Redundant array of independent disks controller, method, device and host apparatus for controlling the same

Abstract

This invention discloses an independent disk redundant array controller and its control method, device, and host device, relating to the field of storage technology. The RAID controller includes: a communication endpoint unit, used to determine the target operating mode according to the control instructions sent by the host device; if the target operating mode is XOR accelerator mode, then calling the XOR calculation engine to process the direct-connect disk management operation tasks issued by the host device; wherein, the direct-connect disk management operation tasks include parity check operation tasks corresponding to the disk devices connected to the host device; this invention utilizes the RAID controller connected to the host device to realize hardware acceleration of the host device's hybrid hardware and software RAID, inheriting the high performance, high reliability, and low latency characteristics of hardware RAID, and possessing the convenience of hybrid RAID in terms of scalability and maintenance upgrades, which can significantly extend the life cycle of the storage system while reducing the cost of configuration and maintenance upgrades.

Description

Technical Field

[0001] This invention relates to the field of storage technology, and in particular to an independent disk redundant array controller and its control method, device, computer program product, host device, storage system and server. Background Technology

[0002] RAID (Redundant Array of Independent Disks) combines multiple independent physical disks in a specific way to form a logical storage unit, thereby improving the data read and write speed of the storage system and enhancing data redundancy to improve fault tolerance.

[0003] Currently, RAID implementation solutions for host devices can be categorized into three types: hardware RAID, software RAID, and hybrid hardware / software RAID. Hardware RAID uses a specially designed hardware RAID controller; however, RAID functionality is limited by hardware, typically only supporting specific RAID levels at the factory, resulting in poor scalability. Software RAID utilizes the host device's operating system or third-party software to implement RAID functionality, requiring no additional hardware. It offers low cost, high flexibility, and broad compatibility; however, software RAID runs directly at the operating system level, with the host CPU (Central Processing Unit) handling RAID-related computations, such as parity calculations. This consumes significant host system resources, severely impacting the efficiency of other host system services in high-load or I / O-intensive applications. Hybrid hardware / software RAID utilizes dedicated hardware to accelerate complex computationally intensive tasks, while software handles RAID-related management and configuration tasks, such as RAID level configuration, data striping management, data reconstruction, hot spares, and data migration. This approach significantly reduces the burden on the host CPU and improves storage performance through hardware acceleration, but requires additional hardware, resulting in higher costs.

[0004] Therefore, how to implement RAID functionality on host devices without setting up additional dedicated hardware, while providing high performance and high reliability, maintaining flexibility and compatibility, and reducing configuration costs, is an urgent problem to be solved. Summary of the Invention

[0005] The purpose of this invention is to provide an independent disk redundant array controller and its control method, device, computer program product, host device, storage system and server, which can provide high performance and high reliability, maintain flexibility and compatibility, and reduce configuration costs.

[0006] To address the aforementioned technical problems, this invention provides an independent redundant disk array controller, which is communicatively connected to both a host device and a hard disk device. The independent redundant disk array controller includes:

[0007] A communication endpoint unit connected to a host device is used to determine a target operating mode based on control commands sent by the host device; if the target operating mode is an XOR accelerator mode, the XOR computing engine is invoked to process the direct-connect disk management operation task issued by the host device; wherein, the target operating mode is any preset operating mode, the preset operating modes include independent disk redundant array controller mode and the XOR accelerator mode; the direct-connect disk management operation task includes the parity check operation task corresponding to the disk device connected to the host device;

[0008] An XOR computation engine is used to perform computational processing on the direct-connect disk management task according to the control of the communication endpoint unit.

[0009] On the other hand, the standalone redundant disk array controller also includes: a processor, an input / output manager, a cache, and a physical disk manager;

[0010] The communication endpoint unit is also used to, when the target operating mode is the independent disk redundant array controller mode, invoke the XOR calculation engine, the processor, the input / output manager, the cache and the physical disk manager to control the hard disk device to provide hard disk storage services to the host device.

[0011] On the other hand, the communication endpoint unit is also used to set the processor, the input / output manager, the cache and the physical disk manager to a hibernation state when the target working mode is the XOR accelerator mode.

[0012] On the other hand, the hard disk storage service includes queue configuration management functions for independent disk redundant arrays, storage data management functions, parity calculation functions, data migration functions, and control functions for the hard disk devices.

[0013] On the other hand, the communication endpoint unit is connected to the host device via a peripheral component rapid interconnect bus.

[0014] On the other hand, the communication endpoint unit is also used to provide the host device with independent disk redundant array controller functions and XOR accelerator functions through single root input / output virtualization technology.

[0015] On the other hand, the hard disk device includes a solid-state drive;

[0016] The communication endpoint unit is further configured to, when the target working mode is the XOR accelerator mode, invoke the XOR calculation engine and the hard disk XOR calculation engine to process the direct disk management calculation task issued by the host device; wherein the hard disk XOR calculation engine is the XOR calculation engine in the hard disk controller of the solid-state drive.

[0017] On the other hand, the XOR calculation engine includes at least two processing units; the processing units are used to process the XOR operations corresponding to the direct-connect disk management processing tasks in parallel.

[0018] On the other hand, the preset working mode also includes a multi-functional concurrent mode;

[0019] The communication endpoint unit is also used to time-multiplex the XOR computing engine when the target working mode is the multi-functional concurrent mode, to provide hard disk storage services to the host device and to process the direct-connect disk management computing tasks.

[0020] This invention also provides a control method for an independent disk redundant array controller, applied to a host device, comprising:

[0021] Obtain the direct-connect disk management operation task; wherein, the direct-connect disk management operation task includes the parity check operation task corresponding to the disk device connected to the host device;

[0022] The operating mode of the independent disk redundancy array controller is adjusted, and the direct-connect disk management operation task is sent to the independent disk redundancy array controller so as to utilize the XOR calculation engine of the independent disk redundancy array controller to process the direct-connect disk management operation task; wherein, the independent disk redundancy array controller is the independent disk redundancy array controller as described above.

[0023] On the other hand, when the number of independent redundant disk array controllers connected to the host device is greater than or equal to 2, adjusting the operating mode of the independent redundant disk array controller and sending the direct-connect disk management operation task to the independent redundant disk array controller includes:

[0024] Based on the operating mode information of each independent disk redundant array controller, the target controller corresponding to the current task to be issued is determined; wherein, the current task to be issued is any of the direct-connect disk management and operation tasks, and the target controller is any of the independent disk redundant array controllers;

[0025] Adjust the working mode of the target controller and send the currently pending task to the target controller so that the target controller's XOR calculation engine can process the currently pending task.

[0026] On the other hand, determining the target controller corresponding to the task to be issued based on the operating mode information of each of the independent disk redundant array controllers includes:

[0027] Based on the operating mode information of each independent disk redundant array controller and the preset operating mode priority order, the target controller corresponding to the current task to be issued is determined; wherein, the operating mode information includes the current operating mode of each independent disk redundant array controller, the current operating mode is any preset operating mode, and the preset operating mode priority order is the priority order of each preset operating mode.

[0028] On the other hand, the hard disk devices connected to the independent disk redundancy array controller include solid-state drives (SSDs). Adjusting the operating mode of the independent disk redundancy array controller and sending the direct-connect disk management operation task to the independent disk redundancy array controller includes:

[0029] Based on the operating mode information of the independent disk redundant array controller and the hard disk controller, the target controller corresponding to the current task to be issued is determined; wherein, the current task to be issued is any of the direct-connect disk management and operation tasks, and the target controller is any of the independent disk redundant array controllers or the hard disk controller; the hard disk controller is set in the solid-state drive, and the hard disk controller is equipped with an XOR calculation engine;

[0030] Adjust the working mode of the target controller and send the currently pending task to the target controller so that the target controller's XOR calculation engine can process the currently pending task.

[0031] The present invention also provides a control device for an independent disk redundant array controller, applied to a host device, comprising:

[0032] The task acquisition module is used to acquire direct-connect disk management operation tasks; wherein, the direct-connect disk management operation tasks include parity check operation tasks corresponding to the disk devices connected to the host device;

[0033] The task distribution module is used to adjust the working mode of the independent disk redundancy array controller and send the direct-connect disk management operation task to the independent disk redundancy array controller so as to use the XOR calculation engine of the independent disk redundancy array controller to process the direct-connect disk management operation task; wherein, the independent disk redundancy array controller is the independent disk redundancy array controller as described above.

[0034] The present invention also provides a computer program product, including a computer program / instruction that, when executed by a processor, implements the steps of the control method of the independent disk redundant array controller as described above.

[0035] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the control method of the independent disk redundant array controller described above.

[0036] The present invention also provides a host device, comprising:

[0037] Memory, used to store computer programs;

[0038] A processor for executing the computer program to implement the steps of the control method for a standalone redundant disk array controller as described above.

[0039] The present invention also provides a storage system, comprising: a standalone redundant disk array controller as described above, a host device, a disk device, and a hard disk device as described above;

[0040] The host device is communicatively connected to both the disk device and the independent disk redundancy array controller, and the independent disk redundancy array controller is communicatively connected to the hard disk device.

[0041] On the other hand, the number of independent disk redundant array controllers is greater than or equal to 2.

[0042] In addition, the present invention also provides a server, including: the storage system as described above.

[0043] The present invention provides an independent redundant disk array controller, which is communicatively connected to a host device and a hard disk device. The independent redundant disk array controller includes: a communication endpoint unit connected to the host device, used to determine a target operating mode according to control commands sent by the host device; if the target operating mode is an XOR accelerator mode, then calling an XOR calculation engine to process the direct-connect disk management operation tasks issued by the host device; wherein, the target operating mode is any preset operating mode, including the independent redundant disk array controller mode and the XOR accelerator mode; the direct-connect disk management operation tasks include parity check operation tasks corresponding to the hard disk devices connected to the host device; and an XOR calculation engine used to perform operation processing on the direct-connect disk management operation tasks according to the control of the communication endpoint unit.

[0044] As can be seen, the independent redundant disk array controller in this invention supports two operating modes: RAID controller mode and XOR accelerator mode. In XOR accelerator mode, it can invoke the XOR calculation engine within the independent redundant disk array controller to handle parity calculation tasks for the disk devices managed by the host device. This utilizes the independent redundant disk array controller connected to the host device to achieve hardware acceleration of the host device's hybrid hardware and software RAID, inheriting the high performance, high reliability, and low latency characteristics of hardware RAID, while possessing the flexibility, scalability, compatibility, and ease of maintenance and upgrades of hybrid RAID. This significantly extends the lifespan of the storage system while reducing configuration and maintenance upgrade costs. Furthermore, this invention also provides a control method, apparatus, computer program product, host device, storage system, and server for an independent redundant disk array controller, which also possess the aforementioned beneficial effects. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0046] Figure 1 This is a schematic diagram illustrating a traditional connection method between a RAID controller, the host device, and flash memory.

[0047] Figure 2 This is a schematic diagram of the organizational structure of a traditional hardware and software hybrid RAID storage system.

[0048] Figure 3 This is a structural block diagram of an independent disk redundancy array controller provided in an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram of another independent disk redundant array controller provided in an embodiment of the present invention;

[0050] Figure 5 This is a schematic diagram of the RAID controller mode of a RAID controller provided in an embodiment of the present invention;

[0051] Figure 6 This is a schematic diagram of the XOR accelerator mode of a RAID controller provided in an embodiment of the present invention.

[0052] Figure 7 A flowchart illustrating a control method for an independent disk redundant array controller provided in an embodiment of the present invention;

[0053] Figure 8 This is a structural block diagram of a control device for an independent disk redundancy array controller provided in an embodiment of the present invention;

[0054] Figure 9 This is a schematic diagram of the structure of a host device provided in an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] In related technologies, in traditional hardware RAID solutions, the connection method between the RAID controller and the host device and hard drive devices can be as follows: Figure 1 As shown, the RAID controller interconnects with the host device via a PCIe (Peripheral Component Interconnect Express) bus, and with the hard drives (physical disks #1-n) via SAS (Serial Attached SCSI), SATA (Serial ATA), or PCIe buses. The RAID controller acts as a bridge between the host and the hard drives, managing all independent hard drives in the downstream array and organizing them into one or more logical storage units according to specific RAID levels, such as RAID1 (a RAID storage level) and RAID5 (a RAID storage level). It also schedules and distributes data access operations originating from the host device among multiple hard drives. For RAID levels with data parity, the controller is also responsible for handling parity calculations, data splitting and reassembly, error management, and data recovery.

[0057] The organizational structure of traditional hybrid RAID architecture storage systems is as follows: Figure 2As shown, the host device connects directly to the physical disk devices (physical disks #1-m) via SAS, SATA, or PCIe buses, and communicates with the XOR (Exclusive OR) calculation engine (i.e., dedicated hardware device) using the PCIe interface or a dedicated interface. The XOR calculation engine can be a dedicated hardware circuit used to significantly improve the efficiency of parity calculations in the RAID system. The operating system software running on the host device is responsible for managing the physical disk array, organizing the independent disk devices into logical storage units according to the user-defined RAID level, and coordinating data read and write operations between these disk devices. For RAID levels configured with parity calculations, the operating system software can take on the task of managing and scheduling the XOR calculation engine, ensuring the accurate calculation and verification of parity data, thereby maintaining data integrity and reliability. While this hybrid hardware and software RAID architecture storage system can enhance the overall performance and response speed of the storage system by utilizing the hardware acceleration of the additional XOR calculation engine, the additional XOR calculation engine increases configuration costs.

[0058] Therefore, by reusing the XOR calculation engine in the RAID controller, the present invention enables the host device to achieve hardware acceleration of hybrid RAID using the connected RAID controller, avoiding the need for an additional XOR calculation engine, achieving a balance between high performance and low cost, and possessing the advantages of hybrid RAID in terms of flexibility, scalability, compatibility, and maintenance and upgrades.

[0059] For details, please refer to Figure 3 , Figure 3 This is a structural block diagram of a standalone redundant array controller provided in an embodiment of the present invention. The standalone redundant array controller is communicatively connected to both the host device and the hard disk device. The RAID controller may include:

[0060] The communication endpoint unit 10, which is connected to the host device, is used to determine the target working mode according to the control command sent by the host device. If the target working mode is the XOR accelerator mode, the XOR calculation engine 20 is invoked to process the direct disk management operation task sent by the host device. The target working mode is any preset working mode, which includes the independent disk redundant array controller mode and the XOR accelerator mode. The direct disk management operation task includes the parity check operation task corresponding to the disk device connected to the host device.

[0061] The XOR calculation engine 20 is used to perform calculations on direct-connect disk management tasks according to the control of the communication endpoint unit 10.

[0062] It is understood that in this embodiment, the communication endpoint unit 10 can be a component used for bidirectional communication with the host device. The specific type of the communication endpoint unit 10 in this embodiment, i.e., the communication method between the communication endpoint unit 10 and the host device, can be set by the designer according to the usage scenario and user requirements. For example, the communication endpoint unit 10 can be connected to the host device via a PCIe bus; that is, the communication endpoint unit 10 can specifically be a PCIe Endpoint (EP) unit. Figure 4 PCIe EP in. For example... Figure 4 As shown, the PCIe EP can be a key functional component in the PCIe hierarchy, possessing the ability to communicate efficiently bidirectionally with the Root Complex (master controller) of the host device. Through the PCIe bus, the PCIe EP can transmit data, control commands, and execute configuration commands. The communication endpoint unit 10 can also communicate bidirectionally with the host device via other communication methods such as the SAS bus or SATA bus; this embodiment does not impose any limitations on this.

[0063] Correspondingly, the communication endpoint unit 10 in this embodiment can provide RAID controller functions (such as...). Figure 4 Function A) and XOR accelerator functions (such as Figure 4 Function B in the RAID controller enables the host device to control the start and stop of two functions, corresponding to the control of the current working mode of the RAID controller (i.e., the target working mode); the target working mode can be a preset working mode. In other words, in this embodiment, the RAID controller function can control the hard drive devices connected to the RAID controller to provide hard drive storage services to the host device, such as the conventional storage function provided by the RAID controller in related technologies; the XOR accelerator function can call the XOR calculation engine 20 in the RAID controller to provide an XOR accelerator function to the host device; that is, a single RAID controller can not only implement the conventional RAID controller function, but also provide the XOR accelerator function, thereby optimizing the utilization efficiency of hardware resources, reducing the need for additional hardware, and lowering configuration costs.

[0064] Accordingly, in this embodiment, the communication endpoint unit 10 can determine the target operating mode based on the control commands sent by the host device. For example, the host device can send control commands to the communication endpoint unit 10 to control the start and stop of the RAID controller function and the XOR accelerator function, enabling the communication endpoint unit 10 to determine the target operating mode of the RAID controller from preset operating modes. The preset operating modes may include a RAID controller mode and an XOR accelerator mode. For instance, the communication endpoint unit 10 can determine the target operating mode as the RAID controller function when the host device starts the RAID controller function and stops the XOR accelerator function (i.e., the XOR accelerator function remains silent); and determine the target operating mode as the XOR accelerator mode when the host device stops the RAID controller function and starts the XOR accelerator function. Furthermore, the preset operating modes may also include a multi-functional concurrent mode, meaning the communication endpoint unit 10 can determine the target operating mode as a multi-functional concurrent mode when the host device starts the RAID controller function and the XOR accelerator function.

[0065] In order to facilitate the host device's control of the RAID controller, such as Figure 4 As shown, when the communication endpoint unit 10 is specifically a PCIe EP, the PCIe EP can provide RAID controller functions and XOR accelerator functions to the host device through PCIe's SR-IOV (Single Root I / O Virtualization) technology. This allows the host device to detect and recognize the RAID controller function (Function A) and XOR accelerator function (Function B) provided by the PCIe EP. In other words, SR-IOV technology allows a single RAID controller to support two or more PCIe physical functions or virtual functions. The PCIe EP can use SR-IOV technology to enable the host device to detect and recognize the RAID controller device corresponding to the RAID controller function and the XOR accelerator device corresponding to the XOR accelerator function. By calling these two virtualized devices, the host device can control the start and stop of the RAID controller function and the XOR accelerator function.

[0066] It is understood that in this embodiment, the communication endpoint unit 10 can call the XOR calculation engine 20 in the RAID controller to process the direct-connect disk management calculation tasks issued by the host device when the target working mode is XOR accelerator mode. This allows the XOR calculation engine 20 to work with the host's operating system or third-party software to implement a hybrid hardware and software RAID. In other words, the direct-connect disk management calculation tasks in this embodiment can be calculation tasks that require hardware acceleration by the XOR calculation engine 20 when the host device manages directly connected disk devices through RAID functionality. For example, the direct-connect disk management calculation tasks may include parity calculation tasks corresponding to the disk devices connected to the host device, such as parity calculation tasks for RAID5 (a type of RAID storage level) and RAID6 (a type of RAID storage level).

[0067] Correspondingly, in this embodiment, the communication endpoint unit 10 can also call the XOR calculation engine 20 and other hardware components in the RAID controller to control the hard disk device connected to the flash memory controller and provide hard disk storage services for the host device when the target working mode is XOR accelerator mode.

[0068] For example, a RAID controller may also include: a processor 30 (such as a central processing unit CPU), an input / output manager 40 (such as... Figure 4 The IO manager in the middle), cache 50 (such as Figure 4 The communication endpoint unit 10 is also used to, when the target operating mode is RAID controller mode, invoke the XOR calculation engine 20, processor 30, input / output manager 40, cache 50, and physical disk manager 60 to control the hard disk devices connected to the physical disk manager 60 and provide hard disk storage services to the host device. For example, Figure 4 As shown, the processor 30 in the RAID controller can run RAID controller software to perform tasks such as RAID array control and management, RAID level configuration, array status monitoring and reporting, error handling and recovery.

[0069] The input / output manager 40 in the RAID controller can be responsible for tasks such as I / O request scheduling, data striping and reassembly, and load balancing. It receives I / O requests from the host device and intelligently schedules these requests to appropriate hard disk devices based on factors such as the read / write type, priority, and RAID strategy. For read operations, it can also reassemble data blocks from multiple hard disk devices into the original data and return it to the host device.

[0070] The physical disk manager 60 within the RAID controller is a key component responsible for communicating and managing connected hard drive devices (such as HDDs or SSDs). It translates data read / write commands into a command set that the hard drive devices can understand, such as SAS, SATA, or NVMe (Non-Volatile Memory Express) protocols, and sends these commands to the corresponding hard drive devices for execution. Specifically, when the RAID controller is connected to an HDD, it can connect to the HDD (or hard disk drive) via a SAS or SATA bus; when it is connected to an SSD, it can connect to the SSD. In other words, the hard drive devices connected to the RAID controller can include HDDs and / or SSDs with SAS, SATA, or NVMe standard interfaces.

[0071] The XOR calculation engine 20 in the RAID controller is a dedicated hardware circuit that accelerates parity calculations required for RAID 5 and RAID 6 levels. It effectively offloads the computational load from the processor 30 in the RAID controller, significantly improving the write performance and data reconstruction speed of the RAID system, thus enhancing overall efficiency. Furthermore, the XOR calculation engine 20 may include at least two processing units that can process the XOR operations required for parity calculations in parallel, thereby achieving efficient data verification and processing capabilities.

[0072] The cache 50 in the RAID controller (such as dynamic random access memory DRAM) can be used to temporarily store frequently accessed data blocks, improving data access speed, reducing access latency, reducing I / O operations on hard drives, and extending the lifespan of hard drives. In the event of a hard drive failure in the RAID array, the cache 50 can also accelerate the RAID data reconstruction process.

[0073] In other words, Figure 4 The RAID controller shown integrates two independent functions: Function A acts as the RAID controller device, and Function B acts as an XOR accelerator device. These two functions have their own independent resource allocation and can operate independently, each performing its specific tasks to optimize the overall system performance and efficiency. This is achieved by reusing the XOR calculation engine resources within the RAID controller, thereby optimizing resource utilization efficiency.

[0074] For example, such as Figure 5As shown, when the communication endpoint unit 10 determines that the target working mode is RAID controller mode, that is, the RAID controller function is activated and the XOR accelerator function remains silent, the RAID controller can play the role of a normal RAID controller, managing the physical hard disk devices (disks R.1-n) to provide comprehensive hard disk storage services for the host device; that is, the processor 30, input / output manager 40, XOR calculation engine 20, cache 50 (high-speed cache) and physical disk manager 60 inside the RAID controller can work together to realize hard disk storage services in order to realize the RAID controller function.

[0075] Accordingly, hard disk storage services can include RAID array queue configuration management, storage data management, parity calculation, data migration, and hard disk device control functions. Specifically, the RAID array queue configuration management function manages specific RAID array configurations; the storage data management function performs data read / write operations, processes data striping and reassembly, schedules and optimizes I / O operations, and performs data reconstruction in case of hard disk device failure; the parity calculation function performs parity checks; the data migration function supports data migration; and the hard disk device control function controls the operation of the hard disk devices.

[0076] like Figure 6 As shown, when the communication endpoint unit 10 determines that the target operating mode is XOR accelerator mode, that is, the XOR accelerator function is activated and the RAID controller function remains silent, the RAID controller can act as a hardware XOR accelerator, specifically providing hardware acceleration services for the host device to improve data processing efficiency. In this case, the processor 30, input / output manager 40, cache 50 (high-speed cache), and physical disk manager 60 inside the RAID controller can all be put into a hibernation state to reduce system power consumption and optimize resource utilization. That is to say, the communication endpoint unit 10 can also be used to set the processor 30, input / output manager 40, cache 50, and physical disk manager 60 to a hibernation state when the target operating mode is XOR accelerator mode.

[0077] Correspondingly, Figure 6The XOR acceleration engine in the RAID controller can work with the host device's operating system software or third-party software to implement a hybrid hardware and software RAID solution. The operating system software or third-party software running on the host device can take over the management responsibilities of the RAID array, and be responsible for the following tasks: managing the disk devices (disks D.1-m) in the RAID array; performing data read and write operations to ensure the accuracy and efficiency of data access; scheduling I / O operations to optimize data flow and improve system response speed; handling data striping and reassembly to adapt to different data access modes; being responsible for data reconstruction when disk devices fail, ensuring data integrity and availability; and scheduling the XOR acceleration engine to accelerate parity calculations for RAID5 and RAID6 levels. That is, the host device can send parity calculation tasks to the RAID controller to utilize the XOR acceleration engine inside the RAID controller to perform the XOR calculations required for parity calculations.

[0078] Furthermore, Figure 6 This demonstration uses a host device connected to one RAID controller as an example. For host devices connected to two or more RAID controllers, such as host devices connected to their respective hard drives through two RAID controllers, the host device can also distribute parity calculation tasks to all or some of the connected RAID controllers to further improve the RAID storage performance of the host device by utilizing the XOR acceleration engines in multiple RAID controllers.

[0079] For example, the host device can determine the target controller based on the operating mode information of each RAID controller; send the current direct-connect disk management operation task (such as parity operation task) to the target controller, and control the adjustment of the target controller's operating mode; wherein, the target controller can be any RAID controller among the RAID controllers connected to the host device; for example, if there is a non-working RAID controller among the RAID controllers connected to the host device, a RAID controller is selected as the target controller from the non-working RAID controllers to execute the current direct-connect disk management operation task; if there is no non-working RAID controller, the target controller is determined from the RAID controllers according to the preset operating mode priority order; for example, the preset operating mode priority order can be that the priority of the XOR accelerator mode is higher than the priority of the RAID controller mode is higher than the priority of the multi-function concurrent mode, so as to preferentially call the RAID controllers that have not started their RAID controller functions for hardware acceleration; the preset operating mode priority order can also be that the priority of the RAID controller mode is higher than the priority of the XOR accelerator mode is higher than the priority of the multi-function concurrent mode, or other priority orders, and this embodiment does not impose any restrictions on this.

[0080] Furthermore, when the RAID controller connected to the host device includes a solid-state drive (SSD), since the SSD's hard drive controller (such as an NVMe SSD controller) contains an XOR calculation engine (i.e., a hard drive XOR calculation engine), in some embodiments, the host device can also send parity calculation tasks to the RAID controller and the hard drive controller to further improve the host device's RAID storage performance by utilizing the XOR calculation engine in the hard drive controller. In other words, when the target operating mode is XOR accelerator mode (or multi-functional concurrent mode), the communication endpoint unit 10 in the RAID controller can also call the XOR calculation engine 20 and the hard drive XOR calculation engine to process the direct-connect disk management calculation tasks sent by the host device; wherein, the hard drive XOR calculation engine is the XOR calculation engine in the SSD's hard drive controller.

[0081] Furthermore, the preset operating mode also includes a multi-functional concurrent mode. When the communication endpoint unit 10 determines that the target operating mode is the multi-functional concurrent mode, that is, both the XOR accelerator function and the RAID controller function are activated, the communication endpoint unit 10 time-division multiplexes the XOR calculation engine 20 to provide hard disk storage services to the host device and handle direct-connect disk management calculation tasks, achieving efficient resource sharing. In other words, the XOR accelerator function and the RAID controller function can share the XOR calculation engine resources inside the RAID controller through a time-division multiplexing mechanism, optimizing resource utilization efficiency. Figure 4 As shown, the RAID controller function (Function A) manages its subordinate hard disk devices (disks R.1-n), performing tasks including data read / write, array management, and I / O scheduling to ensure the high-performance operation of the hardware RAID controller. Meanwhile, the XOR accelerator function (Function B) utilizes the XOR calculation engine to provide hardware acceleration for parity calculations, enhancing data integrity and reliability. The host device's operating system software manages another set of disk devices (disks D.1-m), forming an independent RAID array. This RAID array is isolated from the RAID array managed by Function A, operating independently, with the operating system software responsible for its respective data management and maintenance tasks. This design allows different functions to be optimized for different needs, providing the host device with a flexible, efficient, and reliable storage solution.

[0082] Correspondingly, the specific method by which the communication endpoint unit 10 uses the time-division multiplexing XOR calculation engine 20 to provide hard disk storage services and handle direct-connect disk management calculation tasks for the host device can be set by the designer according to the practical scenario and user needs. For example, the communication endpoint unit 10 can utilize the processor 30 in the RAID controller to call the XOR calculation engine 20 in chronological order according to the issuance time of the parity calculation task issued by the host device and the generation time of its own generated parity calculation task; the communication endpoint unit 10 can utilize the processor 30 in the RAID controller to call the XOR calculation engine 20 to complete the parity calculation tasks issued by the host device and its own generated parity calculation tasks according to the preset multiplexing priority order; for example, the preset multiplexing priority order can be that the priority of the parity calculation task generated by itself is higher (or lower) than the priority of the parity calculation task issued by the host device. This embodiment does not impose any restrictions on this.

[0083] In this embodiment, the independent redundant disk array controller supports two working modes: RAID controller mode and XOR accelerator mode. In XOR accelerator mode, it can call the XOR calculation engine 20 inside the independent redundant disk array controller to handle the parity calculation task of the disk device managed by the host device. This enables hardware acceleration of the host device's hardware and software hybrid RAID by utilizing the independent redundant disk array controller connected to the host device. It inherits the high performance, high reliability and low latency characteristics of hardware RAID, and has the flexibility, scalability, compatibility and maintenance and upgrade convenience of hybrid RAID. It can significantly extend the life cycle of the storage system, while reducing the cost of configuration and maintenance upgrades.

[0084] Corresponding to the RAID controller embodiment above, this embodiment of the invention also provides a control method for an independent disk redundant array controller. The control method for an independent disk redundant array controller described below can be referred to in correspondence with the independent disk redundant array controller described above.

[0085] Please refer to Figure 7 , Figure 7 A flowchart illustrating a control method for a standalone redundant disk array controller provided in an embodiment of the present invention. This method is applied to a host device and may include:

[0086] Step 101: Obtain the direct-attached disk management operation task; wherein, the direct-attached disk management operation task includes the parity check operation task corresponding to the disk device connected to the host device.

[0087] In this embodiment, the direct-connect disk management operation task can be a hardware-accelerated operation task that the host device needs to perform through the XOR calculation engine when managing the directly connected disk device through the RAID function. For example, the direct-connect disk management operation task includes the parity check operation task corresponding to the disk device connected to the host device, such as the parity check operation tasks of RAID5 (a RAID storage level) and RAID6 (a RAID storage level).

[0088] Correspondingly, the specific method and content of the host device obtaining the direct-connect disk management computation task in this embodiment can be set by the designer according to the practical scenario and user needs. For example, it can be implemented in the same or similar way as the method of obtaining the task processed by calling a dedicated XOR calculation engine in the hardware and software hybrid RAID scheme in related technologies. For example, the direct-connect disk management computation task can include parity calculation tasks of RAID5 and / or RAID6 levels. This embodiment does not impose any restrictions on this.

[0089] Step 102: Adjust the operating mode of the independent disk redundant array controller and send the direct-connect disk management operation task to the independent disk redundant array controller so that the XOR calculation engine of the independent disk redundant array controller can be used to process the direct-connect disk management operation task.

[0090] In this embodiment, the independent redundant array of disks (RAID) controller is the independent redundant array of disks controller provided in the above embodiment.

[0091] It is understood that in this embodiment, the host device can adjust the working mode of the RAID controller (such as adjusting the working mode to XOR accelerator mode or multi-function concurrent mode) and send the direct-connect disk management operation task to the RAID controller, thereby controlling the RAID controller to call its internal XOR calculation engine to process the direct-connect disk management operation task.

[0092] In this embodiment, the specific method for adjusting the operating mode of the independent redundant disk array controller and sending the direct-connect disk management operation tasks to the independent redundant disk array controller can be set by the designer according to the practical scenario and user needs. For example, when the number of RAID controllers connected to the host device is 1, the host device can directly adjust the operating mode of the RAID controller to XOR accelerator mode or multi-function concurrent mode and send the direct-connect disk management operation tasks to the RAID controller. For example, the host device can adjust the operating mode of the RAID controller to XOR accelerator mode or multi-function concurrent mode by controlling the XOR accelerator function of the RAID controller.

[0093] Accordingly, when the number of RAID controllers connected to the host device is greater than or equal to 2, in this step, the host device determines the target controller corresponding to the current task to be sent based on the operating mode information of each RAID controller. The current task to be sent is any directly connected disk management operation task, and the target controller is any independent redundant disk array controller. The operating mode of the target controller is adjusted, and the current task to be sent is sent to the target controller so that the target controller's XOR calculation engine can process the current task. In other words, for each directly connected disk management operation task, the host device can determine the RAID controller (i.e., the target controller) where the XOR calculation engine for executing the task is located based on the current operating mode information of each RAID controller (such as their respective operating modes or the start / stop status of their respective RAID controller functions and XOR accelerator functions).

[0094] For example, the process of determining the target controller corresponding to the current task based on the operating mode information of each RAID controller may include: determining the target controller corresponding to the current task based on the operating mode information of each independent RAID controller and a preset operating mode priority order; wherein, the operating mode information includes the current operating mode of each independent RAID controller, the current operating mode being any preset operating mode, and the preset operating mode priority order being the priority order of each preset operating mode. For example, if there is an inactive RAID controller among the RAID controllers connected to the host device, a RAID controller is selected from the inactive RAID controllers as the target controller to execute the current direct-connect disk management operation task; if there is no inactive RAID controller, the target controller is determined from the RAID controllers according to the preset operating mode priority order; for example, the preset operating mode priority order may be that the priority of the XOR accelerator mode is higher than the priority of the RAID controller mode, which is higher than the priority of the multi-function concurrent mode, so as to preferentially call the RAID controller that has not started its RAID controller function for hardware acceleration; the preset operating mode priority order may also be that the priority of the RAID controller mode is higher than the priority of the XOR accelerator mode, which is higher than the priority of the multi-function concurrent mode, or other priority orders, and this embodiment does not impose any restrictions on this.

[0095] Correspondingly, when the hard drive device connected to the RAID controller includes a solid-state drive (SSD), the host device can also utilize the XOR calculation engine in the SSD's hard drive controller and the RAID controller's XOR calculation engine to jointly process direct-connect disk management tasks. For example, in this step, the host device can determine the target controller corresponding to the current task to be sent based on the operating mode information of the independent disk redundant array controller and the hard drive controller; where the current task to be sent is any direct-connect disk management task, and the target controller is any independent disk redundant array controller or hard drive controller; the hard drive controller is located in the SSD, and the hard drive controller has an XOR calculation engine; the operating mode of the target controller is adjusted, and the current task to be sent is sent to the target controller so that the target controller's XOR calculation engine can process the current task to be sent. In other words, for each direct-connect disk management task, the host device can determine the RAID controller or hard drive controller (i.e., the target controller) where the XOR calculation engine for executing the task is located based on the current operating mode information of the RAID controller and the hard drive controller (such as their respective operating modes or the start / stop status of their respective hardware controller functions and XOR accelerator functions).

[0096] Accordingly, the method for determining the target controller corresponding to the current task based on the operating mode information of the independent disk redundant array controller and the hard disk controller can be set in a similar manner to the method for determining the target controller corresponding to the current task based on the operating mode information of each RAID controller. For example, the target controller corresponding to the current task can be determined from the RAID controllers and hard disk controllers according to the pre-set priority order of their operating modes. This embodiment does not impose any limitations on this.

[0097] In this embodiment, the present invention adjusts the working mode of the independent redundant disk array controller and sends the direct-connect disk management operation task to the independent redundant disk array controller. It can utilize the XOR calculation engine inside the RAID controller to process the parity calculation task of the host device managing the disk device. This enables hardware acceleration of the host device's hardware and software hybrid RAID by utilizing the independent redundant disk array controller connected to the host device. It inherits the high performance, high reliability and low latency characteristics of hardware RAID, and has the flexibility, scalability, compatibility and maintenance and upgrade convenience of hybrid RAID. It can significantly extend the life cycle of the storage system, while reducing the cost of configuration and maintenance upgrades.

[0098] Corresponding to the above method embodiments, this invention also provides a control device for an independent redundant disk array controller. The control device for an independent redundant disk array controller described below and the control method for an independent redundant disk array controller described above can be referred to in correspondence.

[0099] Please refer to Figure 8 , Figure 8 This is a structural block diagram of a control device for an independent disk redundant array controller provided in an embodiment of the present invention. The device is applied to a host device and may include:

[0100] The task acquisition module 100 is used to acquire direct-connect disk management operation tasks; wherein, the direct-connect disk management operation tasks include parity check operation tasks corresponding to the disk devices connected to the host device;

[0101] The task distribution module 200 is used to adjust the working mode of the independent disk redundant array controller and send the direct-connect disk management operation task to the independent disk redundant array controller so as to use the XOR calculation engine of the independent disk redundant array controller to process the direct-connect disk management operation task; wherein, the independent disk redundant array controller is as described above.

[0102] In some embodiments, when the number of independent redundant disk array controllers connected to the host device is greater than or equal to 2, the task distribution module 200 may include:

[0103] The first determination submodule is used to determine the target controller corresponding to the current task to be issued based on the working mode information of each independent disk redundant array controller; wherein the current task to be issued is any direct-connect disk management operation task, and the target controller is any independent disk redundant array controller.

[0104] The first task dispatch submodule is used to adjust the working mode of the target controller and send the currently pending task to the target controller so that the target controller's XOR calculation engine can process the currently pending task.

[0105] In some embodiments, the first determining submodule may be specifically used to determine the target controller corresponding to the current task to be issued based on the working mode information of each independent disk redundant array controller and the preset working mode priority order; wherein, the working mode information includes the current working mode of each independent disk redundant array controller, the current working mode is any preset working mode, and the preset working mode priority order is the priority order of each preset working mode.

[0106] In some embodiments, the task distribution module 200 may include:

[0107] The second determination submodule is used to determine the target controller corresponding to the current task to be issued based on the working mode information of the independent disk redundant array controller and the hard disk controller; wherein, the current task to be issued is any direct-connect disk management operation task, and the target controller is any independent disk redundant array controller or hard disk controller; the hard disk controller is set in the solid-state drive, and the hard disk controller is equipped with an XOR calculation engine;

[0108] The second task dispatch submodule is used to adjust the working mode of the target controller and send the currently pending task to the target controller so that the target controller's XOR calculation engine can process the currently pending task.

[0109] In this embodiment, the present invention adjusts the working mode of the independent disk redundant array controller through the task distribution module 200 and sends the direct-connect disk management operation task to the independent disk redundant array controller. It can utilize the XOR calculation engine inside the RAID controller to process the parity calculation task of the disk device managed by the host device, so as to realize the hardware acceleration of the host device's hardware and software hybrid RAID by using the independent disk redundant array controller connected to the host device. It inherits the high performance, high reliability and low latency characteristics of hardware RAID, and has the flexibility, scalability, compatibility and maintenance and upgrade convenience of hybrid RAID. It can significantly extend the life cycle of the storage system, while reducing the cost of configuration and maintenance upgrades.

[0110] Corresponding to the above method embodiments, this invention also provides a computer program product. The computer program product described below and the control method of the independent disk redundant array controller described above can be referred to each other.

[0111] A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of a control method for a standalone redundant disk array controller as provided in the embodiments.

[0112] Corresponding to the above method embodiments, this invention also provides a computer-readable storage medium. The computer-readable storage medium described below and the control method of the independent disk redundant array controller described above can be referred to and correspond to each other.

[0113] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the control method of the independent disk redundant array controller described above.

[0114] Corresponding to the above method embodiments, this invention also provides a host device. The host device described below and the control method of the independent disk redundant array controller described above can be referred to in correspondence.

[0115] Please refer to Figure 9 , Figure 9 This is a schematic diagram of a host device provided in an embodiment of the present invention. The host device may include:

[0116] Memory D1 is used to store computer programs;

[0117] Processor D2 is used to execute computer programs to implement the steps of the control method for the independent disk redundant array controller provided in the above method embodiments.

[0118] In this embodiment, the host device can specifically be a server host or a computer.

[0119] Corresponding to the above embodiments, this invention also provides a storage system. The storage system described below can be referred to in conjunction with the host device and independent disk redundant array controller described above.

[0120] A storage system includes: a standalone redundant disk array controller as provided in the above embodiments, a host device, a disk device, and a hard disk device as provided in the above embodiments;

[0121] The host device is connected to both the disk device and the independent disk redundant array controller, and the independent disk redundant array controller is connected to the hard disk device.

[0122] In some embodiments, the number of independent redundant disk array controllers is greater than or equal to 2.

[0123] In some embodiments, the hard disk device connected to the independent disk redundancy array controller may include a solid-state drive (SSD); the hard disk controller in the SSD may include an XOR computing engine.

[0124] Corresponding to the above system embodiments, this invention also provides a server. The server described below and the storage system described above can be referred to in correspondence.

[0125] A server includes a storage system as described in the above embodiments.

[0126] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus, host device, storage system, computer program product, and computer-readable storage medium disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant details can be found in the method section.

[0127] The present invention has been described in detail above as follows: an independent disk redundant array controller and its control method, apparatus, computer program product, host device, storage system, and server. Specific examples have been used to illustrate the principles and implementation methods of the invention. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and core ideas of the invention. It should be noted that those skilled in the art can make various improvements and modifications to the invention without departing from its principles, and these improvements and modifications also fall within the protection scope of the invention.

Claims

1. A standalone redundant disk array controller, wherein the standalone redundant disk array controller is communicatively connected to a host device and a hard disk device respectively, characterized in that, The independent disk redundancy array controller includes: A communication endpoint unit connected to a host device is used to determine a target operating mode based on control commands sent by the host device; if the target operating mode is an XOR accelerator mode, the XOR computing engine is invoked to process the direct-connect disk management operation task issued by the host device; wherein, the target operating mode is any preset operating mode, the preset operating modes include independent disk redundant array controller mode and the XOR accelerator mode; the direct-connect disk management operation task includes the parity check operation task corresponding to the disk device connected to the host device; An XOR computation engine is used to perform computational processing on the direct-connect disk management task according to the control of the communication endpoint unit.

2. The independent disk redundancy array controller according to claim 1, characterized in that, Also includes: Processor, I / O manager, cache, and physical disk manager; The communication endpoint unit is also used to, when the target operating mode is the independent disk redundant array controller mode, invoke the XOR calculation engine, the processor, the input / output manager, the cache and the physical disk manager to control the hard disk device to provide hard disk storage services to the host device.

3. The independent disk redundancy array controller according to claim 2, characterized in that, The communication endpoint unit is also used to set the processor, the input / output manager, the cache and the physical disk manager to a hibernation state when the target working mode is the XOR accelerator mode.

4. The independent disk redundancy array controller according to claim 2, characterized in that, The hard disk storage service includes queue configuration management functions for independent disk redundant arrays, storage data management functions, parity calculation functions, data migration functions, and control functions for the hard disk devices.

5. The independent disk redundancy array controller according to claim 1, characterized in that, The communication endpoint unit is connected to the host device via a peripheral component rapid interconnect bus.

6. The independent disk redundancy array controller according to claim 5, characterized in that, The communication endpoint unit is also used to provide the host device with independent disk redundant array controller function and XOR accelerator function through single root input / output virtualization technology.

7. The independent disk redundancy array controller according to claim 1, characterized in that, The hard disk device includes a solid-state drive; The communication endpoint unit is further configured to, when the target working mode is the XOR accelerator mode, invoke the XOR calculation engine and the hard disk XOR calculation engine to process the direct disk management calculation task issued by the host device; wherein the hard disk XOR calculation engine is the XOR calculation engine in the hard disk controller of the solid-state drive.

8. The independent disk redundancy array controller according to claim 1, characterized in that, The XOR calculation engine includes at least two processing units; the processing units are used to process the XOR operations corresponding to the direct-connect disk management processing tasks in parallel.

9. The independent disk redundancy array controller according to any one of claims 1 to 8, characterized in that, The preset working mode also includes a multi-functional concurrent mode; The communication endpoint unit is also used to time-multiplex the XOR computing engine when the target working mode is the multi-functional concurrent mode, to provide hard disk storage services to the host device and to process the direct-connect disk management computing tasks.

10. A control method for an independent disk redundant array controller, characterized in that, Applied to host devices, including: Obtain the direct-connect disk management operation task; wherein, the direct-connect disk management operation task includes the parity check operation task corresponding to the disk device connected to the host device; Adjust the operating mode of the independent disk redundancy array controller and send the direct-connect disk management operation task to the independent disk redundancy array controller so as to use the XOR calculation engine of the independent disk redundancy array controller to process the direct-connect disk management operation task; wherein, the independent disk redundancy array controller is the independent disk redundancy array controller as described in any one of claims 1 to 9.

11. The control method for an independent disk redundancy array controller according to claim 10, characterized in that, When the number of independent redundant disk array controllers connected to the host device is greater than or equal to 2, adjusting the operating mode of the independent redundant disk array controllers and sending the direct-connect disk management operation task to the independent redundant disk array controllers includes: Based on the operating mode information of each independent disk redundant array controller, the target controller corresponding to the current task to be issued is determined; wherein, the current task to be issued is any of the direct-connect disk management and operation tasks, and the target controller is any of the independent disk redundant array controllers; Adjust the working mode of the target controller and send the currently pending task to the target controller so that the target controller's XOR calculation engine can process the currently pending task.

12. The control method for an independent disk redundancy array controller according to claim 11, characterized in that, The step of determining the target controller corresponding to the task to be issued based on the operating mode information of each independent disk redundant array controller includes: Based on the operating mode information of each independent disk redundant array controller and the preset operating mode priority order, the target controller corresponding to the current task to be issued is determined; wherein, the operating mode information includes the current operating mode of each independent disk redundant array controller, the current operating mode is any preset operating mode, and the preset operating mode priority order is the priority order of each preset operating mode.

13. The control method for an independent disk redundancy array controller according to any one of claims 10 to 12, characterized in that, The hard disk devices connected to the independent disk redundancy array controller include solid-state drives (SSDs). Adjusting the operating mode of the independent disk redundancy array controller and sending the direct-connect disk management computation task to the independent disk redundancy array controller includes: Based on the operating mode information of the independent disk redundant array controller and the hard disk controller, the target controller corresponding to the current task to be issued is determined; wherein, the current task to be issued is any of the direct-connect disk management and operation tasks, and the target controller is any of the independent disk redundant array controllers or the hard disk controller; the hard disk controller is set in the solid-state drive, and the hard disk controller is equipped with an XOR calculation engine; Adjust the working mode of the target controller and send the currently pending task to the target controller so that the XOR calculation engine of the target controller can process the currently pending task.

14. A control device for an independent disk redundancy array controller, characterized in that, Applied to host devices, including: The task acquisition module is used to acquire direct-connect disk management operation tasks; wherein, the direct-connect disk management operation tasks include parity check operation tasks corresponding to the disk devices connected to the host device; The task dispatch module is used to adjust the working mode of the independent disk redundancy array controller and send the direct-connect disk management operation task to the independent disk redundancy array controller so as to use the XOR calculation engine of the independent disk redundancy array controller to process the direct-connect disk management operation task; wherein, the independent disk redundancy array controller is the independent disk redundancy array controller as described in any one of claims 1 to 9.

15. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the control method of the independent disk redundant array controller as described in any one of claims 10 to 13.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the control method of the independent disk redundant array controller as described in any one of claims 10 to 13.

17. A host device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the control method of the independent disk redundant array controller as described in any one of claims 10 to 13.

18. A storage system, characterized in that, include: The standalone redundant disk array controller as described in any one of claims 1 to 9, the host device, the disk device, and the hard disk device as described in claim 17; The host device is communicatively connected to both the disk device and the independent disk redundancy array controller, and the independent disk redundancy array controller is communicatively connected to the hard disk device.

19. The storage system according to claim 18, characterized in that, The number of independent disk redundant array controllers is greater than or equal to 2.

20. A server, characterized in that, include: The storage system as described in claim 19.

Images