure of the disaster recovery system.

[0074] The disaster recovery system includes four data centers, namely, data center A, data center B, data center C, and data center D, wherein data center A and data center B are in the first area 151, and data center C and data center D are in the second area 152. The normal operation mechanism of the disaster recovery system is that data centers A, B, C, and D all operate normally. At this time, the first synchronous replication link between data center A and data center B can operate normally, the asynchronous replication link between data center A and data center C can operate normally, and the second synchronous replication link between data center C and data center D can operate normally, so as to achieve snapshot consistency of the four data centers.

[0075] See also Figure 2A-2D , Figure 2A This is a structural diagram of a disaster recovery system under a normal operating mechanism provided by this application. Figure 2B , Figure 2C , Figure 2D They are respectively structural schematic diagrams of a disaster recovery system under another normal operating mechanism provided by the present application.

[0076] In such Figure 2AIn the disaster recovery system 10 shown, data center A is DC11, data center B is DC12, data center C is DC13, and data center D is DC14. Therefore, the first synchronous replication link 171, the second synchronous replication link 172, and the first asynchronous replication link 173 can be set to a normal state manually or by software, and the second asynchronous replication link 174, the third asynchronous replication link 175, and the fourth asynchronous replication link 176 can be set to a ready state. Among them, the normal state means normal operation, and the ready state means non-operation. On this basis, the normal operation mechanism of the disaster recovery system 10 is as follows:

[0077] First, DC11 performs a snapshot operation according to a first preset period, generates a first snapshot 114, and sends the first snapshot 114 to DC12 through a first synchronous replication link 171, so that the snapshot 125 stored in DC12 and the snapshot 115 stored in DC11 are synchronized in real time to maintain consistency. The first preset period is determined by the user and can be 1 week, 1 day, 1 hour, etc.

[0078] Then, within the asynchronous replication cycle, DC11 sends the first snapshot 114 to DC13 via the first asynchronous replication link 173, so that the snapshot 135 stored in DC13 and the snapshot 115 stored in DC11 are periodically synchronized to maintain consistency. The asynchronous replication cycle is determined by the user and can be 3 weeks, 3 days, 3 hours, etc.

[0079] Finally, after receiving the first snapshot 114 sent by DC11 through the first asynchronous replication link 173, DC13 performs a snapshot operation to generate a third snapshot 134, and sends the third snapshot 134 to DC14 through the second synchronous replication link 172, so that the snapshot 145 stored in DC14 and the snapshot 135 stored in DC13 are synchronized in real time to maintain consistency. Alternatively, DC13 performs a snapshot operation according to a second preset period to generate a third snapshot 134. The second preset period is determined by the user and can be 1 week, 1 day, 1 hour, etc.

[0080] In some possible implementations, the first snapshot 114 may be one or more. For example, when the first preset period is 0:00 to 24:00 of the current day and the asynchronous replication period is 12:00 to 22:00 of the current day, DC11 performs a snapshot operation at 0:00 every day, generates one first snapshot 114, and sends the first snapshot 114 to DC13 at 12:00 through the first asynchronous replication link 173. When the first preset period is 0:00 to 24:00 of the current day and the asynchronous replication period is 12:00 to 22:00 of the third day, DC11 performs a snapshot operation at 0:00 every day, generates a first snapshot 114, and thus DC11 needs to transmit a total of three first snapshots 114 during the asynchronous replication period. DC11 can send the first snapshot 114 to DC13 through the first asynchronous replication link 173 each time it generates a first snapshot 114; it can also send the three first snapshots 114 together to DC13 through the first asynchronous replication link 173 after generating three first snapshots 114; it can also perform a snapshot operation on the three first snapshots 114 again after generating the three first snapshots 114 to generate a new snapshot, and send the new snapshot to DC13 through the first asynchronous replication link 173. For DC13, it can perform a snapshot operation after receiving each first snapshot 114 to generate a third snapshot 134, and send the third snapshot 134 to DC14 through the second synchronous replication link 172; it can also perform a snapshot operation after receiving three first snapshots 114 to generate a third snapshot 134, and send the third snapshot 134 to DC14 through the second synchronous replication link 172; it can also perform a snapshot operation after receiving a new snapshot including the three first snapshots 114 to generate a third snapshot 134, and send the third snapshot 134 to DC14 through the second synchronous replication link 172.

[0081] In summary, Figure 2A In the disaster recovery system 10 shown, snapshots of four data centers can be synchronized.

[0082] It should be understood that the transmission of the first snapshot 114 between DC11 and DC12 is only an example and is not specifically limited here. Figure 2A In the disaster recovery system 10 shown, DC 12 may also perform a snapshot operation according to a first preset period to generate a second snapshot 124 , and send the second snapshot 124 to DC 11 through the first synchronous replication link 171 .

[0083] In such Figure 2BIn the disaster recovery system 10 shown, data center A is DC11, data center B is DC12, data center C is DC14, and data center D is DC13. Therefore, the first synchronous replication link 171, the second synchronous replication link 172, and the second asynchronous replication link 174 can be set to a normal state manually or by software, and the first asynchronous replication link 173, the third asynchronous replication link 175, and the fourth asynchronous replication link 176 can be set to a ready state. On this basis, the normal operation mechanism of the disaster recovery system 10 is as follows:

[0084] First, DC11 performs a snapshot operation according to a first preset period, generates a first snapshot 114, and sends the first snapshot 114 to DC12 through the first synchronous replication link 171, so that the snapshot 125 stored in DC12 and the snapshot 115 stored in DC11 are synchronized in real time to maintain consistency. In a specific implementation, DC11 performs a snapshot operation for the first time, generates a full snapshot, and uses the full snapshot as the first snapshot 114, and generates an incremental snapshot when performing a snapshot operation subsequently, and uses the incremental snapshot as the first snapshot 114.

[0085] Next, within the asynchronous replication cycle, DC11 sends the first snapshot 114 to DC14 via the second asynchronous replication link 174, so that the snapshot 145 stored in DC14 and the snapshot 115 stored in DC11 are periodically synchronized to maintain consistency.

[0086] Finally, after receiving the first snapshot 114 sent by DC11 through the second asynchronous replication link 174, DC14 performs a snapshot operation to generate a fourth snapshot 144, and sends the fourth snapshot 144 to DC13 through the second synchronous replication link 172, so that the snapshot 135 stored in DC13 and the snapshot 145 stored in DC14 are synchronized in real time to maintain consistency. Alternatively, DC14 performs a snapshot operation according to the second preset period to generate a fourth snapshot 144.

[0087] In summary, Figure 2B In the disaster recovery system 10 shown, snapshots of four data centers can be synchronized.

[0088] In such Figure 2C In the disaster recovery system 10 shown, data center A is DC12, data center B is DC11, data center C is DC13, and data center D is DC14. Therefore, the first synchronous replication link 171, the second synchronous replication link 172, and the third asynchronous replication link 175 can be set to a normal state manually or by software, and the first asynchronous replication link 173, the second asynchronous replication link 174, and the fourth asynchronous replication link 176 can be set to a ready state. On this basis, the normal operation mechanism of the disaster recovery system 10 is the same as the above Figure 2AThe normal operation mechanism of the disaster recovery system 10 is similar, and for the sake of brevity of the specification, it will not be elaborated here.

[0089] In such Figure 2D In the disaster recovery system 10 shown, data center A is DC12, data center B is DC11, data center C is DC14, and data center D is DC13. Therefore, the first synchronous replication link 171, the second synchronous replication link 172, and the fourth asynchronous replication link 176 can be set to a normal state manually or by software, and the first asynchronous replication link 173, the second asynchronous replication link 174, and the third asynchronous replication link 175 can be set to a ready state. On this basis, the normal operation mechanism of the disaster recovery system 10 is the same as the above Figure 2B The normal operation mechanism of the disaster recovery system 10 is similar, and for the sake of brevity of the specification, it will not be elaborated here.

[0090] In summary, when all four DCs are operating normally, the disaster recovery system can select the working asynchronous replication link, so the disaster recovery system has high flexibility.

[0091] Based on the structure and normal operation mechanism of the aforementioned disaster recovery system, a data synchronization method provided by the present application is specifically introduced below. The data synchronization method can be applied to the entire process of asynchronous replication, including asynchronous replication of the first cycle, asynchronous replication of the second cycle, ..., asynchronous replication of the kth cycle, where k is a positive integer, so that the backup data of two data centers from different regions are consistent, and then the backup data of multiple data centers in the disaster recovery system are consistent. Taking the asynchronous replication of the kth cycle as an example, a data synchronization method provided by the present application is specifically introduced below.

[0092] See also Figure 3 , Figure 3 This is a flow chart of a data synchronization method provided by this application. Figure 3 As shown, the data synchronization method provided by the present application is applied to a disaster recovery system, wherein the disaster recovery system includes data center A, data center B, data center C and data center D, and the data synchronization method includes:

[0093] S301: Synchronize the snapshot A(t) of data center A and the snapshot B(t) of data center B in real time through the first synchronous replication link.

[0094] The first synchronous replication link may be Figure 1 The first synchronous replication link 171 in the disaster recovery system 10. Data center A and data center B are in the first region. Data center A can be Figure 1 DC11 or DC12 in the disaster recovery system 10. When data center A is Figure 1When DC11 in the disaster recovery system 10, data center B is Figure 1 DC12 in the disaster recovery system 10. When data center A is Figure 1 When DC12 in the disaster recovery system 10, data center B is Figure 1 DC11 in the disaster recovery system 10. The first area can be Figure 1 The first area 151 in the disaster recovery system 10.

[0095] In some possible implementations, a snapshot includes metadata and data, wherein the metadata represents the attributes of the snapshot, including the name of the snapshot, the generation time of the snapshot, the name of the storage server, the file name, the file modification time, etc. The data is the data in the storage server when the snapshot operation is performed. Taking snapshot A(t) as an example, snapshot A(t) includes metadata A(t) and data A(t). Metadata A(t) represents the attributes of snapshot A(t), including the name of snapshot A(t), the generation time t of snapshot A(t), the name of storage server A, the file name, the file modification time, etc. The data A(t) is the data A(t) in the storage server A when the snapshot operation is performed.

[0096] In a specific implementation, the snapshot also includes a global identifier, and the global identifier is used to identify the data center that generates the snapshot. Taking snapshot A(t) as an example, the specific process of setting the global identifier t in snapshot A(t) is as follows: First, data center A can obtain the identifier of snapshot A(t). In a specific implementation, data center A sorts all snapshots A according to the generation time of snapshot A, thereby obtaining the serial number of snapshot A(t), and uses it as the identifier of snapshot A(t). It should be understood that the identifier of snapshot A(t) can also be the name, generation time, etc. of snapshot A(t). Then, since data center A is deployed on one or more storage servers A, data center A can obtain the identifier of storage server A. The identifier of storage server A can be a world wide name (WWN), a universally unique identifier (UUID), a device unique identifier (DUID), etc. In particular, when there are multiple storage servers A, data center A obtains the identifier of storage server A that generates snapshot A(t). Subsequently, data center A combines the identifier of snapshot A(t) and the identifier of storage server A into a global identifier t, and adds the global identifier t to the metadata A(t) of snapshot A(t), thereby obtaining snapshot A(t) containing the global identifier t. In this scheme, multiple snapshots generated by the same storage server can be identified by the identifier of the snapshot; and multiple snapshots generated by different storage servers can be identified by the identifier of the storage server. Therefore, in the case where the global identifier includes the identifier of the snapshot and the identifier of the storage server, all snapshots generated by all data centers can be uniquely identified by the global identifier, and all data centers can store snapshots containing the same global identifier to ensure the consistency of snapshots in all data centers.

[0097] In some possible implementations, real-time synchronization of the snapshot A(t) of data center A and the snapshot B(t) of data center B through the first synchronous replication link includes at least the following two methods: (1) Data center B receives the snapshot A(t) generated by data center A through the first synchronous replication link, and uses the snapshot A(t) as snapshot B(t). Specifically, after data center A generates the snapshot A(t) containing the global identifier t, it sends the snapshot A(t) to data center B through the first synchronous replication link. After data center B receives the snapshot A(t) sent by data center A through the first synchronous replication link, storage server B performs a write operation on the metadata A(t) and data A(t) in the snapshot A(t), and writes the metadata A(t) and data A(t) to storage server B. After the write operation is completed, data center B obtains snapshot B(t). (2) Data center A receives the snapshot B(t) generated by data center B through the first synchronous replication link, and uses the snapshot B(t) as snapshot A(t). The specific process can refer to the above method (1) in which data center B receives the snapshot A(t) generated by data center A through the first synchronous replication link and uses the snapshot A(t) as the full snapshot B(t). For the sake of brevity, it will not be elaborated here.

[0098] S302: Periodically synchronize the snapshot A(t) of data center A and the snapshot C(t) of data center C through an asynchronous replication link.

[0099] In some possible implementations, when data center A is DC11 and data center C is DC13, the asynchronous replication link can be Figure 1 The first asynchronous replication link 173 in the disaster recovery system 10. When data center A is DC11 and data center C is DC14, the asynchronous replication link can be Figure 1 The second asynchronous replication link 174 in the disaster recovery system 10. When data center A is DC12 and data center C is DC13, the asynchronous replication link can be Figure 1 The third asynchronous replication link 175 in the disaster recovery system 10. When data center A is DC12 and data center C is DC14, the asynchronous replication link can be Figure 1 The fourth asynchronous replication link 176 in the disaster recovery system 10.

[0100] In some possible implementations, during the asynchronous replication cycle, data center A sends snapshot A(t) to data center C via an asynchronous replication link. After data center C receives snapshot A(t) sent by data center A via the asynchronous replication link, storage server C performs a write operation on metadata A(t) and data A(t) in snapshot A(t). After completing the write operation, storage server C performs a snapshot operation to generate snapshot C(t).

[0101] S303: Synchronize the snapshot C(t) of data center C and the snapshot D(t) of data center D in real time through the second synchronous replication link.

[0102] The second synchronous replication link can be Figure 1 The second synchronous replication link 172 in the disaster recovery system 10. Data center C and data center D are in a second area different from the first area. Data center C can be Figure 1 DC13 or DC14 in the disaster recovery system 10. When data center C is Figure 1 In the case of DC13 in the disaster recovery system 10, data center D is Figure 1 DC14 in the disaster recovery system 10. When data center C is Figure 1 In the case of DC14 in the disaster recovery system 10, data center D is Figure 1 DC 13 in the disaster recovery system 10. The second area can be Figure 1 The second area 152 in the disaster recovery system 10.

[0103] In some possible implementations, after data center D receives snapshot C(t) sent by data center C through the second synchronous replication link, storage server D performs a write operation on metadata C(t) and data C(t) in snapshot C(t). After the write operation is completed, data center D obtains snapshot D(t).

[0104] In summary, the data synchronization method proposed in this application is based on the realization of full interconnection of four data centers in the disaster recovery system, synchronous replication of snapshots between two data centers in the same area, and asynchronous replication of snapshots between two data centers in different areas. The two data centers in different areas are used as driving centers, and the two data centers are made to generate snapshots containing the same data during the asynchronous replication cycle to ensure that the snapshots of the two data centers are consistent. At the same time, combined with the synchronous replication between the two data centers in the same area, the snapshots of the other two data centers are also consistent. Therefore, any two data centers have consistent snapshots.

[0105] However, when a fault occurs in the disaster recovery system 10, the disaster recovery system 10 can adjust the operating mechanism of the disaster recovery system 10 according to the type and number of faulty data centers, based on the structure of the disaster recovery system 10 and the data synchronization method, to achieve a step-by-step reduction in the disaster recovery performance of the disaster recovery system 10.

[0106] The present application will now describe the operating mechanism of the disaster recovery system 10 under eleven different failure scenarios.

[0107] (1) Data Center A fails

[0108] Assume that data center A suddenly experiences a catastrophic failure and is completely unavailable, while data centers B, C, and D are all operating normally. At this time, since data centers A and B in the first region are active-active data centers, client 16a that is processing business in data center A can be seamlessly switched to data center B through LUN to continue processing business, while client 16b that was originally processing business in data center B is not affected.

[0109] After a failure occurs in data center A, the disaster recovery system 10 includes at least the following two operating mechanisms:

[0110] a) Asynchronous replication of snapshots is achieved through the asynchronous replication link between data center B and data center C;

[0111] b) Asynchronous replication of snapshots is achieved through the asynchronous replication link between data center B and data center D.

[0112] It can be seen that the asynchronous replication link between data center B and data center C and the asynchronous replication link between data center B and data center D are both backup asynchronous replication links.

[0113] See also Figure 4A , Figure 4A FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data center A. Figure 4A As shown, for the above situation a), the asynchronous replication link between data center B and data center C is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, within the asynchronous replication cycle, data center B sends snapshot B(n) to data center C through the asynchronous replication link. After data center C receives snapshot B(n) sent by data center B through the asynchronous replication link, storage server C performs a write operation on metadata B(n) and data B(n) in snapshot B(n). After completing the write operation, storage server C performs a snapshot operation to generate snapshot C(n), so that the snapshot C(n) of data center C and the snapshot B(n) of data center B are synchronized in period to keep consistent. Subsequently, data center C sends snapshot C(n) to data center D through the second synchronous replication link. After data center D receives snapshot C(n) sent by data center C through the second synchronous replication link, storage server D performs a write operation on metadata C(n) and data C(n) in snapshot C(n). After the write operation is completed, data center D obtains snapshot D(n), so that snapshot D(n) of data center D and snapshot C(n) of data center C are synchronized in real time to maintain consistency.

[0114] See also Figure 4B , Figure 4BFIG. 1 is a schematic diagram of another disaster recovery system provided by the present application, showing an operation mechanism under a failure of data center A. Figure 4B As shown, for the above situation b), the asynchronous replication link between data center B and data center D is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, within the asynchronous replication cycle, data center B sends snapshot B(n) to data center D through the asynchronous replication link. After data center D receives snapshot B(n) sent by data center B through the asynchronous replication link, storage server D performs a write operation on metadata B(n) and data B(n) in snapshot B(n). After completing the write operation, storage server D performs a snapshot operation to generate snapshot D(n), so that the snapshot D(n) of data center D and the snapshot B(n) of data center B are synchronized in period to keep consistent. Subsequently, data center D sends snapshot D(n) to data center C through the second synchronous replication link. After data center C receives snapshot D(n) sent by data center D through the second synchronous replication link, storage server C performs a write operation on metadata D(n) and data D(n) in snapshot D(n). After the write operation is completed, data center C obtains snapshot C(n), so that snapshot C(n) of data center C and snapshot D(n) of data center D are synchronized in real time to maintain consistency.

[0115] (2) Data Center B fails

[0116] Assume that data center B suddenly encounters a catastrophic failure, while data centers A, C, and D are all operating normally. At this time, since data centers A and B in the first area are active-active data centers, client 16b that is processing business in data center B can be seamlessly switched to data center A through LUN to continue processing business, and client 16a that was originally processing business in data center A is not affected.

[0117] See also Figure 4C , Figure 4C FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data center B. Figure 4C As shown in the figure, after the failure of data center B, since the asynchronous replication link between data center A and data center C is still working normally, the snapshot A(n) of data center A and the snapshot C(n) of data center C can be synchronized periodically to maintain consistency. In addition, with the synchronous replication between data center C and data center D, the snapshot C(n) of data center C and the snapshot D(n) of data center D can be synchronized in real time to maintain consistency.

[0118] (3) Data Center C fails

[0119] Assume that a catastrophic failure occurs in data center C, while data centers A, B, and D are all operating normally. At this time, the client 16a that is processing business on data center A and the client 16b that is processing business on data center B are not affected. In addition, since the first synchronous replication link between data centers A and B is still operating normally, the snapshot A(n) of data center A and the snapshot B(n) of data center B can be synchronized in real time to maintain consistency.

[0120] After a failure occurs in data center C, the disaster recovery system 10 includes at least the following two operating mechanisms:

[0121] a) Asynchronous replication of snapshots is achieved through the asynchronous replication link between data center A and data center D;

[0122] b) Asynchronous replication of snapshots is achieved through the asynchronous replication link between data center B and data center D.

[0123] It can be seen that the asynchronous replication link between data center A and data center D and the asynchronous replication link between data center B and data center D are backup asynchronous replication links.

[0124] See also Figure 4D , Figure 4D FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the condition of a failure in data center C. Figure 4D As shown, for the above situation a), the asynchronous replication link between data center A and data center D is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, during the asynchronous replication cycle, data center A sends snapshot A(n) to data center D through the asynchronous replication link. After data center D receives snapshot A(n) sent by data center A through the asynchronous replication link, storage server D performs a write operation on metadata A(n) and data A(n) in snapshot A(n). After completing the write operation, storage server D performs a snapshot operation to generate snapshot D(n), so that the snapshot D(n) of data center D and the snapshot A(n) of data center A are periodically synchronized to maintain consistency.

[0125] See also Figure 4E , Figure 4E FIG. 1 is a schematic diagram of another operation mechanism of a disaster recovery system provided by the present application under a failure of a data center C. Figure 4EAs shown, for the above situation b), the asynchronous replication link between data center B and data center D is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, during the asynchronous replication cycle, data center B sends snapshot B(n) to data center D through the asynchronous replication link. After data center D receives snapshot B(n) sent by data center B through the asynchronous replication link, storage server D performs a write operation on metadata B(n) and data B(n) in snapshot B(n). After completing the write operation, storage server D performs a snapshot operation to generate snapshot D(n), so that the snapshot D(n) of data center D and the snapshot B(n) of data center B are periodically synchronized to maintain consistency.

[0126] (4) Data Center D fails

[0127] Assume that data center D suddenly experiences a catastrophic failure, while data centers A, B, and C are all operating normally. At this time, client 16a that is processing business on data center A and client 16b that is processing business on data center B are not affected. In addition, since the first synchronous replication link between data center A and data center B is still operating normally, snapshot A(n) of data center A and snapshot B(n) of data center B can be synchronized in real time to maintain consistency.

[0128] See also Figure 4F , Figure 4F FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under a failure of data center D. Figure 4F As shown, after a failure occurs in data center D, since the asynchronous replication link between data center A and data center C is still working normally, the snapshot A(n) of data center A and the snapshot C(n) of data center C can be periodically synchronized to maintain consistency.

[0129] To sum up, after a sudden failure of any data center, there is a disaster recovery protection relationship between any two of the remaining three data centers in the disaster recovery system 10, and snapshot synchronization of the three data centers can be achieved. Therefore, the disaster recovery scale of the disaster recovery system 10 is reduced from four to three, and the disaster recovery performance decreases step by step, which can meet user demands to the greatest extent.

[0130] (5) Data Center A and B fail

[0131] Assume that data centers A and B suddenly experience catastrophic failures, while data centers C and D are operating normally. At this time, the client 16a that is processing services on data center A is forced to interrupt, and the client 16b that is processing services on data center B is also forced to interrupt.

[0132] See also Figure 4G, Figure 4G This is a schematic diagram of the operation mechanism of a disaster recovery system provided by this application under the failure of data centers A and B. Figure 4G As shown, after a failure occurs in data centers A and B, the services of data centers A and B are manually switched to data centers C and D, and data centers C and D form an active-active data center, providing the same LUN and the same services to the outside world, so that clients 16a and 16b can access data center C or data center D through LUN, and re-process user services in data center C or data center D, and also enable the snapshot C(n) of data center C and the snapshot D(n) of data center D to be synchronized in real time to maintain consistency.

[0133] (6) Data Centers A and C fail

[0134] Assume that data centers A and C suddenly experience a catastrophic failure, while data centers B and D are operating normally. At this time, since data centers A and B in the first area are active-active data centers, client 16a that is processing business in data center A can be seamlessly switched to data center B through LUN to continue processing business, while client 16b that was originally processing business in data center B is not affected.

[0135] See also Figure 4H , Figure 4H FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers A and C. Figure 4H As shown, after a failure occurs in data centers A and C, the asynchronous replication link between data center B and data center D is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, within the asynchronous replication cycle, data center B sends snapshot B(n) to data center D through the asynchronous replication link. After data center D receives snapshot B(n) sent by data center B through the asynchronous replication link, storage server D performs a write operation on metadata B(n) and data B(n) in snapshot B(n). After completing the write operation, storage server D performs a snapshot operation to generate snapshot D(n), so that the snapshot D(n) of data center D and the snapshot B(n) of data center B are periodically synchronized to maintain consistency.

[0136] It can be seen that the asynchronous replication link between data center B and data center D is a backup asynchronous replication link.

[0137] (7) Data Centers A and D fail

[0138] Assume that data centers A and D suddenly experience a catastrophic failure, while data centers B and D are operating normally. At this time, since data centers A and B in the first area are active-active data centers, client 16a that is processing business in data center A can be seamlessly switched to data center B through LUN to continue processing business, while client 16b that was originally processing business in data center B is not affected.

[0139] See also Fig. 4I , Fig. 4I FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers A and D. Fig. 4I As shown, after a failure occurs in data centers A and D, the asynchronous replication link between data center B and data center C is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, within the asynchronous replication cycle, data center B sends snapshot B(n) to data center C through the asynchronous replication link. After data center C receives snapshot B(n) sent by data center B through the asynchronous replication link, storage server C performs a write operation on metadata B(n) and data B(n) in snapshot B(n). After completing the write operation, storage server C performs a snapshot operation to generate snapshot C(n), so that the snapshot C(n) of data center C and the snapshot B(n) of data center B are periodically synchronized to maintain consistency.

[0140] It can be seen that the asynchronous replication link between data center B and data center C is a backup asynchronous replication link.

[0141] (8) Data Centers B and C fail

[0142] Assume that data centers B and C suddenly experience a catastrophic failure, while data centers A and D are operating normally. At this time, since data centers A and B in the first area are active-active data centers, client 16b that is processing business in data center B can be seamlessly switched to data center A through LUN to continue processing business, while client 16a that was originally processing business in data center A is not affected.

[0143] See also Figure 4J , Figure 4J FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers B and C. Figure 4JAs shown, after a failure occurs in data centers B and C, the asynchronous replication link between data center A and data center D is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, during the asynchronous replication cycle, data center A sends snapshot A(n) to data center D through the asynchronous replication link. After data center D receives snapshot A(n) sent by data center A through the asynchronous replication link, storage server D performs a write operation on metadata A(n) and data A(n) in snapshot A(n). After completing the write operation, storage server D performs a snapshot operation to generate snapshot D(n), so that the snapshot D(n) of data center D and the snapshot A(n) of data center A are periodically synchronized to maintain consistency.

[0144] It can be seen that the asynchronous replication link between data center A and data center D is a backup asynchronous replication link.

[0145] (9) Data Centers B and D fail

[0146] Assume that data centers B and D suddenly experience a catastrophic failure, while data centers A and C are operating normally. At this time, since data centers A and B in the first area are active-active data centers, client 16b that is processing business in data center B can be seamlessly switched to data center A through LUN to continue processing business, while client 16a that was originally processing business in data center A is not affected.

[0147] See also Figure 4K , Figure 4K FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers B and D. Figure 4K As shown, after data centers B and D fail, since the asynchronous replication link between data center A and data center C is still working normally, the snapshot A(n) of data center A and the snapshot C(n) of data center C can be synchronized periodically to maintain consistency.

[0148] (10) Data Centers C and D fail

[0149] Assume that data centers C and D suddenly experience catastrophic failures, while data centers A and B are operating normally. At this time, the client 16a that is processing services in data center A and the client 16b that is processing services in data center B are not affected.

[0150] See also Figure 4L , Figure 4L FIG. 1 is a schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the failure of data centers C and D. Figure 4LAs shown, after data centers C and D fail, since the first synchronous replication link between data center A and data center B is still working normally, snapshot A(n) of data center A and snapshot B(n) of data center B can be synchronized in real time to maintain consistency.

[0151] To sum up, after any two data centers suddenly fail, the remaining two data centers in the disaster recovery system 10 still have a disaster recovery protection relationship, and snapshot synchronization of the two data centers can be achieved. Therefore, the disaster recovery scale of the disaster recovery system 10 is reduced from four to two, and the disaster recovery performance decreases step by step, which can meet user demands to the greatest extent.

[0152] (11) The asynchronous replication link between data center A and data center C fails

[0153] Assume that the asynchronous replication link between data center A and data center C fails, while data centers A, B, C, and D are all operating normally. At this time, the client 16a that is processing business on data center A and the client 16b that is processing business on data center B are not affected. In addition, since the first synchronous replication link between data center A and data center B is still operating normally, the snapshot A(n) of data center A and the snapshot B(n) of data center B can be synchronized in real time to maintain consistency.

[0154] After the asynchronous replication link between data center A and data center C fails, the disaster recovery system 10 includes at least the following three operating mechanisms:

[0155] a) Asynchronous replication of snapshots is achieved through the asynchronous replication link between data center A and data center D;

[0156] b) Asynchronous replication of snapshots is achieved through the asynchronous replication link between data center B and data center C;

[0157] c) Asynchronous replication of snapshots is achieved through the asynchronous replication link between data center B and data center D.

[0158] It can be seen that the asynchronous replication link between data center A and data center D, the asynchronous replication link between data center B and data center C, and the asynchronous replication link between data center B and data center D are all backup asynchronous replication links.

[0159] See also Figure 4M , Figure 4M Schematic diagram of the operation mechanism of a disaster recovery system provided by the present application under the condition of an asynchronous replication link failure between data centers A and C. Figure 4MAs shown, for the above situation a), the asynchronous replication link between data center A and data center D is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, within the asynchronous replication cycle, data center A sends snapshot A(n) to data center D through the asynchronous replication link. After data center D receives snapshot A(n) sent by data center A through the asynchronous replication link, storage server D performs a write operation on metadata A(n) and data A(n) in snapshot A(n). After completing the write operation, storage server D performs a snapshot operation to generate snapshot D(n), so that the snapshot D(n) of data center D and the snapshot A(n) of data center A are synchronized in period to keep consistent. Subsequently, data center D sends snapshot D(n) to data center C through the second synchronous replication link. After data center C receives snapshot D(n) sent by data center D through the second synchronous replication link, storage server C performs a write operation on metadata D(n) and data D(n) in snapshot D(n). After the write operation is completed, data center C obtains snapshot C(n), so that snapshot C(n) of data center C and snapshot D(n) of data center D are synchronized in real time to maintain consistency.

[0160] See also Figure 4N , Figure 4N FIG. 1 is a schematic diagram of another disaster recovery system provided by the present application, in which the asynchronous replication link between data centers A and C fails. Figure 4N As shown, for the above situation b), the asynchronous replication link between data center B and data center C is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, within the asynchronous replication cycle, data center B sends snapshot B(n) to data center C through the asynchronous replication link. After data center C receives snapshot B(n) sent by data center B through the asynchronous replication link, storage server C performs a write operation on metadata B(n) and data B(n) in snapshot B(n). After completing the write operation, storage server C performs a snapshot operation to generate snapshot C(n), so that the snapshot C(n) of data center C and the snapshot B(n) of data center B are synchronized in period to keep consistent. Subsequently, data center C sends snapshot C(n) to data center D through the second synchronous replication link. After data center D receives snapshot C(n) sent by data center C through the second synchronous replication link, storage server D performs a write operation on metadata C(n) and data C(n) in snapshot C(n). After the write operation is completed, data center D obtains snapshot D(n), so that snapshot D(n) of data center D and snapshot C(n) of data center C are synchronized in real time to maintain consistency.

[0161] See also Fig.4O , Fig.4O FIG. 1 is a schematic diagram of another disaster recovery system provided by the present application, in which the asynchronous replication link between data centers A and C fails. Fig.4O As shown, for the above situation c), the asynchronous replication link between data center B and data center D is set to a normal state manually or by software, so that the asynchronous replication link can work normally. After the asynchronous replication link can work normally, within the asynchronous replication cycle, data center B sends snapshot B(n) to data center D through the asynchronous replication link. After data center D receives snapshot B(n) sent by data center B through the asynchronous replication link, storage server D performs a write operation on metadata B(n) and data B(n) in snapshot B(n). After completing the write operation, storage server D performs a snapshot operation to generate snapshot D(n), so that the snapshot D(n) of data center D and the snapshot B(n) of data center B are synchronized in period to keep consistent. Subsequently, data center D sends snapshot D(n) to data center C through the second synchronous replication link. After data center C receives snapshot D(n) sent by data center D through the second synchronous replication link, storage server C performs a write operation on metadata D(n) and data D(n) in snapshot D(n). After the write operation is completed, data center C obtains snapshot C(n), so that snapshot C(n) of data center C and snapshot D(n) of data center D are synchronized in real time to maintain consistency.

[0162] It should be understood that the failure of the asynchronous replication link between data center A and data center C mentioned above is only an example of a link failure. It may also be the failure of the asynchronous replication link between data center A and data center D, the failure of the asynchronous replication link between data center B and data center C, the failure of the asynchronous replication link between data center B and data center D, etc., which is not specifically limited here.

[0163] In summary, if any one of the asynchronous replication links fails suddenly, the remaining three asynchronous replication links in the disaster recovery system 10 can still implement asynchronous replication of snapshots, thereby realizing snapshot synchronization of four data centers, and ensuring that the disaster recovery scale of the disaster recovery system 10 is still four. Furthermore, if any two asynchronous replication links fail suddenly, for example, two asynchronous replication links fail at the same time, or after one asynchronous replication link fails, the other asynchronous replication link also fails after working normally for a period of time, the remaining two asynchronous replication links in the disaster recovery system 10 can still implement asynchronous replication of snapshots, thereby realizing snapshot synchronization of four data centers, and ensuring that the disaster recovery scale of the disaster recovery system 10 is still four. Furthermore, if any three asynchronous replication links fail suddenly, the remaining one asynchronous replication link in the disaster recovery system 10 can still implement asynchronous replication of snapshots, thereby realizing snapshot synchronization of four data centers, and ensuring that the disaster recovery scale of the disaster recovery system 10 is still four. It can be seen that the four asynchronous replication links in the disaster recovery system 10 are backup links for each other.

[0164] Based on the operation mechanism of the above disaster recovery system, a disaster recovery method provided by the present application is specifically introduced below. The disaster recovery method can be applied to the disaster recovery system provided by the present application, so that the disaster recovery performance of the disaster recovery system can be reduced in a step-by-step manner when a data center fails, thereby meeting user demands to the greatest extent.

[0165] See also Figure 5 , Figure 5 This is a flowchart of a disaster recovery method provided by this application. Figure 5 As shown, the disaster recovery method provided by this application includes:

[0166] S501: Determine whether there is a fault in data center A, B, C, or D. If there is no fault, go to S502; if there is a fault, go to S505.

[0167] S502: Synchronize the snapshot A(n) of data center A and the snapshot B(n) of data center B in real time through the first synchronous replication link.

[0168] S503: Periodically synchronize the snapshot A(n) of data center A and the snapshot C(n) of data center C through an asynchronous replication link.

[0169] S504: Synchronize the snapshot C(n) of data center C and the snapshot D(n) of data center D in real time through the second synchronous replication link.

[0170] S505: Use a preset disaster recovery solution to keep the snapshot of the normally operating data center consistent.

[0171] The first synchronous replication link may be Figure 1The first synchronous replication link 171 in the disaster recovery system 10, the second synchronous replication link can be Figure 1 The second synchronous replication link 172 in the disaster recovery system 10. The above data center A can be Figure 1 DC11 or DC12 in the disaster recovery system 10. When data center A is DC11 and data center C is DC13, data center B is DC12, and the asynchronous replication link is the first asynchronous replication link 173. When data center A is DC11 and data center C is DC14, data center B is DC12, and the asynchronous replication link is the second asynchronous replication link 174. When data center A is DC12 and data center C is DC13, data center B is DC11, and the asynchronous replication link is the third asynchronous replication link 175. When data center A is DC12 and data center C is DC14, data center B is DC11, and the asynchronous replication link is the fourth asynchronous replication link 176.

[0172] The execution process of the above steps S502 to S504 can refer to the above Figure 3 The execution process of step S301 to step S303 will not be described in detail here for the sake of brevity of the description.

[0173] See also Figure 5A , Figure 5A This is a flowchart of a preset disaster recovery solution provided by the present application when data center A fails. Figure 5 Step S505: adopting a preset disaster recovery solution to keep the snapshot of the normally operating data center consistent, specifically including:

[0174] S511: Set the asynchronous replication link between data center B and data center a to a normal state, where data center a is any one of data centers C and D.

[0175] S512: Periodically synchronize the snapshot B(n) of data center B with the snapshot a(n) of data center a through an asynchronous replication link.

[0176] S513: Synchronize the snapshot a(n) of data center a and the snapshot b(n) of data center b in real time through the second synchronous replication link, where data center b is the other data center among data centers C and D.

[0177] This embodiment and Figure 4A , Figure 4B And the related instructions correspond to each other, you can refer to each other for implementation, and the repeated parts will not be repeated.

[0178] See also Figure 5B , Figure 5BThis is a flow chart of a preset disaster recovery solution provided by the present application when data center C fails. Figure 5 Step S505: adopting a preset disaster recovery solution to keep the snapshot of the normally operating data center consistent, specifically including:

[0179] S521: Set the asynchronous replication link between data center a and data center D to a normal state, where data center a is any one of data centers A and B.

[0180] S522: Periodically synchronize the snapshot a(n) of data center a and the snapshot D(n) of data center D through an asynchronous replication link.

[0181] This embodiment and Figure 4D , Figure 4E And the related instructions correspond to each other, you can refer to each other for implementation, and the repeated parts will not be repeated.

[0182] See also Figure 5C , Figure 5C This is a flowchart of a preset disaster recovery solution provided by the present application when two data centers fail. In the case that both data centers A and C fail, or both data centers A and D fail, or both data centers B and C fail, the above Figure 5 Step S505: adopting a preset disaster recovery solution to keep the snapshot of the normally operating data center consistent, specifically including:

[0183] S531: Set the asynchronous replication link between data center a and data center b to a normal state, where data center a is the data center that has not failed among data centers A and B, and data center b is the data center that has not failed among data centers C and D.

[0184] S532: Periodically synchronize the snapshot a(n) of data center a and the snapshot b(n) of data center b through an asynchronous replication link.

[0185] This embodiment and Figure 4H , Fig. 4I , Figure 4J And the related instructions correspond to each other, you can refer to each other for implementation, and the repeated parts will not be repeated.

[0186] See also Figure 5D , Figure 5DThis is a flowchart of a preset disaster recovery solution provided by the present application when any asynchronous replication link fails. In the case of an asynchronous replication link failure between data center A and data center C, or an asynchronous replication link failure between data center A and data center D, or an asynchronous replication link failure between data center B and data center C, or an asynchronous replication link failure between data center B and data center D, the above Figure 5 Step S505: adopting a preset disaster recovery solution to keep the snapshot of the normally operating data center consistent, specifically including:

[0187] S541: Set the asynchronous replication link that has not failed to be in a normal state.

[0188] S542: Periodically synchronize snapshot a(n) of data center a and snapshot b(n) of data center b through an asynchronous replication link that has not failed, wherein data centers a and b are data centers at both ends of the asynchronous replication link that has not failed, and data center a is in the first region and data center b is in the second region.

[0189] This embodiment and Figure 4M , Figure 4N , Fig.4O And the related instructions correspond to each other, you can refer to each other for implementation, and the repeated parts will not be repeated.

[0190] To summarize, the disaster recovery method proposed in the present application, based on the full interconnection of four data centers in the disaster recovery system, can realize snapshot synchronization of the data centers at both ends through the normally operating asynchronous replication link as long as there is still an asynchronous replication link that can work normally, so that the snapshots of the data centers that are still operating normally remain consistent, and a disaster recovery protection relationship exists between any two data centers that are still operating normally, thereby realizing a step-by-step decrease in the disaster recovery performance of the disaster recovery system and satisfying user demands to the greatest extent.

[0191] See also Figure 6 , Figure 6 600 is a schematic diagram of a disaster recovery device provided by the present application. The disaster recovery device 600 is used as a first disaster recovery device and can implement the above-mentioned disaster recovery method. Figure 6 As shown, the disaster recovery device 600 includes: a synchronous replication module 601 and an asynchronous replication module 602 .

[0192] The synchronous replication module 601 is used to synchronously replicate the backup data to the second disaster recovery device through the first synchronous replication link;

[0193] The asynchronous replication module 602 is used to asynchronously replicate the backup data to the third disaster recovery device through the first asynchronous replication link when the third disaster recovery device is normal;

[0194] The asynchronous replication module 602 is also used to asynchronously replicate the backup data to the fourth disaster recovery device through the second asynchronous replication link when the third disaster recovery device fails;

[0195] Among them, disaster recovery device 600 and the second disaster recovery device are both production centers, disaster recovery device 600 and the second disaster recovery device are located in the first area, the third disaster recovery device and the fourth disaster recovery device are both disaster recovery centers, and the third disaster recovery device and the fourth disaster recovery device are located in the second area different from the first area.

[0196] In some possible implementations, the backup data is a snapshot or a log.

[0197] In some possible implementations, the disaster recovery device 600 and the second disaster recovery device provide the same services and the same logical unit number LUN to the outside, so that the client accesses the disaster recovery device 600 or the second disaster recovery device through the LUN, and in the event of a failure of the disaster recovery device 600 or the second disaster recovery device, the client's access is switched to the surviving disaster recovery device through the LUN.

[0198] In some possible implementations, the synchronous replication module 601 is specifically used to synchronously replicate the backup data and the data generated according to the client's access to the disaster recovery device 600 to the second disaster recovery device through the first synchronous replication link.

[0199] The synchronous replication module 601 and the asynchronous replication module 602 can be implemented by software or hardware. For example, the implementation of the synchronous replication module 601 is described below by taking the synchronous replication module 601 as an example. Similarly, the implementation of the asynchronous replication module 602 can refer to the implementation of the synchronous replication module 601.

[0200] As an example of a software functional unit, the synchronous replication module 601 may include code running on a computing instance. Among them, the computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Further, the above-mentioned computing instance may be one or more. For example, the synchronous replication module 601 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same region (region) or in different regions. Furthermore, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone (AZ) or in different AZs, each AZ including one data center or multiple data centers with close geographical locations. Among them, usually a region may include multiple AZs.

[0201] Similarly, multiple hosts / virtual machines / containers used to run the code can be distributed in the same virtual private cloud (VPC) or in multiple VPCs. Usually, a VPC is set up in a region. For cross-region communication between two VPCs in the same region and between VPCs in different regions, a communication gateway needs to be set up in each VPC to achieve interconnection between VPCs through the communication gateway.

[0202] As an example of a hardware functional unit, the synchronous replication module 601 may include at least one computing device, such as a server, etc. Alternatively, the synchronous replication module 601 may also be a device implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The PLD may be a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0203] The multiple computing devices included in the synchronous replication module 601 can be distributed in the same region or in different regions. The multiple computing devices included in the synchronous replication module 601 can be distributed in the same AZ or in different AZs. Similarly, the multiple computing devices included in the synchronous replication module 601 can be distributed in the same VPC or in multiple VPCs. The multiple computing devices can be any combination of computing devices such as servers, ASICs, PLDs, CPLDs, FPGAs, and GALs.

[0204] It should be noted that, in other embodiments, the synchronous replication module 601 can be used to execute any step in the disaster recovery method, and the asynchronous replication module 602 can be used to execute any step in the disaster recovery method. The steps that the synchronous replication module 601 and the asynchronous replication module 602 are responsible for implementing can be specified as needed. The full functions of the disaster recovery device 600 are realized by respectively implementing different steps in the disaster recovery method through the synchronous replication module 601 and the asynchronous replication module 602.

[0205] According to the disaster recovery method provided in the present application, the present application also provides a disaster recovery system, which includes the aforementioned disaster recovery device 600.

[0206] See also Figure 7 , Figure 7 This is a schematic diagram of the structure of a storage server provided by this application. Figure 7 As shown, the storage server 700 includes: a bus 701, a processor 702, a memory 703 and a communication interface 704. The processor 702, the memory 703 and the communication interface 704 communicate with each other through the bus 701. It should be understood that the present application does not limit the number of processors and memories in the storage server 700.

[0207] The bus 701 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 The bus 701 is represented by only one line, but it does not mean that there is only one bus or one type of bus. The bus 701 may include a path for transmitting information between various components of the storage server 700 (for example, the memory 703, the processor 702, and the communication interface 704).

[0208] The processor 702 may include any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0209] The memory 703 may include a volatile memory, such as a random access memory (RAM). The memory 703 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0210] The memory 703 stores executable program codes, and the processor 702 executes the executable program codes to respectively implement the functions of the aforementioned synchronous replication module 601 and the asynchronous replication module 602, thereby implementing the disaster recovery method. That is, the memory 703 stores instructions for executing the disaster recovery method.

[0211] The communication interface 704 uses a transceiver unit such as, but not limited to, a network interface card or a transceiver to implement communication between the storage server 700 and other devices or a communication network.

[0212] According to the disaster recovery method provided in the present application, the present application also provides a disaster recovery system, which includes the aforementioned storage server 700.

[0213] The present application also provides a computer program product including instructions. The computer program product may be software or a program product including instructions that can be run on a computing device or stored in any available medium. When the computer program product is run on at least one computing device, the at least one computing device executes a disaster recovery method.

[0214] The present application also provides a computer-readable storage medium. The computer-readable storage medium may be any available medium that can be stored by a computing device or a data storage device such as a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk). The computer-readable storage medium includes instructions that instruct the computing device to execute a disaster recovery method.

[0215] It should be understood that in the embodiments of the present invention, "when..." and "when..." both mean that the device will perform corresponding processing under certain objective circumstances, and it does not limit the time, nor does it require the device to perform a judgment action when it is implemented, nor does it mean that there are other limitations.

[0216] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A disaster recovery system, characterized in that: include: a first data center, a second data center, a third data center, and a fourth data center, wherein the first data center and the second data center are both production centers, the third data center and the fourth data center are both disaster recovery centers, the first data center and the second data center are located in a first area, and the third data center and the fourth data center are located in a second area different from the first area, There is a first synchronous replication link between the first data center and the second data center; There is a second synchronous replication link between the third data center and the fourth data center; There is a first asynchronous replication link between the first data center and the third data center; Furthermore, the disaster recovery system further includes one, two or three of the following backup asynchronous replication links: A second asynchronous replication link is provided between the first data center and the fourth data center; A third asynchronous replication link is provided between the second data center and the third data center; A fourth asynchronous replication link is provided between the second data center and the fourth data center.

2. The disaster recovery system according to claim 1, characterized in that: The synchronous replication link is used to synchronously replicate the backup data of the data centers at both ends of the synchronous replication link; the asynchronous replication link is used to asynchronously replicate the backup data of the data centers at both ends of the asynchronous replication link.

3. The disaster recovery system according to claim 1 or 2, characterized in that: In the event that a failure occurs in the first data center or the third data center, the backup asynchronous replication link is used to asynchronously replicate the backup data between the first area and the second area.

4. The disaster recovery system according to any one of claims 1 to 3, characterized in that: The plurality of backup asynchronous replication links serve as backup links for each other.

5. The disaster recovery system according to any one of claims 2 to 4, characterized in that: The backup data is a snapshot or a log.

6. The disaster recovery system according to any one of claims 1 to 5, characterized in that: The first data center and the second data center are active-active data centers. The first data center and the second data center are used to provide the same services and the same logical unit number LUN to the outside world, so that the client can access the first data center or the second data center through the LUN, and in the event of a failure of the first data center or the second data center, the client's access is switched to the surviving data center through the LUN.

7. The disaster recovery system according to any one of claims 1 to 6, characterized in that: The third data center and the fourth data center are active-active data centers, and the third data center and the fourth data center are used to provide the same services and the same logical unit number LUN to the outside world, so that the client can access the third data center or the fourth data center through the LUN, and in the event of a failure of the third data center or the fourth data center, the client's access is switched to a surviving data center through the LUN.

8. A disaster recovery method, characterized in that: include: The first data center synchronously replicates backup data to the second data center through a first synchronous replication link, wherein the first data center and the second data center are both production centers, and the first data center and the second data center are located in a first region; When the third data center is normal, the first data center asynchronously replicates the backup data to the third data center through a first asynchronous replication link, wherein the third data center is a disaster recovery center and is located in a second area different from the first area; In the event of a failure of the third data center, the first data center asynchronously replicates the backup data to a fourth data center via a second asynchronous replication link, wherein the fourth data center is a disaster recovery center and is located in the second region.

9. The method according to claim 8, characterized in that The method further comprises: In the event of a failure of the first data center, the second data center asynchronously replicates the backup data to the third data center via a third asynchronous replication link, or the second data center asynchronously replicates the backup data to the fourth data center via a fourth asynchronous replication link.

10. The method according to claim 8 or 9, characterized in that: The backup data is a snapshot or a log.

11. The method according to any one of claims 8 to 10, characterized in that: The first data center and the second data center are active-active data centers. The first data center and the second data center provide the same services and the same logical unit number LUN to the outside world, so that the client accesses the first data center or the second data center through the LUN, and in the event of a failure of the first data center or the second data center, the client's access is switched to the surviving data center through the LUN.

12. The method according to claim 11, characterized in that The first data center synchronously replicates the backup data to the second data center through the first synchronous replication link, including: The first data center synchronously replicates the backup data and data generated according to the client's access to the first data center to the second data center through the first synchronous replication link.

13. The method according to any one of claims 8 to 12, characterized in that: The third data center and the fourth data center are active-active data centers. The third data center and the fourth data center provide the same services and the same logical unit number LUN to the outside world, so that the client accesses the third data center or the fourth data center through the LUN, and in the event of a failure of the third data center or the fourth data center, the client's access is switched to a surviving data center through the LUN.

14. A disaster recovery device, characterized in that: Used as a first disaster recovery device, including: a synchronous replication module and an asynchronous replication module, The synchronous replication module is used to synchronously replicate the backup data to the second disaster recovery device through the first synchronous replication link, wherein the first disaster recovery device and the second disaster recovery device are both production centers, and the first disaster recovery device and the second disaster recovery device are located in the first area; The asynchronous replication module is used to asynchronously replicate the backup data to the third disaster recovery device through the first asynchronous replication link when the third disaster recovery device is normal, wherein the third disaster recovery device is a disaster recovery center and is located in a second area different from the first area; The asynchronous replication module is also used to asynchronously replicate the backup data to a fourth disaster recovery device through a second asynchronous replication link when the third disaster recovery device fails, wherein the fourth disaster recovery device is a disaster recovery center and is located in the second area.

15. A storage server, characterized in that: It includes a processor and a memory, the memory is used to store instructions, and the processor is used to execute the instructions. When the processor executes the instructions, the disaster recovery method according to any one of claims 8 to 13 is implemented.

16. A computer program product comprising instructions, characterized in that When the instructions are executed by a computing device, the computing device executes the disaster recovery method according to any one of claims 8 to 13.

17. A computer-readable storage medium, characterized in that: The method comprises computer program instructions. When the computer program instructions are executed by a computing device, the computing device executes the disaster recovery method according to any one of claims 8 to 13.

Images