WO2023160378A1

WO2023160378A1 - Storage device, storage method, computing device, and storage medium

Info

Publication number: WO2023160378A1
Application number: PCT/CN2023/074779
Authority: WO
Inventors: 高军; 黎亮; 屈欢
Original assignee: 华为技术有限公司
Priority date: 2022-02-24
Filing date: 2023-02-07
Publication date: 2023-08-31
Also published as: CN116700594A

Abstract

The present application belongs to the technical field of computer storage. Disclosed are a storage device, a storage method, a computing device, and a storage medium. In the technical solution provided in the embodiments of the present application, a plurality of mutually isolated operating systems are respectively operated on the basis of different portions of hardware resources of a storage device, and when a first operating system, which is operated on a first hardware resource group, fails, an access task of the first operating system is migrated to a second operating system, which is operated on a second hardware resource group, for execution. By means of the technical solution, the isolation of faults among a plurality of operating systems in a single node of a data storage system is guaranteed, and when a device fails, an access task may be migrated to other mutually isolated operating systems for continuous execution without triggering a controller to reset, thereby achieving an effect of a front end having no awareness regarding a fault and greatly improving the reliability of the device.

Description

Storage device, storage method, computing device and storage medium

technical field

The present application relates to the technical field of computer storage, and in particular to a storage device, a storage method, a computing device and a storage medium.

Background technique

With the development of hardware production process, the precision of all kinds of hardware in computing equipment is getting higher and higher, followed by the problem of reduced reliability of hardware. influence becomes larger.

Computing devices are often used as storage devices in data storage systems. At present, storage devices commonly used in the industry usually run an operating system (OS) on a physical machine, and the operating system is based on the central Hardware resources such as processor (central processing unit, CPU), memory, and external devices run.

However, in the storage device based on the above-mentioned architecture, when the reliability of the hardware is reduced and a failure occurs, for example, a CPU core in a multi-core CPU fails, it will trigger a reset of the CPU controller, which will affect the access task and make the device reliability is greatly reduced.

Contents of the invention

The present application provides a storage device, a storage method, a computing device, and a storage medium, which can effectively improve the reliability of the device. The technical solution is as follows:

In a first aspect, a storage device is provided, the storage device includes a management device and hardware resources, the hardware resources are divided into multiple hardware resource groups, each of the hardware resource groups runs a set of operating systems; the The management device is used to:

When a first hardware resource group in the multiple hardware resource groups fails, migrating an access task executed by the first operating system running on the first hardware resource group to a second hardware resource in the multiple hardware resource groups group, the second operating system run by the second hardware resource group executes the access task.

Wherein, the hardware resources include the memory of the storage device, CPU and external devices, for example, the memory of the storage device can be a dual inline memory module (dual inline memory module, DIMM), and the external devices of the storage device include external memory, network card and display devices.

Through the above technical solution, the fault isolation between multiple operating systems in a single node of the data storage system is guaranteed. In the case of a device failure, the access task can be migrated to other isolated operating systems to continue execution without It will trigger the reset of the controller, achieving the effect that the front end is not aware of faults, and greatly improving the reliability of the equipment.

In a possible implementation manner, specifications of the first hardware resource group and the second hardware resource group are the same.

Through the above technical solution, the data processing capability of each operating system is balanced, and the usability of the device is guaranteed when a failure occurs.

In a possible implementation manner, the management device is further configured to: monitor states of the multiple hardware resource groups.

Through the above technical solution, by monitoring the hardware resources of the management device, the hardware resources of the storage device can be reasonably planned, and the failure of the hardware resource group and the operating system running by the hardware resource group can be found in time, further improving the reliability of the device.

In a possible implementation manner, the storage device further includes a network card, the network card supports a single-root input-output virtualization SR-IOV function, and the first virtual function VF of the network card is allocated to the first hardware resource group, The second virtual function VF of the network card is allocated to the second hardware resource group.

Through the above technical solution, use the hardware sharing technology to virtually assign the network card to each hardware resource group, thereby providing reliable physical communication resources for the operating systems running in each hardware resource group. Isolation effectively improves the reliability of the equipment.

In a possible implementation manner, the management device is further configured to receive failure information sent by the first hardware resource group; the failure information indicates that a failure occurs in the first operating system.

Through the above technical solution, the first operating system can promptly inform the target server of the interruption of the access task through the fault information, so as to switch the operating system, thereby realizing the migration of the access task, and ensuring Service continuity effectively improves device reliability.

In a possible implementation manner, the management device is further configured to receive a first takeover request sent by the second operating system, where the first takeover request is used to take over the access task executed by the first operating system.

In a possible implementation manner, the second operating system is configured to determine that the first operating system fails if the heartbeat message of the first operating system is not received within a first period of time.

Wherein, the heartbeat message refers to a message sent regularly by the sender, and the receiver can determine that the sender is in a running state after receiving the heartbeat message.

Through the above technical solution, each operating system can perceive each other, so that when any operating system fails, the operating system that has not failed can respond quickly and initiate a takeover request to ensure business continuity in a flexible manner. While effectively improving the reliability of the device, the availability of the storage device is also improved.

In a possible implementation manner, the management device is further configured to receive a second takeover request sent after the first operating system recovers from a fault, and the second takeover request indicates to take over the The access task.

Through the above technical solution, after troubleshooting, the first operating system can take over the access tasks that are migrated to the second operating system to reduce the load on the second operating system and ensure the load balance among the operating systems in the storage device , to further guarantee the availability of the storage device, and effectively improve the reliability of the device.

In a possible implementation manner, the management device corresponds to a target hardware resource group, and the management device is configured to configure the multiple sets of operating systems.

Through the above technical solution, when the management device fails, hardware resource groups other than the target hardware resource group will not be affected, so that the operating system running on the hardware resource group can normally perform access tasks, reducing the impact of the management device on business processes. It realizes the decoupling of the management side and the business side, and effectively improves the stability and reliability of the storage device.

In a possible implementation manner, the management device is further configured to determine that the first operating system fails if it does not receive a heartbeat message from the first operating system within a second period of time.

In a possible implementation manner, the management device is further configured to pull up the first operating system after the first operating system recovers from a fault.

In a possible implementation manner, the target hardware resource group includes fewer hardware resources than the hardware resource group.

In the above technical solution, the target hardware resource group corresponding to the management device includes less hardware resources than the above hardware resource group. Therefore, the possibility of the management device providing management and configuration functions being affected by hardware failures is greatly reduced, thereby effectively improving the reliability of the storage device.

In a possible implementation manner, the target hardware resource group includes a part of memory and a part of CPU; the The hardware resource group includes a part of memory, a part of CPU, and a part of external devices.

In the second aspect, a storage method is provided, which is applied to a storage device. The storage device includes a management device and hardware resources, and the hardware resources are divided into multiple hardware resource groups. Each hardware resource group runs a sets of operating systems;

The methods include:

In a possible implementation manner, the method further includes: the management device monitors states of the multiple hardware resource groups.

In a possible implementation manner, the method further includes: the management device receiving fault information sent by the first hardware resource group; the fault information indicates that the first operating system is faulty.

In a possible implementation manner, the method further includes: the management device receiving a first takeover request sent by the second operating system, where the first takeover request is used to take over the access performed by the first operating system Task.

In a possible implementation manner, the method further includes: if the second operating system does not receive a heartbeat message from the first operating system within a first period of time, determining that the first operating system fails.

In a possible implementation manner, the method further includes: the management device receiving a second takeover request sent after the first operating system recovers from a failure, the second takeover request indicating that the takeover has been migrated to the second The access task of the operating system.

In a possible implementation manner, the management device corresponds to a target hardware resource group, and the method further includes: the management device configures the multiple hardware resource groups.

In a possible implementation manner, the method also includes:

If the management device does not receive the heartbeat message of the first operating system within the second time period, it determines that the first operating system fails.

In a possible implementation manner, the method also includes:

The management device pulls up the first operating system after the failure of the first operating system recovers.

In a possible implementation manner, the target hardware resource group includes a part of a memory and a part of a CPU; the hardware resource group includes a part of a memory, a part of a CPU, and a part of an external device.

In a third aspect, a computing device is provided, the computing device includes a management device and hardware resources, the hardware resources are divided into multiple hardware resource groups, the hardware resources include a processor and a memory, and the memory is used to store The management device and at least one piece of program code corresponding to each of the hardware resource groups, the at least one piece of program code is loaded by the processor and executes the storage method according to the second aspect.

In a fourth aspect, a computer-readable storage medium is provided, the computer-readable storage medium is used to store at least one piece of program code, and the at least one piece of program code is used to execute the storage method as described in the second aspect.

In a fifth aspect, a computer program product is provided. When the computer program product is run on a computer, the computer is made to execute the storage method as described in the second aspect.

Description of drawings

FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application;

FIG. 2 is a schematic diagram of a hardware structure of a computing device provided by an embodiment of the present application;

FIG. 3 is a schematic structural diagram of a storage device provided by an embodiment of the present application;

FIG. 4 is a schematic diagram of hardware sharing provided by an embodiment of the present application;

FIG. 5 is a flow chart of a storage method provided by an embodiment of the present application;

FIG. 6 is a schematic diagram of an operating system switching process provided by an embodiment of the present application;

FIG. 7 is a flow chart of a storage method provided by an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application clearer, the implementation manners of the present application will be further described in detail below in conjunction with the accompanying drawings.

FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application. Referring to FIG. 1 , the implementation environment includes a storage device 110 , a storage unit 120 and a target server 130 . Wherein, the storage device 110 is used to provide data storage services for the target server 130, and the storage device 110 writes data into the storage unit 120 based on the access task received from the target server 130, or reads data from the storage unit 120 and returns to the target server 130.

Wherein, the storage device 110 includes a management device and hardware resources, and the hardware resources are divided into multiple hardware resource groups, each of which runs an operating system. In some embodiments, the storage device 110 provides data storage services for the multiple target servers 130 in parallel by sending the access tasks of the multiple target servers 130 to multiple operating systems running on multiple hardware resource groups for execution. In some embodiments, the access task includes data read and write instructions for data stored in the storage unit 120 .

In some embodiments, the aforementioned hardware resources include CPU, memory, and external devices. In some embodiments, the external device includes a network interface card (network interface card, NIC). A network interface card, referred to as a network card or a network adapter, is an expansion card inserted into an expansion slot of a computing device (such as a server) to connect the computing device to the network. Optionally, the network card is connected to a physical transmission medium (such as twisted pair, coaxial cable or optical fiber), and exchanges data with the network through the network transmission medium. In some embodiments, the network card may be a wireless network card capable of wirelessly connecting to a network within the coverage of a wireless local area network. The network is usually the Internet, but can be any network, including but not limited to local area network (LAN), metropolitan area network (MAN), wide area network (WAN), mobile, wired or wireless Any combination of a network, a private network, or a virtual private network is not limited in this embodiment of the present application.

Optionally, the above-mentioned network interface card is a PCIE card supporting the high-speed serial computer expansion bus standard (peripheral component interconnect express, PCIE). The PCIE card supports high-speed serial point-to-point dual-channel high-bandwidth transmission. Each device connected to the PCIE card is assigned an independent channel bandwidth for data transmission, and does not share the bus bandwidth.

In some embodiments, the storage device 110 and the storage unit 120, and between the storage device 110 and the target server 130 can be communicated through a wired network or a wireless network, and the wireless network or wired network uses standard communication technologies and/or or agreement. For example, the storage device 110 receives an access task from the target server 130 through the first network card, and reads and writes data to the storage unit 120 through the second network card according to the data read and write instructions indicated by the access task.

In some embodiments, the storage unit 120 is a distributed storage device composed of multiple scattered storage resources, that is, a virtual storage device constructed by integrating storage resources on multiple physical machines through a network connection. Optionally, the storage unit 120 may include a solid-state hard disk, a mechanical hard disk, or other types of storage media, which is not limited in this embodiment of the present application.

On the basis of introducing the above-mentioned implementation environment and device architecture, the application scenarios of the embodiments of the present application are introduced next.

The storage device provided by the embodiment of the present application can be used as a node connecting the target server and the storage unit in the data storage system, for example, a storage server used to manage and control storage resources, and the storage server can control the connected disk array (that is, is a storage unit) to efficiently provide reliable data storage services for client hosts (that is, target servers).

FIG. 2 is a schematic diagram of a hardware structure of a computing device provided by an embodiment of the present application. Referring to FIG. 2 , the computing device 200 can serve as a storage device, a storage unit, and a target server in the above-mentioned data storage system. Wherein, the computing device 200 may have relatively large differences due to different configurations or performances, and may include one or more than one processor (central processing units, CPU) 201 and one or more than one memory 202, and the above processor 201 and memory 202 Collectively referred to as hardware resources.

Wherein, the processor 201 may be a network processor (network processor, NP), a central processing unit (central processing unit, CPU), an application-specific integrated circuit (application-specific integrated circuit, ASIC), or a integrated circuits. The processor 201 may be a single-core (single-CPU) processor, or a multi-core (multi-CPU) processor. Memory 202 may be read-only memory (read-only memory, ROM) or other types of static storage devices that can store static information and instructions, random access memory (random access memory, RAM) or other types that can store information and instructions It can also be an electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or store desired program code in the form of instructions or data structures and can be programmed by a computer Any other medium accessed, but not limited to. Wherein, the processor 201 and the memory 202 may be provided separately, or may be integrated together. Certainly, the computing device 200 may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface for input and output, and the computing device 200 may also include other components for implementing device functions, which will not be described in detail here.

In an exemplary embodiment, there is also provided a computer-readable storage medium, such as a memory including instructions, which can be executed by a processor in a computing device to implement the storage method in the following embodiments. For example, the computer-readable storage medium can be a read-only memory (read-only memory, ROM), a random access memory (random access memory, RAM), a compact disc read-only memory (CD-ROM), Magnetic tapes, floppy disks, and optical data storage devices, etc.

Next, the storage device provided by the embodiment of the present application will be introduced.

FIG. 3 is a schematic diagram of the architecture of a storage device provided by the embodiment of the present application. Referring to FIG. 3 , the storage device 300 includes a management device and hardware resources, and the hardware resources are divided into multiple hardware resource groups. There is an operating system running.

Wherein, the hardware resources include memory, CPU and external devices of the storage device 300 . In some embodiments, the memory of the storage device 300 can be a dual inline memory module (dual inline memory module, DIMM), DIMM Signals are transmitted independently between each signal path of the system, so it can meet the transmission requirements of multiple data signals in complex scenarios. In some embodiments, the external devices of the storage device 300 include devices such as external memory, a network card, and a display.

Referring to FIG. 3 , the hardware resource includes multiple hardware resource groups, and each hardware resource group includes a part of the hardware resources. Wherein, a set of operating systems runs on each hardware resource group, therefore, the multiple hardware resource groups correspond to multiple sets of operating systems. The multiple sets of operating systems include a first operating system 320 run by the first hardware resource group 310 and a second operating system 340 run by the second hardware resource group 330 . Wherein, the management device 350 corresponds to a target hardware resource group 360 included in the hardware resources, and the target hardware resource group 360 includes a part of memory and a part of CPU in the hardware resources of the storage device 300 .

Wherein, the first hardware resource group 310 corresponding to the first operating system 320 includes: a part of the memory, for example, the memory bar 311 occupied by the first operating system 320; a part of the CPU, for example, the first operating system 320 in the multi-core CPU The occupied CPU core 312 ; a part of the external device, for example, the external memory 313 occupied by the first operating system 320 . In some embodiments, the specifications of the hardware resources included in the multiple hardware resource groups are the same, for example, each hardware resource group includes the same number of CPU cores, the same size of memory space, and the same number of external memories. Based on this, the data processing capability of each operating system is balanced to ensure the availability of the device when a failure occurs.

Of course, according to different business processing requirements and the actual situation of hardware resources, the actual hardware resources included in each hardware resource group can be adjusted accordingly. For example, in order to improve the data processing capability of a certain operating system, increase its corresponding hardware The memory included in the resource group, and the memory included in the other hardware resource groups remain the same, which is not limited in this embodiment of the present application.

In some embodiments, based on mutually isolated hardware resource groups, the kernel-mode processes of each operating system run in independent kernel spaces, and run user-mode processes in their corresponding upper-layer user spaces, for example, data operation processes and Control process, etc., see Figure 3.

It can be understood that the target hardware resource group corresponding to the management device includes fewer hardware resources than the above hardware resource group. Based on this, the possibility of the management device providing management and configuration functions being affected by hardware failures is greatly reduced, thereby effectively improving the reliability of the storage device.

In some embodiments, the management device 350 is used to configure the multiple sets of operating systems. Wherein, the initial configuration process includes: dividing the hardware resources of the storage device 300 according to business processing requirements to obtain multiple hardware resource groups; and configuring the hardware resource groups to each operating system. Optionally, after the initial configuration, each operating system persists its own configuration information in the corresponding memory, so that the operating system can be directly read after restarting.

In some embodiments, the management device 350 is also used to monitor the states of the above-mentioned multiple hardware resource groups. The management device can timely discover whether the hardware resource group reaches a performance bottleneck or fails by monitoring the use of hardware resources such as CPU, memory, and external devices included in the hardware resource group. In some other embodiments, the management device monitors the consumption of hardware resources by the process, so as to know the running status of the operating system running on each hardware resource group in a timely manner, for example, whether a fault occurs. Through the above technical solution, by monitoring the hardware resources of the management device, the hardware resources of the storage device can be reasonably planned, and the failure of the hardware resource group and the operating system running by the hardware resource group can be found in time, further improving the reliability of the device.

In some embodiments, the configuration process of the management device 350 also includes the allocation of external devices. For example, the external device includes a network card. Communication resources are allocated to the above multiple operating systems. In some embodiments, the network card is a PCIE device that supports single root I/O virtualization (single root I/O virtualization, SR-IOV), through the The SR-IOV management driver can virtualize multiple available virtual functions (Virtualization Function, VF) based on PCIE devices, and assign the virtual functions to the corresponding hardware resource groups, so that the operating system running on the corresponding hardware resource groups does not need to By managing the driver of the device, the PCIE device can be directly accessed to receive the access task. Optionally, the manner of assigning the virtual function to the operating system includes: establishing a mapping relationship between the number of the virtual function and the information of the operating system; and storing the mapping relationship in the memory corresponding to the operating system in the form of configuration information. In some embodiments, the information of the operating system may be a serial number of the operating system, which is not limited in this embodiment of the present application. Through the above technical solution, the network card is virtually assigned to each hardware resource group by using the hardware sharing technology, so as to provide reliable physical communication resources for the operating systems running on each hardware resource group, and based on this, the hardware resources of each operating system are guaranteed. They are isolated from each other, which effectively improves the reliability of the equipment.

For ease of understanding, the embodiment of the present application provides a schematic diagram of hardware sharing. Referring to FIG. 4 , a PCIE device supporting SR-IOV is driven by a management device. The SR-IOV management driver running in the management device is used to virtualize the physical functions (physical functions, PF) provided by the PCIE device into N VFs, and assign them to the operating system running on each hardware resource group by way of mapping , where N is a positive integer. Referring to Figure 4, operating system 1 can use VF0 to access PCIE devices, and operating system 2 can use VF1 to access PCIE devices. Both operating system 1 and operating system 2 have device management processes running to manage the devices accessed through VF. PCIe devices.

Through the above configuration process, the operating system can receive and execute access tasks, and, in the event of a failure of the management device, hardware resource groups other than the target hardware resource group will not be affected, so that the operating system running on the hardware resource group Access tasks can be performed normally, the impact of management devices on business processes is reduced, the management side and business side are decoupled, and the stability and reliability of storage devices are effectively improved.

In some embodiments, after configuring multiple sets of operating systems based on multiple hardware resource groups, the management device pulls up each operating system, that is, sets each operating system to a running state, so that each operating system can receive and Execute access tasks.

In order to facilitate the understanding of the process of the above-mentioned storage device performing the access task, based on the implementation environment provided in FIG. 1 and the architecture of the storage device provided in FIG. The storage method provided by the embodiment of this application is described. Fig. 5 is a flow chart of a storage method provided by an embodiment of the present application. Referring to Fig. 5, the method includes:

501. A network card of a storage device sends an access task from a target server to a first operating system running on a first hardware resource group.

For the storage device in the embodiment of the present application, reference is made to the architecture provided in FIG. 3 , and details are not described here.

In some embodiments, the network card of the storage device forwards the received access task to the operating system specified by the target server. For example, the network card of the storage device reads the information of the operating system from the received access task, and based on the information of the operating system, queries the mapping relationship between the virtual function of the network card and the information of the operating system, so as to pass the corresponding virtual The function forwards the access task to the first operating system specified by the target server. In some embodiments, the information of the operating system is a serial number of the operating system, which is not limited in this embodiment of the present application.

In some embodiments, when the target server establishes a connection with the storage device, the two parties will exchange relevant information for subsequent data read and write processes. For example, the target server obtains the serial numbers of the respective operating systems from the storage device. Certainly, the target server can also store serial numbers of various operating systems in advance, which is not limited in this embodiment of the present application.

502. The first operating system reads and writes the storage unit based on the access task.

In some embodiments, the access task carries information such as the identifier of the target server and the task identifier, and the information can be used to determine the storage location of the data in the access task in the storage unit. For example, the logical unit number (logical unit number, LUN) of the storage unit where the data to be read is located, the position offset of the data to be read in the storage unit, and the like.

Optionally, the first operating system executes the access task based on multiple processes. In some embodiments, the management and control process in the first operating system queries the storage location information of the data to be read in the storage unit of the access task based on information such as the identification of the target server and the task identification in the access task; The data processing process in the first operating system reads the data to be read from the corresponding storage location in the storage unit according to the storage location information queried by the management and control process.

503. When a failure occurs, the first operating system sends failure information to the target server, where the failure information indicates that the first operating system fails.

In some embodiments, the failure may include but not limited to any of the following items: 1. The CPU core of the operating system fails, and the access task is interrupted; error, UCE), the access task is interrupted; 3, the kernel security hole or software vulnerability (bug) of the operating system, the access task is interrupted.

In the embodiment of the present application, when a failure occurs, the first operating system sends failure information to the target server, indicating that the first operating system fails, resulting in interruption of the access task. In some embodiments, the failure information indicates that the execution of the access task fails, and carries an error code to indicate information about the failure that occurred. The embodiment of the present application does not limit the form of the fault information.

The above process is described by taking the first operating system as an example in the process of executing the access task. In other embodiments, if the first operating system fails when the access task is not executed, the first operating system The operating system can send failure information to the node with which it maintains communication, indicating that the first operating system has failed. When the node maintaining communication with the first operating system receives the fault information, it will not send the access task to the first operating system until the first operating system recovers from the fault.

504. In response to receiving the fault information, the target server sends the access task to the second operating system running on the second hardware resource group.

Wherein, the second hardware resource group refers to the architecture of the storage device provided in FIG. 3 , which will not be repeated here.

In some embodiments, the fault information includes information of the second operating system. Based on this, the target server can determine to send the access task of the first operating system to the second operating system after receiving the fault information. In this example, the target server, in response to receiving the fault information, updates the information of the operating system in the access task to the information of the second operating system, so as to instruct the network card of the storage device to forward the access task to the second operating system. The second operating system executes.

In some embodiments, information of each operating system is stored in the target server, and upon receiving the fault information, the target server reads the information of other operating systems except the first operating system, and determines to take over the access task therefrom the second operating system. In some embodiments, the target server stores a takeover relationship between operating systems, and the takeover relationship indicates which operating system will take over its access task when the operating system fails. For example, if there is a takeover relationship between the first operating system and the second operating system, the target server determines to migrate the access task to the second operating system based on the takeover relationship after receiving the failure information of the first operating system. The second operating system executes.

It should be noted that the above-mentioned process of replacing the information of the operating system in the access task on the target server with the information of the second operating system occurs at the data link layer, and the application program in the upper application layer will not receive the failure information, it is impossible to perceive that the first operating system has failed. Therefore, through the above technical solution, the effect that the front end is not aware of the fault is realized, and the reliability of the device is effectively improved.

The foregoing steps 503 to 504 are illustrated by taking the target server receiving the fault information and switching the operating system for executing the access task as an example. In some embodiments, the data storage system further includes a forwarding node, which can be used to receive fault information and switch operating systems, so as to implement migration of access tasks. Optionally, the forwarding node is a switch in the data storage system. In this example, based on the fault information received from the storage device, the forwarding node updates the operating system information in the access task received from the target server, so as to forward the access task of the first operating system to the storage device's Execute in the second operating system. Optionally, the forwarding node can detect whether the operating system of the storage device is faulty through, for example, a connectivity test, so as to switch the operating system in a timely and fast manner. Through the above technical solution, the forwarding node is used to sense the fault of the operating system and switch the operating system to improve the flexibility of operating system switching in the fault scenario, thereby efficiently realizing the migration of access tasks and effectively improving the reliability of the device.

In some other embodiments, the network card of the storage device receives the fault information returned by the first operating system, and switches the operating system that executes the access task based on the fault information. In this example, when the network card receives the fault information, the front-end can continue to receive access tasks from the target server (or forwarding node) to maintain business continuity, and based on the fault information, the back-end switch An interface for forwarding the access task, so as to forward the access task to the second operating system for execution. Through the above technical solution, service continuity can be further maintained, so as to improve the reliability of equipment.

505. The second operating system receives the access task from the target server, and reads and writes the storage unit based on the access task.

For this step, refer to steps 501 to 502, and details are not repeated here.

In some embodiments, the operating system switching process described in the above step 503 to step 504 can reduce the service interruption time caused by the fault from 2 to 10 minutes to 1 to 3 seconds, and realize the second-level switching. The front-end can eliminate the fault within a short period of time when the fault cannot be detected, and can maintain business continuity, significantly improving the reliability of the equipment.

Through the above technical solution, the first operating system can timely inform the target server of the interruption of the access task through the fault information, and realize the switching of the operating system at the data link layer, thereby realizing the migration of the access task, and using the front-end without The sensing method ensures service continuity and effectively improves device reliability.

In order to facilitate the understanding of the above-mentioned process of switching from the first operating system to the second operating system, the embodiment of the present application provides a schematic diagram of the switching process of the operating system, see FIG. 6, wherein the access task A and the access task A sent by the target server B is distributed to the first operating system 620 and the second operating system 630 for execution through the front-end network card 610, and the first operating system 620 and the second operating system 630 perform operations on the storage unit through the back-end network card 650 based on the received access tasks respectively. read and write. When the first operating system 620 fails, the front-end network card 610 sends the access task A to the second operating system 630 for execution. The storage unit is read and written by the second operating system 630 based on the access task A and the access task B. Wherein, the first operating system and the second operating system run a management process and a data processing process (refer to the description in step 502).

506. If the fault recovers, the first operating system sends a second takeover request to the target server, where the second takeover request indicates to take over the access task that has been migrated to the second operating system.

In some embodiments, the second takeover request carries the information of the first operating system and the information of the second operating system, so as to instruct the target server to resend the access task of the first operating system to the first operating system implement. For example, the second takeover request carries the operating system serial number of the first operating system and the operating system serial number of the second operating system serial number.

In some embodiments, the method for troubleshooting by the first operating system may include any of the following items: 1. Manual replacement of faulty components; 2. Resetting and restarting within the fault domain; 3. Memory error checking and correction functions. Understandably, the hardware resource group of the first operating system is isolated from the hardware resource groups of other operating systems. Therefore, the first operating system performs troubleshooting only in its own fault domain without affecting to normal operation of other operating systems. For example, when the CPU core of the first operating system fails, since the CPU cores occupied by each operating system are different, replacing the faulty CPU core of the first operating system will not affect the cores of other operating systems. For another example, UCE occurs in the memory of the first operating system. Since the memory occupied by each operating system is different, the process of isolating UCE in the memory occupied by the first operating system for troubleshooting will not affect the Other operating systems use their corresponding memory normally.

In some embodiments, after troubleshooting, the first operating system sends a recovery message to the management device. When the management device receives the restoration message of the first operating system, it determines that the first operating system has recovered from a failure, and restarts the first operating system, that is, resets the operating system to a running state. In some other embodiments, after troubleshooting, the first operating system restores to the running state by reading the persistent configuration information from the memory, and sends a recovery message to the management device to inform the management device of recovery from its own failure. Based on this, the first operating system sends a second takeover request to the target server after returning to the running state.

Through the above technical solution, the faults of each operating system are guaranteed to be isolated from each other, so that the second operating system that takes over its access tasks can run normally during the troubleshooting process of the first operating system, and the reliability of the device is effectively improved.

507. The target server sends the access task of the first operating system to the first operating system in response to receiving the second takeover request.

For this step, refer to step 504, which will not be repeated here.

508. The first operating system receives the access task from the target server, and reads and writes the storage unit based on the access task.

For this step, refer to step 505, which will not be repeated here.

Further, through the above technical solution, based on different parts of the hardware resources of the storage device, multiple sets of mutually isolated operating systems are run separately, so that the fault domain of the device is reduced to a single operating system, and the fault isolation between each operating system is ensured. At the same time, the time required for troubleshooting is greatly shortened, and the efficiency of troubleshooting is improved, thereby effectively improving the reliability of the equipment.

In the above-mentioned embodiment corresponding to FIG. 5 , the process from step 503 to step 505 is to illustrate the switching mode in which the first operating system that has failed actively reports the failure to the target server, and the target server switches the operating system that executes the access task. In some embodiments, the multiple sets of operating systems are aware of each other by regularly sending heartbeat messages, and then when any operating system fails, other operating systems can take the initiative to initiate a takeover and switch the operating system to achieve The migration of the current access task, that is, the above steps 503 to 505 can be replaced by the following steps 701 to 704 . Fig. 7 is a flow chart of a storage method provided by the embodiment of the present application. Referring to Fig. 7, the method includes:

701. If the second operating system does not receive a heartbeat message from the first operating system within a first period of time, determine that the first operating system is faulty.

In the embodiment of the present application, the heartbeat message refers to a message sent regularly by the sender, and the receiver can determine that the sender is in the running state after receiving the heartbeat message.

In step 701, the first operating system and the second operating system can send and receive heartbeat messages to detect whether the other is running. In some embodiments, the management device can also determine the running status of each operating system by sending and receiving heartbeat messages, so that it can determine the operating status of the first operating system if no heartbeat message from the first operating system is received within the second An operating system malfunctions.

702. When the first operating system fails, the second operating system sends a first takeover request to the target server, where the first takeover request indicates to take over the access task of the first operating system.

In some embodiments, the first takeover request carries information of the first operating system and information of the second operating system, so as to instruct the target server to send the access task of the first operating system to the second operating system for execution. For example, the first takeover request carries the operating system serial number of the first operating system and the operating system serial number of the second operating system.

In some embodiments, the storage device is provided with an arbitration unit. When multiple second operating systems initiate a takeover request for the first operating system, the arbitration unit can, according to preset rules, select from multiple second operating systems Determine the target second operating system to take over the access task in. Wherein, the process of determining the target second operating system according to preset rules may be: according to the hardware resource configuration or the current operation status of each operating system, determine the second operating system with the best resource configuration or the most stable operation status as the target second operating system. For the operating system, the embodiment of this application does not limit the preset rules.

It can be understood that each operating system communicates with the target server through the network card. Therefore, before the network card forwards the request of each operating system, the arbitration unit can process the takeover request of each operating system to resolve possible conflicts. Based on this, the arbitration unit selects an appropriate operating system to take over the access task, which can further guarantee the reliability of the device.

703. The target server sends the access task of the first operating system to the second operating system in response to receiving the first takeover request.

For this step, refer to step 504, which will not be repeated here.

704. The second operating system receives and executes the access task from the target server.

For this step, refer to step 505, which will not be repeated here.

In some embodiments, the first operating system sends a heartbeat message to other operating systems and the management device after troubleshooting. When the management device receives the heartbeat message of the first operating system again, it determines that the first operating system has recovered from the failure, and restarts the first operating system, that is, re-sets the operating system to the running state . In other embodiments, after troubleshooting, the first operating system restores to the running state by reading the persistent configuration information from the memory, and sends a heartbeat message to other operating systems and management devices to inform the receiver Self-failure recovery.

Further, through the above technical solution, each operating system can perceive each other, so that any operating system In the event of a failure, the non-faulty operating system can respond quickly and actively initiate a takeover request to ensure business continuity in a flexible manner, effectively improving the reliability of the device and improving the availability of the device.

It should be noted that the information (including but not limited to user equipment information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application, All are authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions. For example, the data involved in the access tasks in this application are all obtained under full authorization.

In this application, the terms "first" and "second" are used to distinguish the same or similar items with basically the same function and function. It should be understood that "first", "second" and "nth" There are no logical or timing dependencies, nor are there restrictions on quantity or order of execution. It should also be understood that although the following description uses the terms first, second, etc. to describe various elements, these elements should not be limited by the terms. These terms are only used to distinguish one element from another. For example, a first operating system could be termed a second operating system, and, similarly, a second operating system could be termed a first operating system, without departing from the scope of the various described examples. Both the first operating system and the second operating system may be operating systems, and in some cases, separate and distinct operating systems.

The meaning of the term "at least one" in this application refers to one or more, and the meaning of the term "multiple" in this application refers to two or more, for example, multiple operating systems refer to two or more operating system.

The above description is only the specific implementation of the application, but the scope of protection of the application is not limited thereto. Any person familiar with the technical field can easily think of various equivalent modifications within the technical scope disclosed in the application. Or replacement, these modifications or replacements should be covered within the protection scope of this application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

In the above embodiments, all or part of them may be implemented by software, hardware, firmware or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a program product. The program product includes one or more program instructions. When the program instructions are loaded and executed on the computing device, all or part of the processes or functions according to the embodiments of the present application will be generated.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above-mentioned embodiments can be completed by hardware, and can also be completed by instructing related hardware through a program. The program can be stored in a computer-readable storage medium. The above-mentioned The storage medium can be read-only memory, magnetic disk or optical disk and so on.

As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still understand the foregoing The technical solutions described in each embodiment are modified, or some of the technical features are replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the various embodiments of the application.

Claims

A storage device, characterized in that the storage device includes a management device and hardware resources, the hardware resources are divided into multiple hardware resource groups, each of the hardware resource groups runs a set of operating systems; the management device Used for:

When a first hardware resource group in the multiple hardware resource groups fails, migrating an access task executed by the first operating system running on the first hardware resource group to a second hardware resource in the multiple hardware resource groups group, the second operating system run by the second hardware resource group executes the access task.
The device according to claim 1, wherein specifications of the first hardware resource group and the second hardware resource group are the same.
The device according to claim 1 or 2, wherein the management device is further configured to: monitor the states of the multiple hardware resource groups.
The device according to claim 1 or 2, wherein the storage device further includes a network card, the network card supports a single-root input-output virtualization SR-IOV function, and the first virtual function VF of the network card is allocated to the The first hardware resource group, the second virtual function VF of the network card is allocated to the second hardware resource group.
The device according to claim 1, wherein the management device is further configured to receive fault information sent by the first hardware resource group; the fault information indicates that the first operating system is faulty.
The device according to claim 1, wherein the management device is further configured to receive a first takeover request sent by the second operating system, the first takeover request is used to take over execution of the first operating system access tasks.
The device according to claim 6, wherein the second operating system is configured to determine that a heartbeat message from the first operating system has occurred if no heartbeat message from the first operating system is received within a first duration Fault.
The device according to claim 1, wherein the management device is further configured to receive a second takeover request sent after the first operating system fails to recover, the second takeover request indicates that the takeover has been migrated to the Said access task for the second operating system.
A storage method, characterized in that it is applied to a storage device, the storage device includes a management device and hardware resources, the hardware resources are divided into multiple hardware resource groups, each of the hardware resource groups runs a set of operations system; the method comprising:

When a first hardware resource group in the multiple hardware resource groups fails, migrating an access task executed by the first operating system running on the first hardware resource group to a second hardware resource in the multiple hardware resource groups group, the second operating system run by the second hardware resource group executes the access task.
The method according to claim 9, wherein the specifications of the first hardware resource group and the second hardware resource group are the same.
The method according to claim 9 or 10, characterized in that the method further comprises:

The management device monitors the states of the multiple hardware resource groups.
The method according to claim 9 or 10, wherein the storage device further includes a network card, the network card supports a single-root input-output virtualization SR-IOV function, and the first virtual function VF of the network card is allocated to the The first hardware resource group, the second virtual function VF of the network card is allocated to the second hardware resource group.
The method according to claim 9, characterized in that the method further comprises:

The management device receives the fault information sent by the first hardware resource group; the fault information indicates that the first operating system is faulty.
The method according to claim 9, characterized in that the method further comprises:

The management device receives a first takeover request sent by the second operating system, where the first takeover request is used to take over the access task executed by the first operating system.
The method according to claim 14, characterized in that the method further comprises:

If the second operating system does not receive a heartbeat message from the first operating system within a first time period, it is determined that the first operating system is faulty.
The method according to claim 9, characterized in that the method further comprises:

The management device receives a second takeover request sent after the first operating system recovers from a failure, and the second takeover request indicates to take over the access task that has been migrated to the second operating system.
A computing device, characterized in that the computing device includes a management device and hardware resources, the hardware resources include a plurality of hardware resource groups, the hardware resources include a processor and a memory, and the memory is used to store the management At least one piece of program code corresponding to the device and each of the hardware resource groups, the at least one piece of program code is loaded by the processor and executes the storage method according to any one of claims 9 to 16.
A computer-readable storage medium, characterized in that the computer-readable storage medium is used to store at least one piece of program code, and the at least one piece of program code is used to execute any one of claims 9 to 16. storage method.
A computer program product, characterized in that, when the computer program product is run on a computer, the computer is made to execute the storage method according to any one of claims 9 to 16.