A method and related device for improving storage system reliability

By realizing dual write and backup controller groups of target controllers and their mirror controllers in a distributed storage system, the business interruption caused by storage control device failure is solved, and the reliability and availability of the system are improved.

CN112543922BActive Publication Date: 2024-11-15HUAWEI TECH CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN201980005665.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-07-22
Publication Date
2024-11-15
Estimated Expiration
2039-07-22

AI Technical Summary

Technical Problem

When the existing distributed storage system fails, the host service interruption will result in low storage system reliability.

Method used

By double writing data between the target controller and its mirror controller, and establishing a backup controller group between the control devices, ensuring that data is still accessible when any control device fails, and the host service is not interrupted.

Benefits of technology

Improves the reliability and availability of the storage system, ensuring that the continuity of host services can be maintained in the event of control device failure.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112543922B_ABST
    Figure CN112543922B_ABST
Patent Text Reader

Abstract

A method and related device for improving the reliability of a storage system. The storage system includes a first control device and a second control device, and the method includes: a target controller receives a write request (S401), the write request includes data to be written, and the target controller belongs to the first control device; the target controller writes the data to be written into the memory of the target controller (S420); the target controller writes the data to be written into the memory of a mirror controller of the target controller (S430), and at least one of the mirror controllers belongs to the second control device. Based on this method, the data to be written can be written and stored across control devices. When a control device fails, the host can still access other control devices to ensure that the business is not interrupted, effectively improving the reliability and availability of the storage system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of computer distributed storage systems, and in particular to a method and related devices for improving the reliability of a storage system. Background Art

[0002] Distributed data storage architecture is increasingly used in current cloud computing and big data platforms. Distributed storage technology means that a storage device includes multiple storage controllers. When writing data to a storage device, data can be written through multiple storage controllers.

[0003] When writing data, the distributed storage system generally disperses the data and writes it to the memory of different controllers. At the same time, multiple copies of the data written to each controller's memory are configured to prevent data from being inaccessible or lost due to the failure of a controller. When a controller fails, the data can still be accessed due to the existence of copies.

[0004] At present, distributed storage systems all adopt the method of separating storage control devices from storage. If the storage controller storing the target data and the storage controller configured with the target data copy fail at the same time, or the storage control device corresponding to the storage controller storing the target data fails and becomes inoperable, the target data will be inaccessible and the data in the storage connected to the storage control device will also be inaccessible, thus causing the host service to be interrupted and the reliability of the storage system to be reduced.

[0005] Therefore, how to ensure that host services are not interrupted in the event of a storage control device failure and improve the reliability of the storage system is a technical problem that needs to be solved urgently. Summary of the invention

[0006] The present application provides a method and related apparatus for improving the reliability of a storage system, which can ensure that host services are not interrupted when a storage control device fails, thereby effectively improving the reliability and availability of the storage system.

[0007] In a first aspect, a method for improving the reliability of a storage system is provided, wherein the storage system includes a first control device and a second control device, and the method includes: a target controller receives a write request, the write request includes data to be written, and the target controller belongs to the first control device; the target controller writes the data to be written into the memory of the target controller; the target controller writes the data to be written into the memory of a mirror controller of the target controller, and at least one of the mirror controllers belongs to the second control device.

[0008] Optionally, after writing the data to be written into the memory of the target controller and the mirror controller of the target controller, the target controller sends a write completion message to the host, where the write completion message is used to indicate that the data to be written has been successfully written.

[0009] In an embodiment of the present application, when writing the data to be written carried in the write request, it is necessary not only to write it into the memory of the target controller, but also to write it into the memory of the mirror controller of the target controller. In addition, there is a controller in the mirror controller that is not the same control device as the target controller. This ensures that the data to be written can be written to different control devices, and further, in the event of a failure in any control device, the host can still access the data to be written through other control devices, thereby ensuring the continuity of the host business and improving the reliability and availability of the storage system.

[0010] In combination with the first aspect, in a possible implementation manner, the first control device and the second control device are connected to the storage device, and when the first control device fails, the host accesses the storage device through the second control device.

[0011] In an embodiment of the present application, the first control device and the second control device can access the storage device at the same time, which can ensure that when any one of the control devices fails, the host can still access the storage device through other control devices, thereby ensuring the continuity of the host business and improving the reliability of the storage system.

[0012] In combination with the first aspect, in a possible implementation, the target controller and the mirror controller of the target controller constitute a backup controller group, the backup controller group includes N controllers, N is an integer greater than or equal to 2, when M controllers in the backup controller group fail, and the M controllers include the target controller, the main controller selects a controller from the mirror controllers of the target controller as a new target controller, the new target controller is used to receive the write request, the main controller is a controller selected from the controllers in the first control device and the second control device for managing other controllers; the main controller establishes M new mirror controllers for the new target controller, and mirrors the data in the memory of the new target controller to the M new mirror controllers, M is an integer greater than or equal to 1 and less than N.

[0013] In an embodiment of the present application, when multiple controllers including a target controller fail, the main controller can select a new target controller from the mirror controllers of the target controller and establish a new mirror controller for it, ensuring that when multiple controllers fail, the host can still access the target data, ensuring that the host business is not interrupted, and improving the reliability of the storage system.

[0014] In combination with the first aspect, in a possible implementation, the target controller and the mirror controller of the target controller constitute a backup controller group, and the backup controller group includes N controllers, where N is an integer greater than or equal to 2. When M controllers in the backup controller group fail and the M controllers do not include the target controller, the main controller establishes M new mirror controllers for the target controller and mirrors the data in the memory of the target controller to the M new mirror controllers. The main controller is a controller selected from the controllers in the first control device and the second control device to manage other controllers, and M is an integer greater than or equal to 1 and less than N.

[0015] In an embodiment of the present application, when multiple controllers excluding the target controller fail, the main controller can establish a new mirror controller for the target controller to ensure that the target controller can maintain multiple mirror controllers, and when multiple controllers fail, the host can still access the target data, ensuring that the host business is not interrupted and improving the reliability of the storage system.

[0016] In combination with the first aspect, in a possible implementation manner, N is equal to 3.

[0017] In combination with the first aspect, in one possible implementation, when the first control device fails, one of the mirror controllers of the target controller in the second control device is used as a new target controller, and a mirror controller is established for the new target controller, and the data in the memory of the new target controller is mirrored to the memory of the new mirror controller.

[0018] In an embodiment of the present application, the controller in each control device has a mirror controller in other control devices. In the event of a failure in the control device, the main controller upgrades the mirror controller of the target controller in the other control device to a new target controller and establishes a mirror controller for it. This ensures that when a control device fails, the host can still access the target data, thereby ensuring the continuity of the host business and improving the reliability and availability of the storage system.

[0019] In a second aspect, a storage control system is provided, which includes a first control device and a second control device, and is characterized in that the first control device includes a target controller, and the target controller includes: a receiving module for receiving a write request, the write request including data to be written; a processing module for writing the data to be written into the memory of the target controller; the processing module is also used to write the data to be written into the memory of a mirror controller of the target controller, and at least one of the mirror controllers belongs to the second control device.

[0020] In combination with the second aspect, in a possible implementation, the target controller and the mirror controller of the target controller constitute a backup controller group, the backup controller group includes N controllers, N is an integer greater than or equal to 2, the storage control system includes a main controller, the main controller is a controller selected from the controllers in the first control device and the second control device for managing other controllers, the main controller includes: a fault handling module, which is used when M controllers in the backup controller group fail and the M controllers include the target controller, the main controller selects a controller from the mirror controllers of the target controller as a new target controller, and the new target controller is used to receive the write request; a mirroring module, which is used to establish M new mirror controllers for the new target controller, and mirror the data in the memory of the new target controller to the M new mirror controllers, M is an integer greater than or equal to 1 and less than N.

[0021] In combination with the second aspect, in a possible implementation, the target controller and the mirror controller of the target controller constitute a backup controller group, the backup controller group includes N controllers, N is an integer greater than or equal to 2, the storage control system also includes a main controller, the main controller is a controller selected from the controllers in the first control device and the second control device for managing other controllers, the main controller includes: a fault handling module, when M controllers in the backup controller group fail and the M controllers do not include the target controller, the main controller establishes M new mirror controllers for the target controller; a mirroring module, used to mirror the data in the memory of the target controller to the M new mirror controllers, M is an integer greater than or equal to 1 and less than N.

[0022] In combination with the second aspect, in a possible implementation manner, N is equal to 3.

[0023] In combination with the second aspect, in one possible implementation, when the first control device fails, one of the mirror controllers of the target controller in the second control device is used as a new target controller, and a mirror controller is established for the new target controller, and the data in the memory of the new target controller is mirrored to the memory of the new mirror controller.

[0024] According to a third aspect, a storage control system is provided, which includes a first control device and a second control device, and is characterized in that the first control device includes a target controller, the target controller includes a processor and a memory, and the processor executes computer instructions stored in the memory, so that the target controller executes the method provided in the first aspect and in combination with any one of the implementations of the first aspect.

[0025] In combination with the third aspect, in a possible implementation, the storage control system includes a main controller, which is a controller selected from the controllers in the first control device and the second control device to manage other controllers, and the main controller includes a processor and a memory, and the processor executes computer instructions stored in the memory so that the main controller executes the above-mentioned first aspect and the method provided in combination with any one of the implementations of the above-mentioned first aspect.

[0026] In a fourth aspect, a computer storage medium is provided, wherein the computer storage medium stores a computer program, and when the computer program is executed by a processor, the process of the method provided in the first aspect and in combination with any one of the implementation methods in the first aspect can be implemented.

[0027] In a fifth aspect, a computer program product is provided, which includes computer instructions. When the computer program is executed by a computer, the computer can execute the process of the method provided in the above-mentioned first aspect and in combination with any one of the implementation methods of the above-mentioned first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1A A schematic diagram of a series connection model provided in an embodiment of the present application;

[0029] Figure 1B A schematic diagram of a parallel model provided in an embodiment of the present application;

[0030] Figure 2A A schematic diagram of the structure of a storage system provided in an embodiment of the present application;

[0031] Figure 2B A schematic diagram of dual-copy data storage provided in an embodiment of the present application;

[0032] Figure 3 A schematic diagram of an application scenario provided for an embodiment of the present application;

[0033] Figure 4 A flowchart of a method for improving the reliability of a storage system provided in an embodiment of the present application;

[0034] Figure 5 A schematic diagram of writing multiple copies of data provided in an embodiment of the present application;

[0035] Fig. 6A A schematic diagram of a dual controller failure at the same time provided in an embodiment of the present application;

[0036] Figure 6B A schematic diagram of a control device failure provided in an embodiment of the present application;

[0037] Figure 6C A schematic diagram of a controller failure in sequence provided in an embodiment of the present application;

[0038] Figure 7 A schematic diagram of a reliability model of a control device provided in an embodiment of the present application;

[0039] Figure 8 A schematic diagram of a reliability model of a storage system provided in an embodiment of the present application;

[0040] Fig. 9 A schematic diagram of the structure of a storage control system provided in an embodiment of the present application;

[0041] Fig.10 A schematic diagram of the structure of another storage control system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application are described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0043] First, some of the terms and related technologies involved in this application are explained in conjunction with the accompanying drawings to facilitate understanding by those skilled in the art.

[0044] The high-speed serial computer expansion bus standard (Peripheral Component Interconnect Express, PCIe) is a high-speed serial point-to-point dual-channel high-bandwidth transmission. The connected devices are allocated exclusive channel bandwidth and do not share bus bandwidth. The data transmission rate is high. The connection between any two PCIe devices is called a link, which is constructed by a collection of one or more channels.

[0045] Small Computer System Interface (SCSI) is an independent processor standard for system-level interfaces between computers and smart devices (such as hard disks, optical drives, printers, scanners, etc.). The SCSI interface is a universal interface that can connect host adapters and SCSI peripheral controllers on the SCSI bus. Multiple peripherals hanging on a SCSI bus can work simultaneously. The SCSI interface can transmit data synchronously or asynchronously. The SCSI interface controls the read and write operations of data through an independent high-speed SCSI card. The central processing unit (CPU) does not need to waste time waiting, which can improve the overall performance of the system.

[0046] Serial attached SCSI (SAS) is a new generation of SCSI technology. It uses serial technology to achieve higher transmission speeds, improves internal space by shortening the link, and provides compatibility with serial advanced technology attachment (SATA) hard disks. SAS is a point-to-point structure that can establish a direct connection from the disk to the controller. In addition, the SAS structure has very good expansion capabilities and can connect up to 16,384 disk devices. Each SAS cable has 4 cables, 2 inputs and 2 outputs, which can read and write data simultaneously.

[0047] Non-volatile memory express (NVMe) is a logical device interface specification, a protocol similar to the serialATA advanced host controller interface (AHCI) designed specifically for flash storage. Its essence is to establish multiple paths between computers and storage devices. Multiple paths are actually multiple queues. The NVMe protocol can have up to 64k queues, which greatly improves concurrency and performance.

[0048] Direct memory access (DMA) is a mode of operation in which I / O exchanges are completely performed by hardware. In this mode, the DMA controller completely takes over the control of the bus from the CPU, and data exchange is performed directly between the memory and the IO device without passing through the CPU. Using DMA can reduce the CPU overhead during large-scale data transmission. The DMA controller is used to generate memory access addresses and control the memory access process. The transmission speed is fast, and the CPU basically does not intervene, only participating in initialization and termination. The CPU and peripherals work in parallel, which is more efficient.

[0049] Remote direct memory access (RDMA) is a technology that bypasses the kernel of the remote host operating system to access data in its memory. Since RDMA quickly moves data from one system to the remote system memory over the network without going through the operating system, it can save a lot of CPU resources, improve system throughput, eliminate the overhead of external memory copying and context switching, and free up memory bandwidth to improve application system performance.

[0050] Data copies refer to redundant data storage, that is, the same data is stored on different storage devices to improve system availability. When the storage device storing the primary copy data fails, other storage devices storing the copy data can be accessed to obtain the data to ensure uninterrupted business.

[0051] Failure rate, also known as failure rate, is the frequency of failure of an engineering system or part. The unit is usually the number of failures per hour, represented by the letter λ. The mean time between failures (MTBF) refers to the average time from the current failure to the next failure of a repairable device, λ = 1 / MTBF. The mean time to repair (MTTR) refers to the average repair time when the equipment is converted from a faulty state to a working state. Availability (A) refers to the degree to which a system, subsystem or equipment is in a specified operable or submittable state when starting a task. In other words, availability is the proportion of time a system is in an operable state, A = MTBF / (MTBF+MTTR). If you calculate the availability of a system, you can calculate the MTBF of a single unit based on the failure rate of the single unit, thereby calculating the availability of the single unit, and then calculate the overall availability of the system based on the series and parallel models. See. Figure 1A , is a schematic diagram of a series model. Unit 1, unit 2, ..., unit N are connected in series, the availability of unit 1 is A1, the availability of unit 2 is A2, ..., the availability of unit N is An, then the availability of the entire system A = A1*A2*... *An. See Figure 1B , is a schematic diagram of a parallel model. The parallel model is more complicated than the series model. The redundant units of the parallel model are divided into primary and standby redundant units and load sharing redundant units. Figure 1B Unit 1, unit 2, ..., unit N are primary and standby redundant units, and the availability of each unit is a. Then the availability of the entire system is A = 1-{C*{1-a N *(N+1-N*a)} / N+(1-C)*(1-aN ) / N}. Where C represents the switching rate, that is, the probability of successfully switching from the main unit to the backup unit, C = C1*C2, C1 represents the fault detection rate of the main unit, and C2 represents the fault detection rate of the backup unit. In particular, if each unit has only one redundant unit, that is, N = 1, then the formula can be simplified to A = a + C*a*(1-a). If Figure 1B Unit 1, unit 2, ..., unit N are load sharing redundant units, and the availability of each unit is a. Then the availability of the entire system is A = 1-{C A *{1-a N *(N+1-N*a)} / N+(1-C A )*(1-a N+1 ) / (N+1)}. Among them, C A represents the fault detection rate of each unit.

[0052] In general, in order to improve the scalability of the system, enterprise storage systems usually use interconnection to connect multiple control storage nodes to form a unified distributed system, and then provide storage services to applications. For details, see Figure 2A , Figure 2A It is a schematic diagram of the structure of a storage system. Figure 2AAs shown, device 210 is a device that uses storage services, such as a host, a server, etc. Device 210 is connected to control device 220 and control device 230. Control device 220 includes multiple controllers. In this embodiment, four controllers are used as an example for explanation. These four controllers are controller 221, controller 222, controller 223 and controller 224, which are interconnected through a PCIe interface; control device 230 also includes multiple controllers. In this embodiment, four controllers are also used as an example for explanation. These four controllers are controller 231, controller 232, controller 233 and controller 234, which are also interconnected through a PCIe interface. In order to improve system reliability, dual-link redundant connections are used between each controller. The control device 220 and the control device 230 are also connected via dual-link redundant connection via the PCIe interface. The control device 220 and the control device 230 can be storage arrays, and the controllers contained therein can be a structure of a central processing unit CPU plus a memory, wherein the CPU calls the program in the memory to execute the corresponding function, or can be an independent control chip, such as a field-programmable gate array (FPGA), in which the program is burned into the FPGA. The control device 220 is connected to the memory 240 via a SAS interface, and the control device 230 is connected to the memory 250 via a SAS interface. In order to improve system reliability, dual-link redundant connection is adopted between the control device 220 and the memory 240, and between the control device 230 and the memory 250. The memory 240 and the memory 250 are both devices for storing data, such as storage arrays. The memory 240 includes a cascade board 241, a cascade board 242, and N storage media, where N is a positive integer greater than 1. The cascade board 241 is responsible for managing the memory 240, and the cascade board 242 is a redundant cascade board of the cascade board 241. When the cascade board 241 fails, it takes over the work of the cascade board 241 to improve the reliability of the entire system. The structure of the memory 250 is consistent with that of the memory 240, and includes a cascade board 251, a cascade board 252, and N storage media.

[0053] In actual applications, the storage system uses dual memory copy technology to balance system reliability and controller performance. That is, the data written to the memory of each controller is backed up in the memory of another controller belonging to the same controller device. Figure 2BAs shown, data A is stored in controller 221, and data A1, a copy of data A, is stored in controller 222; data B is stored in controller 222, and data B1, a copy of data B, is stored in controller 221; data E is stored in controller 231, and data E1, a copy of data E, is stored in controller 232; data H is stored in controller 234, and data H1, a copy of data H, is stored in controller 233. If controller 221 fails, device 210 will not be able to access the data stored in controller 221, that is, device 210 cannot access data A and B1. Since data A1, a copy of data A, is stored in controller 222, device 210 can access data A1 by accessing controller 222, thereby ensuring that the service is not interrupted.

[0054] If controller 221 and controller 222 fail at the same time, or control device 220 fails, device 210 will not be able to access data A and data A1, which may cause service interruption.

[0055] In addition, since the controller in each control device can only access the memory connected to the control device, for example, the controller 221 in the control device 220 can only access the data in the memory 240, but cannot access the data in the memory 250, because the control device 220 is not connected to the memory 250; similarly, the controller 231 in the control device 230 can only access the data in the memory 250, but cannot access the data in the memory 240. Therefore, when any control device fails due to power failure or hardware failure, the data on the controller in the control device and the data on the memory connected to the control device will not be accessible, resulting in service interruption or failure on the host. For example, if the control device 220 fails, then the device 210 will not be able to access the data in the control device 220 and the memory 240, resulting in service interruption or failure on the device 210.

[0056] It should be noted that the control device and the storage can be connected via a SAS link or an NVMe link, and each control device can also be connected to multiple storage devices. In addition, the control device and the storage can be deployed on the same physical device. For example, the above-mentioned storage 240 can be deployed on the same physical device as the control device 220. Of course, they can also be deployed separately on different physical devices.

[0057] In order to solve the above problems, the present application provides a method and related devices for improving the reliability of a storage system, thereby ensuring the continuity of host services when a control device fails, and improving the reliability and availability of the storage system.

[0058] See also Figure 3 , Figure 3A possible application scenario of the embodiment of the present application is shown. In this application scenario, the host 310 is connected to the control device 320 and the control device 330. The host 310 can be connected to the control device 320 and the control device 330 through an optical fiber, or through a network, such as Ethernet. In addition, there are two switch cards ( Figure 3 (not shown), one of which is a main switching card and the other is a backup. All controllers in the control device 320 and the control device 330 need to be connected to the main switching card, which is used to support the controllers in the control device to communicate with the controllers in other control devices, and to communicate with the memory 340. When the main switching card fails, the backup switching card takes over the work of the main switching card to ensure the reliability of the system. The control device 320 includes a plurality of controllers, and four are used as an example in this embodiment, such as controller 321, controller 322, controller 323 and controller 324. The control device 330 also includes a plurality of controllers, and four are used as an example in this embodiment, such as controller 331, controller 332, controller 333 and controller 334. The controllers in the control device 320 and the control device 330, as well as the control device 320 and the control device 330 are interconnected through RDMA channels. The data between the controllers can be directly transmitted to the opposite end without passing through the CPU. Of course, they can also be connected to each other through the PCIe interface. The control device 320 and the control device 330 are connected to the memory 340 respectively by using the RDMA channel, that is, the memory 340 is shared by the control device 320 and the control device 330, that is, the control device 320 and the control device 330 can access the data in the memory 340 at the same time, and the memory 340 can be a storage server (i.e., a storage node) including a CPU 341 and a memory 342, which includes N storage media for storing data for controller access, and N is a positive integer greater than 1. The memory 342 of the memory 340 stores program instructions, and the CPU 341 calls the program instructions to implement the control of the memory 340, such as performing relevant configuration on the memory 340, implementing the upgrade of the memory 340, etc. The controllers in all control devices can be used to receive IO sent by the host, and determine the target controller to process the IO based on the address carried in the IO. If the target controller is the controller that receives the IO sent by the host, the target controller directly processes the IO. If the target controller is not the controller that receives the IO sent by the host, the controller forwards the IO to the target controller for processing.

[0059] It should be noted that, in order to improve system reliability, all connections are dual-link redundancy, that is, two RDMA channels are used to connect between controllers, between control devices, and between control devices and memories. In addition, the host 310 can also be connected to more control devices, and the control device can also be connected to multiple shared memories. In addition, each controller includes at least one mirror controller, and at least one mirror controller belongs to another control device, that is, the data stored in each controller is backed up in other controllers, and at least one controller belongs to other control devices. Exemplarily, the mirror controller of controller 321 includes controller 322 and controller 331. If the data A stored in controller 321 is backed up, then controller 322 and controller 331 store the copy data A1 and copy data A2 of data A accordingly. Controller 321 and controller 322 belong to control device 320, and controller 331 belongs to the second control device 330. Data A is equivalent to being stored across control devices. The mirror controller of each controller is pre-set, and each controller and its corresponding mirror controller constitute a backup controller group.

[0060] It is easy to understand that when a controller in a control device fails or a control device fails, the host can still access the backup data in other control devices because the data is backed up across control devices, thereby ensuring business continuity. In addition, multiple control devices share a memory, so even if a control device fails, the host can still access the data in the memory through the controller in other control devices, which can ensure that the business on the host will not be interrupted or fail, effectively improving the reliability and availability of the system.

[0061] Combination Figure 3 The application scenario shown below will be combined with Figure 4 The method for improving the reliability of a storage system provided by an embodiment of the present application is described as follows: Figure 4 As shown, the method includes but is not limited to the following steps:

[0062] S410: The target controller receives a write request.

[0063] Specifically, the target controller may be any controller in any control device connected to the host. Figure 3 , the target controller may be the controller 321 in the control device 320. In particular, when the target controller is the controller that currently receives the IO sent by the host, the target controller directly receives the write request sent by the host, and when the target controller is not the controller that currently receives the IO sent by the host, the target controller receives the write request forwarded by the controller.

[0064] Furthermore, the write request includes the data to be written that the host needs to write.

[0065] S420: The target controller writes the data to be written into the memory of the target controller.

[0066] Specifically, the target controller may be a control chip including a CPU and a memory. After receiving a write request, the target controller parses the write request to obtain the data to be written and writes the data to be written into the memory of the target controller.

[0067] S430: The target controller writes the data to be written into the memory of the mirror controller of the target controller.

[0068] like Figure 3 In the embodiment of the present invention, each controller is provided with a mirror controller, and at least one of the mirror controllers does not belong to the same control device as the target controller. Therefore, after the target controller writes the data to be written into the memory, it also writes the data to be written into the memory of the mirror controller. Since the controllers are connected via RDMA, the target controller can use RDMA to write the data to be written directly into the memory of the mirror controller of the target controller.

[0069] It should be understood that data is transmitted between any two controllers through the RDMA channel, and the data does not pass through the CPU, thereby improving the transmission efficiency and the CPU performance of the controller.

[0070] For example, see Figure 5 , Figure 5 This is a schematic diagram of writing multiple copies of data provided by an embodiment of the present application. Figure 5As shown, the mirror controllers of controller 521 are controller 522 and controller 532, so controller 521 first writes data A into the memory, and then writes the copies of data A1 and A2 into the memory of controller 522 and controller 532 respectively; the mirror controllers of controller 523 are controller 524 and controller 534, and after controller 523 writes data B into the memory, it writes the copy of data B1 into the memory of controller 524, and writes the copy of data B2 into the memory of controller 534; similarly, The mirror controllers of controller 531 are controller 522 and controller 532. After controller 531 writes data C into the memory, it writes a copy of data C1 into the memory of controller 532, and writes a copy of data C2 into the memory of controller 522. The mirror controllers of controller 533 are controller 524 and controller 534. After controller 533 writes data D into the memory, it writes a copy of data D1 into the memory of controller 534, and writes a copy of data D2 into the memory of controller 524.

[0071] In particular, there is a master controller among all interconnected controllers. The master controller can be any one selected from all controllers to manage the mirror relationship of all controllers, that is, each controller has mirror controllers, and other controllers have backups of the mirror relationship. When the master controller fails to work, a new master controller will be elected from the remaining controllers to take over the work of the original master controller and manage the mirror relationship of all controllers. In addition, in the mirror relationship managed by the master controller, it will avoid setting the mirror controllers of too many controllers to the same controller. The master controller can use a scattering algorithm to set the mirror relationship to achieve load balancing between the controllers.

[0072] For example, see above Figure 5 , controller 521 is the target controller, controller 522 and controller 532 are mirror controllers of the target controller, and after controller 521 writes data A into the memory, it writes copies of data A1 and A2 into the memory of controller 522 and controller 532 respectively.

[0073] S440: The target controller sends a write completion message to the host.

[0074] Specifically, after the target controller writes the data to be written into the memory of the target controller and the mirror controller of the target controller, it sends a write completion message to the host to notify the host that the data to be written has been successfully written into the memory of the target controller and the mirror controller of the target controller.

[0075] In a possible implementation, a target controller and a mirror controller of the target controller constitute a backup controller group, and the backup controller group includes N controllers, where N is an integer greater than or equal to 2. When M controllers in the backup controller group fail, and the M controllers include the target controller, the main controller selects one controller from the mirror controllers of the target controller as a new target controller, and the new target controller is used to receive a write request, and the main controller is used to receive a write request from a host and forward the write request to the new target controller; the main controller establishes M new mirror controllers for the new target controller, and mirrors data in a memory in the new target controller to the M new mirror controllers, where M is an integer greater than or equal to 1 and less than N.

[0076] Specifically, all controllers in the control devices periodically send heartbeat messages to each other to sense each other's status. If a controller does not receive a heartbeat message sent by another controller for more than a preset time, it can be determined that the controller has failed. At this time, the main control needs to readjust the mirror relationship of each controller. The heartbeat period and the preset time can be set according to actual needs, and this application does not limit this. For example, in the above Figure 5 In the embodiment, the eight controllers in the control device 520 and the control device 530 send heartbeat messages to each other. If the controller 521 does not receive the heartbeat message sent by the controller 522 within a preset time, the controller 521 can determine that the controller 522 fails.

[0077] Optionally, the main controller can set the priority of the mirror controller of the managed controller, that is, when the target controller fails, the mirror controller with a higher priority is selected from the mirror controllers of the target controller as the new target controller according to the priority. Optionally, the main controller can set the mirror controller in the same control device as the target controller as the high-priority controller. Of course, it can also be set according to other conditions, such as the physical state of the controller, the backup power relationship, etc., which is not limited in this application. For example, in Figure 5 In the embodiment, controller 522 and controller 532 are both mirror controllers of controller 521, but controller 522 has a higher priority than controller 532. Therefore, when controller 521 fails, the main controller will select controller 522 as the new target controller. Alternatively, the main controller randomly selects a controller from the mirror controllers of the target controller as the new target controller. The present application does not limit the method used by the main controller to determine the new target controller.

[0078] Furthermore, when M controllers including the target controller fail, the main controller will determine a new target controller from the remaining mirror controllers, and establish M new mirror controllers for the new target controller, ensuring that the number of controllers in the backup controller group remains the same as before (i.e., N controllers), and mirror the data in the new target controller to the M new mirror controllers. In addition, the host controller can receive a write request from the host, and then forward the write request to the new target controller, and the new target controller writes the data to be written in the write request into the memory of the new target controller and the mirror controller of the new target controller (including the M new mirror controllers).

[0079] Optionally, the value of N is 3, that is, the target controller includes two mirror controllers, and at least one of the mirror controllers and the target controller are deployed on different control devices.

[0080] In a possible implementation, the target controller and the mirror controller of the target controller constitute a backup controller group, and the backup controller group includes N controllers, where N is an integer greater than or equal to 2. When M controllers in the backup controller group fail, and the M controllers do not include the target controller, the main controller establishes M new mirror controllers for the target controller, and mirrors the data in the memory of the target controller to the M new mirror controllers, where M is an integer greater than or equal to 1 and less than N.

[0081] Specifically, when M controllers in the backup controller group, excluding the target controller, fail, the master controller updates the global mirroring relationship, that is, the master controller directly establishes M new mirror controllers for the target controller, ensuring that the number of controllers in the backup controller group remains N, and mirrors the data in the target controller to the M new mirror controllers.

[0082] Optionally, the value of N is 3, that is, the target controller includes two mirror controllers, and at least one of the mirror controllers and the target controller are deployed on different control devices.

[0083] For example, see Fig. 6A ,like Fig. 6AAs shown, when the control device 520 and the control device 530 operate normally, the controller 521, the controller 522 and the controller 532 form a backup controller group, which respectively store data A and the replica data A1 and A2 of data A; the controller 522, the controller 531 and the controller 532 form another backup controller group, which respectively store data C and the replica data C1 and C2 of data C. When the controller 521 and the controller 522 fail, for the data A, its target controller is the controller 521. Since the controller 522 and the controller 532 store the replica data A1 and A2 of data A, when the controller 521 and the controller 522 fail at the same time, the main controller will select the controller 532 as the new target controller, which is equivalent to the data stored in the controller 532 becoming the data A. The master controller re-establishes a new mirroring relationship. After controller 532 is determined as the new target controller, the replica data A1 and A2 of data A are mirrored to controller 531 and controller 523 respectively, that is, controller 523 and controller 531 are mirroring controllers of controller 532. Similarly, for data C, its target controller is controller 531. Since controller 522 stores the replica data C2 of data C, when a controller fails, the master controller establishes a new mirroring controller for target controller 531 to establish a new mirroring relationship. The master controller mirrors the replica data of data C to controller 524, that is, controller 524 and controller 532 are mirroring controllers of target controller 531.

[0084] It should be understood that the above description is based on the example that each target controller includes only two mirror controllers. Of course, each target controller may also include more mirror controllers, and the specific implementation logic and process are consistent with the above description. For the sake of brevity, they will not be repeated here.

[0085] It can be seen that in the case of two controllers failing at the same time, since the target data (such as data A) is backed up and the copies of the target data are stored in three different controllers, the host can still access the target data, thereby ensuring the continuity of the host business and improving the reliability and availability of the entire storage system. In addition, the main controller in the control device can also re-establish the mirror relationship and establish a new target controller or mirror controller in the controller that has not failed, ensuring that the target data and the copies of the target data can still be stored in three different controllers.

[0086] In one possible implementation, when the first control device fails, one of the mirror controllers of the target controller in the second control device is used as a new target controller, and a mirror controller is established for the new target controller, and the data in the memory of the new target controller is mirrored to the memory of the new mirror controller.

[0087] Specifically, when the first control device fails, the master controller needs to re-update the global mirror relationship. Since each controller in the first control device has at least one mirror controller in the second control device, the master controller can select one from the mirror controllers in the second control device as a new target controller, and then establish a new mirror controller for the new target controller, and mirror the data in the memory of the new target controller to the memory of the new mirror controller. In the embodiment of the present invention, one of the controllers in the first control device and the controllers in the second control device can be selected as the master controller. The method for selecting the master controller in a distributed system is a prior art and will not be described in detail here.

[0088] For example, see Figure 6B ,like Figure 6B As shown, when control device 520 and control device 530 operate normally, for data A, its target controller is controller 521, controller 532 stores a copy of data A2, and controller 532 is a mirror controller of controller 521; for data B, its target controller is controller 523, controller 534 stores a copy of data B2, and controller 534 is a mirror controller of controller 523. When control device 520 fails, for data A, the main controller uses controller 532 as the new target controller, and the data stored in controller 532 becomes data A. The main controller re-establishes a new mirroring relationship and mirrors the copies of data A1 and A2 to controller 531 and controller 533 respectively; for data B, the main controller uses controller 534 as the new target controller and re-establishes a new mirroring relationship and mirrors the copies of data B1 and B2 to controller 531 and controller 533 respectively; for data C, the main controller directly establishes a new mirroring relationship and mirrors the copy of data C2 to controller 534; for data D, the main controller also directly establishes a new mirroring relationship and mirrors the copy of data D2 to controller 532.

[0089] It should be understood that the above description is only based on two control devices as an example. In the case where there are more control devices, the specific implementation logic is the same as above and will not be repeated here.

[0090] It can be seen that in the event of a control device failure, since all controllers in the control device have at least one mirrored controller in another control device, that is, all data in the control device is backed up in another control device, the host can still access the data in the failed control device by accessing other control devices, thereby ensuring the continuity of the host business and improving the reliability and availability of the storage system.

[0091] In the scenario where the first control device includes 4 controllers and the second control device includes 4 controllers, the method provided by the present invention can support failure of 7 controllers without interrupting the service. Figure 6C The embodiment shown. Figure 6CAs shown, when the control device 520 and the control device 530 operate normally, the controller 521 stores data A, and the controller 522 and the controller 532 store the replica data A1 and A2 of the data A respectively; the controller 523 stores data B, and the controller 524 and the controller 534 store the replica data B1 and B2 of the data B respectively; the controller 531 stores data C, and the controller 532 and the controller 522 store the replica data C1 and C2 of the data C respectively; the controller 533 stores data D, and the controller 534 and the controller 524 store the replica data D1 and D2 of the data D respectively. When the controller 521 fails, the main controller establishes a new mirroring relationship, uses the controller 522 as the target controller of the data A, and mirrors the replica data A1 of the data A to the third controller 523, and the mirroring relationship of the other controllers remains unchanged. Further, controller 522 also fails, the main controller uses controller 523 as the target controller for data A, and mirrors the copy data A1 of data A to controller 524; re-mirrors the copy data C2 of data C to controller 524, and the mirroring relationships of other controllers remain unchanged. Further, controller 523 also fails, the main controller uses controller 524 as the target controller for data A, and mirrors the copy data A1 of data A to controller 534; the main controller uses controller 524 as the target controller for data B, and mirrors the copy data B1 of data B to controller 532, and the mirroring relationships of other controllers remain unchanged. Further, controller 524 also fails, the main controller uses controller 534 as the target controller for data A, and mirrors the copy data A1 of data A to controller 531; the main controller uses controller 532 as the target controller for data B, and mirrors the copy data B1 of data B to the controller; the main controller re-mirrors the copy data C2 of data C to controller 534; the main controller re-mirrors the copy data D2 of data D to controller 532, and the other mirroring relationships remain unchanged. Furthermore, controller 531 also fails, and the main controller re-mirrors the replica data A1 of data A to controller 533; the main controller re-mirrors the replica data B1 of data B to controller 533; the main controller uses controller 532 as the target controller of data C, and mirrors the replica data C1 of data C to controller 533, and other mirroring relationships remain unchanged. Furthermore, controller 532 also fails. Since there are only two controllers left at this time, namely controller 533 and controller 534, each target controller in the mirroring relationship managed by the main controller at this time includes only one mirroring controller. The main controller uses controller 533 as the target controller of data B, and also uses controller 533 as the target controller of data C and data D. Since controller 534 stores the replica data B1 of data B, the replica data C1 of data C, and the replica data D1 of data D, the main controller does not need to establish a new mirroring relationship.Furthermore, controller 533 also fails, and only controller 534 remains among all controllers. All data can only be stored in controller 534, and data backup cannot be performed, that is, the main controller can no longer establish a new mirror controller, and controller 534 serves as the target controller for data A, data B, data C and data D.

[0092] It can be seen that when the controllers fail one after another, by continuously mirroring and constantly establishing and updating the mirroring relationship, it can be ensured that the host can access the target data normally. Even if only one controller is left working, it can still support the host to access the target data and ensure the continuity of the host business.

[0093] It should be noted that the above only takes the failure of each controller in sequence as an example to illustrate the change of the mirror relationship of the controller. Of course, there are also application scenarios in which multiple controllers fail at the same time and multiple controllers fail at the same time and fail in sequence. The specific logic of the change of the mirror relationship is consistent with the above. For the sake of brevity, it will not be repeated here.

[0094] In addition, in order to better illustrate the impact of this application on the availability of the storage system, a storage system including two control devices and 8 controllers is used as an example for explanation. Figure 7 , Figure 7 It is a schematic diagram of the reliability model of a control device, such as Figure 7 As shown in Figure 1, the model includes 4 parallel controllers, 2 parallel backup power units, 2 parallel power modules, 2 parallel fans, a backplane, and 2 parallel management modules, which are connected in series. Figure 2B and Figure 5 The availability of each control device in the scenario shown. It should be noted that in the above two scenarios, other devices are consistent and the connection relationship is consistent, so the availability of the control device can be calculated based on the relevant redundancy of the controller. Assuming that the availability of each controller is 0.999989496, use formula 1 to calculate Figure 2B The availability of each control device in Equation 1 is as follows:

[0095] A=(a+C*a*(1-a))*(a+C*a*(1-a)) Formula 1

[0096] Where a represents the availability of each controller, C represents the switching rate, and the calculated availability of the control device is 0.999998950.

[0097] Calculate using formula 2 Figure 5 The availability of each control device in Equation 2 is as follows:

[0098] A=1-{C*{1-a N*(N+1-N*a)} / N+(1-C)*(1-a N ) / N} Formula 2

[0099] Wherein, N represents the number of redundant controllers of each controller (that is, the number of mirror controllers), and the calculated availability of the control device is 0.999999475.

[0100] It can be seen that Figure 5 The availability of control equipment in Figure 2B The availability of the control equipment in the system is improved, and the reliability of the control equipment is higher.

[0101] For further information, see Figure 8 , Figure 8 It is a schematic diagram of the reliability model of a storage system. Figure 8 As shown in Figure 1, the model includes two parallel control devices, X serial hard disks, and Y serial disk groups, which are connected in series. Figure 8 The model shown calculates Figure 2B and Figure 5 The availability of the entire storage system should be understood. Figure 2B and Figure 5 In the scenario shown, other devices are consistent and their connections are consistent, so the availability of the entire storage system can be calculated based only on the control device. Figure 2B In the scenario shown, since each control device is independent of each other, that is, if one control device fails, the entire storage system will not work. Therefore, the control devices should be in a series model, and the calculation method is as follows: Figure 2B The availability of the storage system in the formula 3 is as follows:

[0102] A=A1*A2 Formula 3

[0103] Among them, A1 and A2 represent the availability of each control device. Because each control device is the same, the values ​​of A1 and A2 are the same, both are 0.999998950, so the availability of the entire storage system is 0.999997900.

[0104] And in Figure 5 In the scenario shown, each control device is related to each other. Even if one control device fails, the other control device can still support the host to access it, and the entire storage system can work normally. Therefore, the control devices should belong to the parallel model. Formula 4 is used to calculate Figure 5 The availability of the storage system in the formula 4 is as follows:

[0105] A=a+C*a*(1-a) Formula 4

[0106] Where a represents the availability of each control device, C represents the switching rate, and the calculated availability of the entire storage system is 0.999999475.

[0107] It can be seen that Figure 5 The availability of the storage system is much higher than Figure 2B The availability of the storage system in the storage system, that is, the implementation of the solution provided in the embodiment of the present application can effectively improve the availability of the control device and the storage system, improve the reliability of the storage system, and ensure the continuity of the host business.

[0108] The method of the embodiment of the present application is described in detail above. In order to facilitate better implementation of the above scheme of the embodiment of the present application, correspondingly, relevant devices for cooperating in implementing the above scheme are also provided below.

[0109] See also Fig. 9 , Fig. 9 This is a schematic diagram of the structure of a storage control system provided in an embodiment of the present application. Fig. 9 As shown, the storage control system 900 includes a first control device 9100 and a second control device 9200. The first control device 9100 includes a target controller 9110. The target controller 9110 includes a receiving module 9111 and a processing module 9112.

[0110] The receiving module 9111 is used to receive a write request, where the write request includes data to be written.

[0111] A processing module 9112, used for writing the data to be written into the memory of the target controller 9110;

[0112] The processing module 9112 is further used to write the data to be written into the memory of the mirror controller of the target controller 9110 , at least one controller of the mirror controller belongs to the second control device 9200 .

[0113] In a possible implementation, the target controller 9110 and the mirror controller of the target controller 9110 constitute a backup controller group, the backup controller group includes N controllers, N is an integer greater than or equal to 2, the storage control system 900 includes a main controller 9210, the main controller 9210 is a controller selected from the controllers in the first control device 9100 and the second control device 9200 for managing other controllers, the main controller 9210 includes: a fault handling module 9211, which is used to select a controller from the mirror controllers of the target controller as a new target controller when M controllers in the backup controller group fail and the M controllers include the target controller, and the new target controller is used to receive the write request; a mirror module 9212, which is used to establish M new mirror controllers for the new target controller, and mirror the data in the memory of the new target controller to the M new mirror controllers, M is an integer greater than or equal to 1 and less than N.

[0114] In a possible implementation, N is equal to 3.

[0115] In one possible implementation, when the first control device 9100 fails, one of the mirror controllers of the target controller 9110 in the second control device 9200 is used as a new target controller, and a mirror controller is established for the new target controller, and the data in the memory of the new target controller is mirrored to the memory of the new mirror controller.

[0116] It should be understood that the structures of the target controller and the main controller and the process of improving the reliability of the storage control system are only examples and should not constitute specific limitations. The modules in the target controller and the main controller can be increased, reduced or merged as needed. In addition, the operations and / or functions of the modules in the target controller and the main controller are respectively to achieve the above Figure 4 For the sake of brevity, the corresponding process of the described method will not be repeated here.

[0117] See also Fig.10 , Fig.10 This is a schematic diagram of the structure of another storage control system provided in an embodiment of the present application. Fig.10 As shown, the storage control system 100 includes a first control device 1100 and a second control device 1200, wherein the first control device 1100 includes a target controller 1110, and the target controller 1110 includes: a processor 1111, a communication interface 1112 and a memory 1113, and the processor 1111, the communication interface 1112 and the memory 1113 are connected via an internal bus 1114.

[0118] In one possible implementation, the storage control system 100 includes a main controller 1210, which is a controller selected from the controllers in the first control device 1100 and the second control device 1200 for managing other controllers. The main controller 1210 includes: a processor 1211, a communication interface 1212 and a memory 1213, and the processor 1211, the communication interface 1212 and the memory 1213 are connected via an internal bus 1214.

[0119] It should be understood that the functions implemented by the receiving module 9111 can be specifically implemented using the communication interface 1112, the functions implemented by the processing module 9112 can be specifically implemented by the processor 1111, and the functions implemented by the fault handling module 9211 and the mirroring module 9212 can be specifically implemented by the processor 1211.

[0120] The memory 1113 can be used to store program codes and data, so that the processor 1111 can call the program codes and data stored in the memory 1113 to implement the functions of the above-mentioned receiving module 9111 and / or processing module 9112. The processor 1111 can call the program codes and data stored in the memory 1113 to implement the above-mentioned Figure 4 The corresponding process of the described method.

[0121] The memory 1213 can be used to store program codes and data, so that the processor 1211 can call the program codes and data stored in the memory 1213 to implement the functions of the above-mentioned fault processing module 9211 and / or the mirror module 9212. The processor 1211 can call the program codes and data stored in the memory 1213 to implement the above-mentioned Figure 4 The corresponding process of the described method.

[0122] It should be noted that the target controller 1110 and the main controller 1210 may be implemented by using a separate logic device with built-in processing logic, such as an FPGA or a digital signal processor (DSP). In addition, the target controller 1110 and the main controller 1210 may include a plurality of logic devices with built-in processing logic. Fig.10 Show more or fewer components, or have different component configurations.

[0123] It should be understood that the above Fig.10 The process of improving the reliability of the storage control system by the target controller 1110 and the main controller 1210 shown in FIG. 1 to FIG. Figure 8 For the sake of brevity, the relevant description is not repeated here.

[0124] An embodiment of the present application also provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is executed on a computer or a processor, the computer or the processor executes part or all of the steps of any one of the above embodiments.

[0125] An embodiment of the present application also provides a computer program, which includes instructions. When the computer program is executed by a computer, the computer can execute part or all of the steps of any method for improving the reliability of a storage system.

[0126] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0127] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, some steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0128] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0129] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

Claims

1. A method for improving the reliability of a storage system, the storage system comprising a first control device and a second control device, the storage system being connected to a host, characterized in that: The first control device and the second control device include a controller, the controller includes a memory and a central processing unit CPU, and the controller is also connected to the memory. The method includes: The target controller receives a write request, the write request includes data to be written, the target controller belongs to the first control device, and is any one of the controllers in the first control device; The target controller writes the data to be written into the memory of the target controller; The target controller writes the data to be written into the memory of the mirror controller of the target controller, at least one controller of the mirror controller belongs to the second control device; The target controller and the mirror controller of the target controller constitute a backup controller group, the backup controller group includes N controllers, N is an integer greater than or equal to 2; the method further includes: When M controllers including the target controller fail, the main controller selects a controller from the mirror controllers of the target controller as a new target controller, and the new target controller is used to receive the write request. The main controller is a controller selected from the controllers in the first control device and the second control device to manage other controllers, wherein a mirror controller with a high priority is selected from the mirror controllers of the target controller as the new target controller according to the priority level, and the main controller sets the mirror controller in the same control device as the target controller as a high-priority controller.

2. The method according to claim 1, characterized in that The first control device and the second control device are connected to a storage device, and the method further includes: When the first control device fails, the storage device is accessed through the second control device.

3. The method according to claim 1, characterized in that The N is equal to 3.

4. The method according to claim 1 or 2, characterized in that: When the first control device fails, one of the mirror controllers of the target controller in the second control device is used as a new target controller, a mirror controller is established for the new target controller, and data in the memory of the new target controller is mirrored to the memory of the new mirror controller.

5. A storage control system, comprising a first control device and a second control device, wherein the storage control system is connected to a host, and characterized in that: The first control device and the second control device include a controller, the controller includes a memory and a central processing unit CPU, and the controller is also connected to the memory. The first control device includes a target controller, which is any one of the controllers in the first control device, and the target controller includes: A receiving module, used for receiving a write request, wherein the write request includes data to be written; A processing module, used for writing the data to be written into the memory of the target controller; The processing module is further used to write the data to be written into the memory of the mirror controller of the target controller, at least one controller of the mirror controller belongs to the second control device; The target controller and the mirror controller of the target controller constitute a backup controller group, and the backup controller group includes N controllers, where N is an integer greater than or equal to 2; A fault handling module, when M controllers including the target controller fail, the main controller selects a controller from the mirror controllers of the target controller as a new target controller, the new target controller is used to receive the write request, the main controller is a controller selected from the controllers in the first control device and the second control device for managing other controllers, wherein a mirror controller with a high priority is selected from the mirror controllers of the target controller as the new target controller according to the priority level, and the main controller sets the mirror controller in the same control device as the target controller as a high-priority controller.

6. The storage control system according to claim 5, characterized in that: The N is equal to 3.

7. The storage control system according to claim 5, characterized in that: When the first control device fails, one of the mirror controllers of the target controller in the second control device is used as a new target controller, a mirror controller is established for the new target controller, and data in the memory of the new target controller is mirrored to the memory of the new mirror controller.

8. A storage control system, comprising a first control device and a second control device, wherein the storage system is connected to a host, and characterized in that: The first control device comprises a target controller, the target controller comprises a processor and a memory, the processor executes computer instructions stored in the memory, so that the target controller executes the method according to any one of claims 1 to 4.

9. A computer storage medium, characterized in that The computer storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

10. A computer program product, comprising computer instructions, which, when executed by a computer, enables the computer to perform the method according to any one of claims 1 to 4.

Citation Information

Patent Citations

  • Data processing method and device for disk array

    CN101833422A

  • RAID (redundant arrays of independent disks) subsystem and data read-write method thereof

    CN103488435A

  • Controller switching method and device

    CN105159802A

  • Data mirroring method and storage system

    CN107168820A

  • Data mirroring method and storage system

    CN108664354A