Storage area management method, device and computer equipment

Abstract

This application relates to a storage area management method, apparatus, and computer device. Upon detecting a preset degradation trigger event, the method selects storage units from among the created storage units that will be degraded to the corresponding preset degradation trigger event. Each storage unit includes hard disk objects belonging to different hard drives, including data disks and / or system disks. The data disks belong to the same storage pool, and the system disks include at least one system disk from each available controller. Each hard disk object stores the same data copy. A degradation identifier is added to each storage unit to be degraded. The degradation identifier is used to trigger a response to a business request, releasing the storage unit to be degraded and requesting a new storage unit based on the maximum number of data copies currently achievable. Selecting storage units to be degraded and adding degradation identifiers avoids the data security degradation caused by directly removing the corresponding data copies; furthermore, it eliminates the need for real-time replacement of storage units, thus balancing data security and system operating efficiency.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a storage area management method, apparatus, and computer device. Background Technology

[0002] Data backup involves creating multiple copies of data and storing them on different media or in different locations to mitigate risks of data loss, corruption, or tampering, ultimately ensuring data availability and integrity. For example, multiple copies of data can be stored on different data disks and system disks.

[0003] However, different carriers or locations where data copies have already been stored may malfunction or change during subsequent use, making it impossible to continue storing the data copies. In traditional technologies, the copy data stored at the different carriers or locations that have changed is usually removed. This reduces the number of data backups and lowers data security. Summary of the Invention

[0004] Therefore, it is necessary to provide a storage area management method, apparatus, and computer equipment that can improve data security in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a storage area management method, including:

[0006] Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0007] Add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0008] In the above embodiments, on the one hand, by detecting preset degradation trigger events to filter storage units to be degraded and adding degradation flags, it is possible to trigger storage unit replacement operations based on the current maximum number of data replicas, ensuring that data replicas always maintain the maximum redundancy that the system can provide. Compared with traditional technologies, when the storage carrier fails or changes, the corresponding data replica is directly removed, avoiding the problem of data security degradation caused by replica removal. On the other hand, by using the flag-triggered method, there is no need to replace storage units in real time, reducing the immediate impact on system operation and balancing data security and system operating efficiency.

[0009] In one embodiment, the detection of a preset degradation trigger event includes any of the following:

[0010] A redundancy anomaly recovery event was detected; wherein, the redundancy anomaly recovery event includes the redundancy level being restored from disk-level redundancy to frame-level redundancy, or the number of hard drives being increased when the number of hard drives is less than a preset redundancy level;

[0011] A hard drive failure event was detected;

[0012] A disk object reclamation event was detected.

[0013] In the above embodiments, by clearly defining the specific types of preset degradation trigger events, the storage system can accurately identify abnormal scenarios that need to be handled, avoid indiscriminately triggering the storage unit degradation process, and reduce the occupation of system resources by invalid operations.

[0014] In one embodiment, the step of filtering the storage units to be downgraded corresponding to the preset downgrade trigger event from the existing storage units when a preset downgrade trigger event is detected includes any one of the following:

[0015] In the event of a redundancy anomaly recovery event, the storage units corresponding to the redundancy anomaly events are selected from the existing storage units to obtain the storage units to be downgraded.

[0016] In the event of a hard disk failure, storage units that include the hard disk objects from the failed hard disk are selected from all created storage units to obtain the storage units to be downgraded.

[0017] Upon detecting a disk object reclamation event, a preset number of storage units, including the disk objects to be reclaimed, are selected from all created storage units to obtain storage units to be downgraded; wherein, the preset number is the number of storage units that meet the reclamation requirements of the disk object reclamation event.

[0018] In the above embodiments, differentiated filtering logic is designed for different preset degradation trigger events, which can accurately locate the storage units that need to be degraded and avoid the waste of system resources caused by an excessively large filtering range.

[0019] In one embodiment, the method further includes:

[0020] In response to a data write request, a storage unit is created based on a preset redundancy and currently available hard disks, and a data copy of the data corresponding to the data write request is written to each hard disk object of the created storage unit;

[0021] If a redundant abnormal event is detected, the corresponding storage unit and the redundant abnormal event are recorded.

[0022] In the above embodiments, when responding to a data write request, storage units are dynamically created based on preset redundancy and available hard disks to ensure that the creation of storage units meets system reliability requirements; at the same time, redundancy anomalies that exist during creation are recorded so as to accurately identify storage units that need to increase redundancy based on redundancy anomalies.

[0023] In one embodiment, when the redundancy anomaly event is that the number of hard disks is less than the preset redundancy, the storage unit created for the corresponding record and the redundancy anomaly event include any one of the following:

[0024] When the created storage unit is a first-type storage unit, an event identifier for the redundant exception event is added to the metadata of the created storage unit; wherein, the metadata is stored in the preset memory of the controller to which the created storage unit belongs; the first-type storage unit is a storage unit that only includes a data disk;

[0025] When the created storage unit is a second type of storage unit, the created storage unit is recorded in the storage unit list corresponding to the redundancy exception event; wherein, the different storage unit lists are used to record created storage units with an actual data disk number less than a different first number and created storage units with an actual system disk number less than a different second number; wherein, the first number is not greater than the target data disk number corresponding to the preset redundancy, the second number is not greater than the target system disk number corresponding to the preset redundancy, and the second type of storage unit is a storage unit that includes at least a system disk.

[0026] In the above embodiments, a differentiated redundant abnormal event recording method is designed based on the different characteristics of the first type of storage unit and the second type of storage unit. This not only ensures the accuracy of the recording and the efficiency of the query, but also adapts to the functional positioning of different storage units, reduces the system resource consumption during the recording and querying process, and provides a query basis for the subsequent screening of different types of storage units, thereby improving the query efficiency of storage units.

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

[0028] In response to a hard disk deletion command, storage units containing the target hard disk object in the hard disk to be deleted indicated by the hard disk deletion command are selected from each created storage unit to obtain the storage unit to be processed;

[0029] If the number of hard disk objects in the storage unit to be processed is greater than a preset number threshold, the target hard disk objects in the storage unit to be processed are released; wherein, the preset number threshold is not less than 2;

[0030] If the number of hard disk objects in the storage unit to be processed is not greater than a preset threshold, the degradation flag is added to the storage unit to be processed.

[0031] In the above embodiments, when deleting a hard drive, the storage units to be processed are screened and dynamically processed according to the number of kObj objects. This avoids the storage unit replica count being too low due to direct hard drive deletion: when the number of kObj objects is sufficient, the target kObj is released directly, ensuring that the storage unit still meets the basic redundancy requirements; when the number of kObj objects is insufficient, a degradation flag is added to trigger the generation of new storage units, ensuring that the data replica count is restored to the maximum. This differentiated processing method not only realizes resource recovery during hard drive deletion but also avoids data reliability degradation, solving the problem of insufficient replica count that is easily caused by traditional hard drive deletion, thus balancing resource management and data security.

[0032] In one embodiment, the method further includes:

[0033] Upon receiving a service request, the current storage unit corresponding to the service request is determined from among the existing storage units.

[0034] If the current storage unit is marked with the degradation flag, release the current storage unit and request a target storage unit that is not marked with the degradation flag;

[0035] Write a copy of the business data stored in the current storage unit to each hard disk object of the target storage unit, and execute the business request based on the target storage unit.

[0036] In the above embodiments, the service request triggers the replacement of the degraded storage unit to ensure that the service always uses the storage unit that is not degraded and meets the maximum redundancy, so as to avoid data loss or read failure caused by the service accessing the storage unit with insufficient reliability.

[0037] In one embodiment, the method further includes:

[0038] At each query time, select the storage units marked with the degradation flag from the created storage units to be released;

[0039] Release the storage unit to be released, and request a corresponding storage unit that is not marked with the downgrade identifier;

[0040] Write a copy of the data stored in the current storage unit to the corresponding storage unit.

[0041] In the above embodiments, the timed query mechanism can proactively replace the storage units to be released with reliable units that provide the maximum redundancy that the storage system can offer, thereby improving data security.

[0042] Secondly, this application also provides a storage area management device, comprising:

[0043] The filtering module is used to filter the storage units to be degraded from each created storage unit when a preset degradation trigger event is detected; wherein each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0044] An add module is used to add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0045] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0046] Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0047] Add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0048] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0049] Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0050] Add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0051] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0052] Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0053] Add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0054] The aforementioned storage area management method, apparatus, and computer equipment, upon detecting a preset degradation trigger event, select storage units to be degraded from among the created storage units corresponding to the preset degradation trigger event; wherein, each storage unit includes hard disk objects belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy; a degradation identifier is added to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data copies currently achievable. The above solution, on the one hand, filters storage units to be degraded by detecting preset degradation trigger events and adds degradation tags, which can trigger storage unit replacement operations based on the current maximum number of data replicas, ensuring that data replicas always maintain the maximum redundancy that the system can provide. Compared with traditional technologies, which directly remove the corresponding data replicas when the storage carrier fails or changes, this avoids the problem of reduced data security caused by replica removal. On the other hand, by using a tag-trigger method, there is no need to replace storage units in real time, reducing the immediate impact on system operation and balancing data security and system operating efficiency. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart illustrating a storage area management method in one embodiment;

[0057] Figure 2 This is a schematic diagram of a storage unit update in one embodiment;

[0058] Figure 3 This is a flowchart illustrating the execution of a business request in one embodiment;

[0059] Figure 4 This is a sequence diagram of executing a service request in one embodiment;

[0060] Figure 5 This is a schematic diagram of the process for replacing a storage unit to be released in one embodiment;

[0061] Figure 6 This is a flowchart illustrating a storage area management method in another embodiment;

[0062] Figure 7 This is a structural block diagram of a storage area management device in one embodiment;

[0063] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0065] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0066] The storage area management method provided in this application can be applied to application environments that store multiple copies of data. This method can be executed by a server or a terminal. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, projection devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc.

[0067] In one exemplary embodiment, such as Figure 1 As shown, a storage area management method is provided. Taking the application of this method to a server as an example, the method includes the following steps:

[0068] S101, when a preset degradation trigger event is detected, select the storage unit to be degraded from each created storage unit corresponding to the preset degradation trigger event.

[0069] Each storage unit comprises disk objects belonging to different hard drives. A disk object, abbreviated as kObj, refers to a specific storage space allocated to the storage unit on a single hard drive. All disk objects within the same storage unit store identical data copies. Different hard drives include data disks and / or system disks. Data disks belong to the same storage pool, and system disks include at least one system disk on each available controller. A storage unit is a logical unit used to store data copies, including ordinary storage units (kSeg) and special storage units (sysSeg). Ordinary storage units only include data disk disk objects, while special storage units contain system disks and / or data disk disk objects. Data disks are used to store user data and metadata bound to the storage pool's lifecycle; they do not run an operating system. System disks are hard drives running the storage array's operating system. A portion of storage space is reserved for sysSeg disk objects, which are evenly distributed across the controllers, providing disk-level redundancy.

[0070] Among them, kSeg's redundancy depends on the distribution of data disks (such as across data disk frames), and the risk of failure is concentrated in the data disk cluster; while sysSeg's kObj at least includes a system disk, which is evenly distributed on each available controller. sysSeg can also add additional data disk redundancy, which is equivalent to having "controller-level redundancy (system disk across controllers) + data disk-level redundancy (optional)" at the same time. This makes the replica distribution more dispersed and the data security higher.

[0071] SysSeg's disk objects are evenly distributed across the controllers. This can be understood as follows: if data copies need to be stored on four system disks, and each controller has two system disks, then in a storage array with two controllers, each controller needs to provide two system disks. In a storage array with four controllers, each controller needs to provide one system disk.

[0072] The preset degradation trigger event refers to the event that triggers the degradation of the storage unit. Storage unit degradation refers to the event that may cause changes to the disk objects in the storage unit, such as the addition, removal or change of disk objects in the storage unit.

[0073] Storage units awaiting degradation refer to storage units whose reliability is insufficient due to preset degradation trigger events and which require subsequent replacement.

[0074] For example, the main management module (Mgr) of the storage system can receive event notifications from the underlying disk management module and disk service management module (DSM). These event notifications include notifications of events such as the addition, reduction, or change of disk objects in the storage unit. For example, when a disk fails, the DSM sends a failure notification to the Mgr, indicating that a preset degradation trigger event has been detected.

[0075] The disk management module is a disk resource management software module deployed in the server operating system. It is used to manage the physical space of the disk (such as dividing the kObj storage area), monitor the disk operation status (such as the number of free kObj and read / write health), perform the allocation and reclamation of kObj, provide resource support for the creation of storage units (kSeg or sysSeg), and at the same time report disk status changes (such as insufficient kObj) to the Mgr.

[0076] Mgr is a software module deployed on the server processor and integrated into the server's operating system process. The server runs a computer program containing Mgr logic, which enables Mgr to perform operations such as detecting trigger events, filtering storage units to be degraded, managing the list of storage units, and coordinating Agents to mark downgrades.

[0077] The Disk Service Management (DSM) module is a disk event and resource coordination software module deployed on the server. It is used to aggregate hard disk failure information (such as hard disk read / write failures), send event notifications (such as pushing hard disk failure events and redundancy anomaly recovery events to the Mgr), and coordinate the scheduling of hard disk resources among various modules.

[0078] Furthermore, the Manager can iterate through the created storage units and, based on the type of the triggering event, filter out the storage units that meet the criteria by querying the disk objects of the storage units associated with the triggering event. For example, if the preset degradation triggering event is a disk failure, then the storage unit containing the disk object from the failed disk will be used as the storage unit to be degraded corresponding to the preset degradation triggering event.

[0079] S102, add a degradation identifier to each storage unit to be degraded.

[0080] The degradation flag is a status marker used to indicate a storage unit to be degraded, stored in the storage unit's metadata or the controller's reserved memory. The degradation flag triggers a response to a business request, releasing the storage unit to be degraded and requesting a new storage unit based on the maximum number of data replicas currently achievable. Before releasing the storage unit to be degraded, the data in it needs to be read out so that it can be written to the new storage unit.

[0081] For example, the Mgr can send the identifiers of the storage units to be downgraded in batches to the agent modules of each controller. The agent writes the downgrade identifier into the metadata of the corresponding storage unit and stores the metadata in the controller's preset reserved memory, thus completing the identifier addition.

[0082] For example, a storage pool P1 in a storage system is configured for frame-level redundancy, meaning it requires four disk objects from four different disk frames. Due to the failure of two disks, storage pool P1 is forced to degrade to disk-level redundancy, for example, having only two valid disk objects. All kSegs created during the period of the two disk failures carry a "frame over-redundancy failure flag". Subsequently, after the failed disks are repaired, the storage pool regains frame-level redundancy. The underlying module sends a "redundancy anomaly recovery event" notification to the Manager, indicating that a preset degradation trigger event has been detected. The Manager iterates through all created kSegs, selecting those with the "frame over-redundancy failure flag" as the storage units to be degraded. It then sends a batch notification to the Agents of each controller, who add a degradation flag to the metadata of these kSegs. When a business module in the server accesses a kSeg, it detects the degradation flag, releases the old kSeg, and requests a new kSeg that meets frame-level redundancy requirements, ensuring data reliability.

[0083] In the above embodiments, on the one hand, by detecting preset degradation trigger events to filter storage units to be degraded and adding degradation flags, it is possible to trigger storage unit replacement operations based on the current maximum number of data replicas, ensuring that data replicas always maintain the maximum redundancy that the system can provide. Compared with traditional technologies, when the storage carrier fails or changes, the corresponding data replica is directly removed, avoiding the problem of data security degradation caused by replica removal. On the other hand, by using the flag-triggered method, there is no need to replace storage units in real time, reducing the immediate impact on system operation and balancing data security and system operating efficiency.

[0084] In some optional implementations, the detected preset degradation trigger event can be any of the following:

[0085] A redundancy anomaly recovery event was detected. This includes situations where the redundancy level is restored from disk-level redundancy to frame-level redundancy, or where the number of hard drives is increased by repairing or adding new hard drives when the number of hard drives is less than the preset redundancy level.

[0086] For example, when the disk management module detects that the number of disks has increased to meet the preset redundancy, or when the redundancy level has been restored from the disk level to the frame level, it sends a status notification to the DSM. The DSM then sends a "redundancy anomaly recovery event" notification to the Mgr, and the Mgr determines that the event has been detected.

[0087] For example, a storage system has a preset redundancy level corresponding to a target number of 4 hard drives. If 3 hard drives go offline, the number of available hard drives becomes 1. After repairing 2 hard drives, the number of available hard drives reaches 3. The hard drive management module sends a hard drive number increase notification to the DSM, and the DSM sends a "redundancy anomaly recovery event" notification to the Mgr.

[0088] Alternatively, a hard drive failure event is detected. A hard drive failure event refers to an event where the hard drive is unable to provide kObj normally due to physical damage, read / write failures, or other reasons. This event can be monitored by the DSM module and the Mgr can be notified.

[0089] For example, the DSM module can monitor the hard drive's operating status in real time. When it detects hard drive read / write failures, physical damage, or other issues, it immediately sends a "hard drive failure event" notification to the Mgr, which then determines that the event has been detected.

[0090] For example, if a hard drive numbered HD-003 in the storage system fails to read or write, the DSM will immediately send a "hard drive failure event" notification to the Mgr after detecting the failure, clearly identifying the faulty hard drive.

[0091] Alternatively, a disk object reclamation event may be detected. A disk object reclamation event refers to an event in which some disks in the storage system have exhausted their kObj resources, requiring the reclamation of kObj resources already allocated to storage units to meet new storage demands. This event can be notified to the Manager by the disk management module.

[0092] For example, the disk management module can count the number of free kObj objects for each disk. When a disk has insufficient free kObj objects and needs to allocate kObj objects for new storage needs, it sends a "disk object reclamation event" notification to the Manager to specify the reclamation need. The Manager determines that the event has been detected.

[0093] For example, all the kObj objects on the hard drive numbered HD-005 in the storage system have been allocated to each kSeg. The hard drive management module needs to reclaim two kObj objects from this hard drive to allocate to new storage requests. It sends a "hard drive object reclamation event" notification to the Manager, specifying the hard drive to which the kObj objects to be reclaimed belong and the number of objects to be reclaimed.

[0094] In the above embodiments, by clearly defining the specific types of preset degradation trigger events, the storage system can accurately identify abnormal scenarios that need to be handled, avoid indiscriminately triggering the storage unit degradation process, and reduce the occupation of system resources by invalid operations.

[0095] For example, before describing how to filter the storage units to be degraded from the existing storage units after detecting a preset degradation trigger event, we will first explain the purpose of filtering the storage units to be degraded. Several scenarios will be used as examples below.

[0096] In the first scenario, within the storage system, the reliability of kSeg frame-level redundancy is higher than that of disk-level redundancy. If, at a certain point, the storage system's kSeg redundancy strategy degrades from high-reliability frame-level redundancy to low-reliability disk-level redundancy, from that moment on, all newly generated kSegs will be disk-level redundant with limited reliability. If, at some point in the future, the storage system supports reverting the kSeg redundancy strategy to frame-level redundancy (i.e., a preset degradation trigger event is detected), these disk-level redundant kSegs must regain their reliability and become frame-level redundant. Therefore, upon detecting the preset degradation trigger event, it is necessary to select the storage units to be degraded from all created storage units, replacing them with new storage units that offer a higher security level and meet the maximum data replication storage requirements.

[0097] In the second scenario, the number of hard drives in the storage system may be less than the expected redundancy of kSeg, making it impossible to allocate a full set of kSeg replicas. In this case, the Manager will allocate downgraded kSeg replicas. For example, if the expected number of kSeg replicas was n, but there are only m hard drives (n>m), only m replicas of kSeg can be allocated. Once the number of hard drives in the system recovers to n, or increases to i (n>i>m), a pre-defined degradation trigger event is detected. At this point, the m replicas of kSeg need to be restored to n replicas or upgraded to i replicas to meet the maximum reliability requirement. Therefore, upon detecting a pre-defined degradation trigger event, it is necessary to select the storage units to be degraded from all created storage units to replace the degraded storage units with new storage units that meet the maximum data replica storage requirements.

[0098] In the third scenario, a hard drive failure occurs in the storage system, triggering a pre-defined degradation event. In this case, the kSeg containing a copy of the failed hard drive lacks reliability and requires timely adjustment of its replica data. The failed hard drive copy should no longer be used, while simultaneously ensuring maximum redundancy for the kSeg. Therefore, upon detecting a pre-defined degradation event, it's necessary to select the corresponding storage unit from among the existing storage units to be degraded. This degraded storage unit should be replaced with a new one that meets the maximum data replication requirements.

[0099] In the fourth scenario, within the storage system, for some reason, it's necessary to reclaim some kObj copies on the hard drive. This occurs when a pre-defined degradation trigger event is detected. If almost all kObj copies on the hard drive are allocated to kSegs, and there aren't enough free kObj copies, then some kObj copies on the allocated kSegs need to be reclaimed. In this case, all kSegs with copies of the kObj from that disk are unstable, and these disk copies need to be removed for reclamation, and the reliability of these kSegs needs to be restored. Therefore, upon detecting a pre-defined degradation trigger event, it's necessary to select the storage units corresponding to the pre-defined degradation trigger event from among the created storage units, so that the storage units to be degraded can be replaced with new storage units that meet the maximum data replication storage requirements.

[0100] Based on this, in some optional implementations, after detecting any one of the preset degradation trigger events, including the detection of a redundancy anomaly recovery event, a hard disk failure event, and a hard disk object reclamation event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit. This can be achieved in the following way:

[0101] In the event of a redundancy anomaly recovery event, the storage units corresponding to the redundancy anomaly events are selected from the existing storage units to obtain the storage units to be downgraded.

[0102] Redundancy anomaly events are those that cause abnormal redundancy status in the storage system, corresponding to redundancy anomaly recovery events. Redundancy anomaly events include insufficient disk quantity to the preset redundancy level and redundancy level downgrade, such as downgrading from frame-level redundancy to disk-level redundancy. Insufficient disk quantity to the preset redundancy level can be insufficient data disk quantity, insufficient system disk quantity, or insufficient data disk quantity and insufficient system disk quantity.

[0103] For example, after receiving a redundancy anomaly recovery event notification, the Manager queries the redundancy anomaly event record corresponding to the redundancy anomaly recovery event, such as the marked "box over-redundancy fault flag" or "disk copy insufficiency flag", traverses all created storage units, and filters out the storage units with the redundancy anomaly event identifier as the storage units to be downgraded.

[0104] For example, the storage system triggered a redundancy anomaly event due to insufficient hard disks, during which all kSegs created were marked with an "insufficient disk replicas" flag. When the number of hard disks recovered, the Mgr received the redundancy anomaly recovery event notification, traversed all kSegs, and selected all kSegs marked with an "insufficient disk replicas" flag as storage units to be degraded.

[0105] Alternatively, in the event of a detected hard disk failure, storage units that include the hard disk objects from the failed hard disk can be selected from the existing storage units to obtain the storage units to be downgraded.

[0106] Among them, the hard disk object in the faulty hard disk is kObj that has been allocated to the storage unit on the faulty hard disk. This kObj cannot provide normal data storage and reading services due to the hard disk failure.

[0107] For example, after receiving a hard disk failure event notification, the Mgr obtains the identifier of the failed hard disk, traverses all created storage units, queries the hard disk information to which the kObj of each storage unit belongs, and filters out the storage units containing the kObj on the failed hard disk as storage units to be downgraded.

[0108] For example, after receiving the HD-003 hard drive failure notification, the Mgr traverses all storage units and finds that kSeg-001, kSeg-005, and sysSeg-002 all contain kObj from HD-003. Therefore, these three storage units are listed as storage units to be downgraded.

[0109] Alternatively, upon detecting a disk object reclamation event, a preset number of storage units, including the disk objects to be reclaimed, are selected from the existing storage units to obtain storage units to be downgraded; wherein, the preset number is the number of storage units that meet the reclamation requirements of the disk object reclamation event.

[0110] The disk objects to be reclaimed are kObj objects that need to be released from storage units to meet new storage demands, typically from disks where kObj has been exhausted. The preset quantity is the number of storage units that can meet the reclamation requirements of a disk object reclamation event, i.e., the total number of storage units to which the kObj objects to be released belong, calculated by the Manager based on the number of kObj objects to be reclaimed and the number of kObj objects to be reclaimed contained in each storage unit.

[0111] For example, after receiving the disk object recycling event notification, the Mgr obtains the disk identifier and recycling quantity of the kObj to be recycled, traverses all created storage units, filters out the storage units containing kObj on the disk, sorts them by creation time or usage frequency, and selects the top N (N is a preset number) storage units as storage units to be downgraded.

[0112] For example, when the Mgr receives a notification to reclaim two kObj objects on HD-005, it finds that kSeg-003, kSeg-007, and kSeg-010 all contain kObj objects from HD-005. The preset quantity N=2, meaning that two storage units are needed to release two kObj objects. Therefore, kSeg-003 and kSeg-007 are selected as storage units to be downgraded.

[0113] In the above embodiments, differentiated filtering logic is designed for different preset degradation trigger events, which can accurately locate the storage units that need to be degraded and avoid the waste of system resources caused by an excessively large filtering range.

[0114] In some optional implementations, a storage unit can be created when responding to a data write request, and in the event of a redundant exception event, the created storage unit and the redundant exception event can be recorded.

[0115] For example, in response to a data write request, a storage unit is created based on a preset redundancy and currently available hard disks, and a data copy of the data corresponding to the data write request is written to each hard disk object of the created storage unit; and if a redundancy exception event is detected, the created storage unit and the redundancy exception event are recorded accordingly.

[0116] Among them, the data write request is a request sent by the business requesting party to the server to store data, which includes the data content to be stored and the storage requirements, including redundancy level, storage pool identifier, etc.

[0117] The preset redundancy is the minimum number of data replicas that the storage system presets to ensure data reliability. It corresponds to the number of target data disks and the number of target system disks. For example, the frame-level redundancy corresponds to 4 data disk replicas, and sysSeg corresponds to 4 system disks + 4 data disk replicas.

[0118] The currently available hard drives are those that are running normally and can be allocated kObj, including data disks and system disks.

[0119] Redundancy anomalies refer to events where the redundancy status of the storage system fails to meet preset redundancy requirements, leading to a decrease in data reliability. Examples include insufficient number of hard drives, redundancy level downgrade, and missing system disk or data disk replicas. Insufficient number of hard drives specifically means that the number of available hard drives (data disks or system disks) is less than the target number of hard drives corresponding to the preset redundancy level. For example, the preset requirement is 4 data disk replicas, but only 3 are actually available. Redundancy level downgrade means that the storage pool's redundancy strategy is downgraded from high-reliability frame-level redundancy (replicas are stored in different hard drive frames) to low-reliability disk-level redundancy (replicas only come from different hard drives, and do not meet the requirement that each replica is stored in a different hard drive frame). Missing system disk or data disk replicas means that the number of system disk replicas and data disk replicas in a special storage unit (sysSeg) does not reach the preset target. For example, sysSeg is preset to have 4 system disk replicas, but only 3 are actually available.

[0120] For example, after receiving a data write request, the Mgr can obtain the information of the currently available hard disks from the hard disk management module, calculate the required number of kObj based on the preset redundancy and the currently available hard disks. If the preset redundancy is 4, then 4 different hard disks' kObj are required. The kObj are allocated from the available hard disks to form a storage unit, and then the data copy corresponding to the data write request is written into each kObj of the storage unit.

[0121] For example, during the creation of a storage unit, if the currently available hard disk does not meet the frame redundancy requirement, a frame-level redundancy fault label is generated, i.e., a redundancy exception event is generated, and the storage unit and the redundancy exception event are associated and recorded.

[0122] For example, during the creation of a storage unit, if the number of available hard disks is less than the target number of hard disks corresponding to the preset redundancy, an insufficient disk copy tag is generated, i.e., a redundancy exception event is generated, and the storage unit and the redundancy exception event are associated and recorded.

[0123] For example, a client sends a data write request to the server, requesting the storage of data A with a preset redundancy of 4, requiring 4 kObj disks. The manager checks the number of available data disks and finds 3, which is less than the target of 4, indicating a redundancy anomaly. In this case, kObj disks can be allocated from the 3 available disks to form kSeg-012, and copies of data A can be written to these 3 kObj disks. Simultaneously, the manager adds a "disk copy insufficiency flag" to the metadata of kSeg-012, completing the redundancy anomaly event recording.

[0124] In the above embodiments, when responding to a data write request, storage units are dynamically created based on preset redundancy and available hard disks to ensure that the creation of storage units meets system reliability requirements; at the same time, redundancy anomalies that exist during creation are recorded so as to accurately identify storage units that need to increase redundancy based on redundancy anomalies.

[0125] In some optional implementations, when the redundancy exception event is that the number of hard disks is less than the preset redundancy, the corresponding storage unit created and the redundancy exception event can be implemented in any of the following ways:

[0126] If the created storage unit is a Class 1 storage unit, add an event identifier for redundant exception events to the metadata of the created storage unit.

[0127] The metadata is stored in the preset memory of the controller to which the created storage unit belongs. The first type of storage unit is a storage unit that only includes a data disk, i.e., a regular kSeg, whose metadata is stored in the preset memory of the controller to which it belongs for easy access. Event identifiers are used to mark redundant abnormal events associated with the storage unit, such as "insufficient disk replicas" and "over-redundancy frame" and are written into the metadata of the first type of storage unit.

[0128] For example, when the created storage unit is a kSeg containing only data disks, if the number of hard disks is less than the preset redundancy, the Mgr adds an event identifier for a redundancy exception event, such as "insufficient disk replicas flag", to the metadata of the storage unit. The metadata is stored in the preset memory of the controller to facilitate quick subsequent queries.

[0129] For example, when creating a kSeg-015 containing only data disks, if the number of available data disks is 2 and the preset redundancy is 4, the Mgr can add a "disk copy shortage flag" to the metadata of kSeg-015. The metadata of kSeg-015 is stored in the preset memory of the controller corresponding to the storage pool where the data disks are located.

[0130] Optionally, if the created storage unit is a second type of storage unit, the created storage unit is recorded in the storage unit list corresponding to the redundancy exception event.

[0131] The different storage unit lists are used to record created storage units with an actual number of data disks less than a different first number and created storage units with an actual number of system disks less than a different second number. The first number is not greater than the target number of data disks corresponding to the preset redundancy, and the second number is not greater than the target number of system disks corresponding to the preset redundancy. The second type of storage unit is a storage unit that includes at least a system disk, i.e., sysSeg, and may additionally include data disks.

[0132] The storage unit list is specifically used to record second-category storage units with redundancy anomalies. Different lists correspond to different anomaly types, such as the actual number of data disks being less than the first number, or the actual number of system disks being less than the second number. The first number is a value that is no greater than the target number of data disks corresponding to the preset redundancy level. For example, if the target number of data disks is 4, the first number could be 3, 2, etc. The second number is a value that is no greater than the target number of system disks corresponding to the preset redundancy level. For example, if the target number of system disks is 4, the second number could be 3, 2, etc.

[0133] For example, when the created storage unit is a sysSeg containing a system disk, the Mgr records the storage unit in the corresponding storage unit list according to the type of redundant exception event, such as the actual number of data disks being less than the first number or the actual number of system disks being less than the second number. Different lists are used to classify and manage different types of exceptions, which facilitates accurate filtering in the future.

[0134] For example, when creating sysSeg-008 containing system disks and data disks, assuming the preset redundancy is 4 system disks and 4 data disks, the number of available data disks is 3, the first number is 4, the number of available system disks is 2, the second number is 4 and 3; the Mgr records sysSeg-008 into three storage unit lists: "number of system disks less than 4", "number of system disks less than 3", and "number of data disks less than 4".

[0135] Furthermore, when using the storage unit list, if a new system disk is added to the storage system, all storage units in the storage unit list with "less than 3 system disks" can be marked with a degradation flag, which is stored in the controller's preset reserved memory. Additionally, all storage units in the storage unit list with "less than 3 system disks" are designated as storage units to be degraded, and all storage units in the storage unit list with "less than 3 system disks" are removed from that list. Moreover, these storage units to be degraded are removed from the storage unit list with "less than 4 system disks".

[0136] Similarly, when a new data disk is added to the storage system, all storage units in the list of storage units with "less than 3 data disks" can be marked with a degradation flag, which is stored in the controller's preset reserved memory. Furthermore, all storage units in the list of storage units with "less than 3 data disks" are designated as storage units to be degraded, and all storage units in the list of storage units with "less than 3 data disks" are removed from that list. Additionally, these storage units to be degraded are removed from the list of storage units with "less than 4 data disks".

[0137] In the above embodiments, a differentiated redundant abnormal event recording method is designed based on the different characteristics of the first type of storage unit and the second type of storage unit. This not only ensures the accuracy of the recording and the efficiency of the query, but also adapts to the functional positioning of different storage units, reduces the system resource consumption during the recording and querying process, and provides a query basis for the subsequent screening of different types of storage units, thereby improving the query efficiency of storage units.

[0138] In some alternative implementations, when deleting a disk, in order to avoid the number of data replicas being too low, the disk deletion operation must be performed only when a certain number of replicas are met.

[0139] For example, in response to a hard disk deletion command, storage units containing the target hard disk object in the hard disk to be deleted as indicated by the hard disk deletion command are selected from each created storage unit to obtain the storage units to be processed.

[0140] The hard drive deletion command is a command issued by the user or system to remove a hard drive, such as a command to remove a faulty hard drive or a retired hard drive, which includes an identifier of the hard drive to be deleted.

[0141] The target hard disk object is kObj that has been allocated to a storage unit on the hard disk to be deleted. It needs to be released from the corresponding storage unit or the storage unit needs to be downgraded.

[0142] The storage unit to be processed is the storage unit containing the target hard disk object, that is, the storage unit that needs to process the target kObj.

[0143] For example, after receiving the hard disk deletion command, the Mgr obtains the identifier of the hard disk to be deleted, traverses all created storage units, queries the hard disk information to which the kObj of each storage unit belongs, and filters out the storage units containing the target kObj on the hard disk as the storage units to be processed.

[0144] If the number of disk objects in the storage unit to be processed exceeds a preset threshold, release the target disk objects in the storage unit to be processed; and if the number of disk objects in the storage unit to be processed does not exceed the preset threshold, add a degradation flag to the storage unit to be processed.

[0145] The preset quantity threshold is the critical value for determining whether the storage unit to be processed needs to be downgraded. It can be set according to data security requirements and minimum redundancy requirements. For example, the preset quantity threshold can be set to be no less than 2 to ensure that there are still enough replicas after the target kObj is released.

[0146] For example, the McGr queries the total number of kObj in each storage unit to be processed. If the total number is greater than a preset threshold, such as 3>2, the target kObj in the storage unit is released directly, the metadata of the storage unit is updated, and the remaining kObj information is recorded. If the total number is not greater than the preset threshold, such as 2≤2, a notification is sent to the Agent to add a downgrade flag to the storage unit to be processed.

[0147] For example, in response to the instruction to delete the HD-007 hard drive, the Manager traverses the storage units and finds kSeg-018 (containing 4 kObj, one of which comes from HD-007) and kSeg-020 (containing 2 kObj, one of which comes from HD-007) as storage units to be processed. The preset quantity threshold is 2: if the total number of kObj in kSeg-018 is 4>2, the Manager releases the kObj from HD-007 and updates the metadata to 3 kObj; if the total number of kObj in kSeg-020 is 2≤2, the Manager notifies the Agent to add a downgrade flag for kSeg-020.

[0148] In the above embodiments, when deleting a hard drive, the storage units to be processed are screened and dynamically processed according to the number of kObj objects. This avoids the storage unit replica count being too low due to direct hard drive deletion: when the number of kObj objects is sufficient, the target kObj is released directly, ensuring that the storage unit still meets the basic redundancy requirements; when the number of kObj objects is insufficient, a degradation flag is added to trigger the generation of new storage units, ensuring that the data replica count is restored to the maximum. This differentiated processing method not only realizes resource recovery during hard drive deletion but also avoids data reliability degradation, solving the problem of insufficient replica count that is easily caused by traditional hard drive deletion, thus balancing resource management and data security.

[0149] In some optional implementations, when initiating a business request, the storage unit marked with a degradation flag can be replaced with a new storage unit without a degradation flag before the business request is processed, thus avoiding data loss or read failure caused by accessing storage units with insufficient reliability.

[0150] For example, see Figure 2 , Figure 2A schematic diagram of storage unit updates is provided. The Manager (Mgr) manages the metadata information of the storage unit (taking kSeg as an example). When the kSeg metadata information changes, the Mgr sends the kSeg metadata information to be updated (kSeg status, etc.) in batches to the Agent. The Agent writes this information into the kSeg metadata reserved memory in the controller (Ctrl) corresponding to the Agent. Upon receiving a business request, the business module in the server reads the kSeg metadata reserved memory, determines whether to replace the kSeg based on its status, and then executes the business request.

[0151] See Figure 3 , Figure 3 A flowchart illustrating the execution of a business request is provided, which includes the following steps:

[0152] S301, after receiving a service request, determines the current storage unit corresponding to the service request from among the existing storage units.

[0153] Here, a business request is a request issued by the business requesting client to perform read and write operations on a storage unit, containing an identifier of the required business data. The current storage unit is the storage unit corresponding to the business request, that is, the storage unit that stores a copy of the required business data.

[0154] For example, after receiving a business request, the storage unit corresponding to the business data can be queried based on the business data identifier in the request, and the current storage unit can be determined.

[0155] S302: If the current storage unit is marked with a degradation flag, release the current storage unit and request a target storage unit that is not marked with a degradation flag.

[0156] The target storage unit is a storage unit that is not marked with a degradation identifier and meets the current system's maximum redundancy requirements, and is used to replace the current storage unit.

[0157] S303 writes a copy of the business data stored in the current storage unit to each hard disk object of the target storage unit, and executes the business request based on the target storage unit.

[0158] For example, before accessing the current storage unit, the business module first reads the metadata of the current storage unit or the state in the controller's reserved memory. If a degradation flag is detected, a replacement process is triggered: that is, the current storage unit is first released, and the business module requests a target storage unit without a degradation flag from the Agent; the Agent creates the target storage unit based on the currently available hard disks and maximum redundancy; the business module writes a copy of the business data in the current storage unit into each kObj in the target storage unit, and then executes the business request based on the target storage unit.

[0159] For example, when a business module receives a request to read data A, it finds that data A is stored in kSeg-012 (with a degradation flag added). After detecting the degradation flag, the business module releases kSeg-012 and requests the target storage unit from the Agent. The Agent creates kSeg-025 based on the currently available hard drives (assuming there are 4 data disks, satisfying frame-level redundancy). The business module writes a copy of data A into the 4 kObj objects in kSeg-025, and then performs a data read operation based on kSeg-025.

[0160] For example, see Figure 4 , Figure 4 A sequence diagram for executing a business request is provided, including the following steps:

[0161] Step 1: The Mgr iterates through all kSegs and filters out the kSegs that need to be downgraded.

[0162] Step 2: The Manager sends the identifiers of the kSegs to be downgraded to the Agent in batches.

[0163] Step 3: After receiving the information, the Agent writes the "downgrade identifier" into the kSeg metadata reserved memory.

[0164] Step 4: After the Agent completes the identification writing, it returns a "Update Complete" notification to the Manager.

[0165] Step 5: The Manager receives feedback from the Agent, confirming that the downgrade flag has been successfully added.

[0166] Branch 1:

[0167] Step 6: Before reading or writing kSeg, the business module first queries the state of the kSeg metadata reserved in memory.

[0168] Step 7: If no degradation flag is detected, kSeg metadata reserved memory returns an "unmarked" status to the business module.

[0169] Step 8: The business module directly performs read and write operations on the kSeg.

[0170] Branch Two:

[0171] Step 9: If a downgrade flag is detected, the business module first requests the Manager to release the unreliable kSeg.

[0172] Step 10: The business module requests a new kSeg from the Manager, without the downgrade flag.

[0173] Step 11: The business module checks the status of the new kSeg in the kSeg metadata reserved memory (the new kSeg has no degradation indicator).

[0174] Step 12: The business module performs read and write operations on the new kSeg (at this time, a reliable unit that meets the current maximum redundancy is used).

[0175] In the above embodiments, the service request triggers the replacement of the degraded storage unit to ensure that the service always uses the storage unit that is not degraded and meets the maximum redundancy, so as to avoid data loss or read failure caused by the service accessing the storage unit with insufficient reliability.

[0176] In some optional implementations, it is also possible to periodically check whether there are any storage units marked with a degradation flag that are to be released. If so, a new storage unit is requested to replace the storage unit marked with the degradation flag that is to be released.

[0177] For example, see Figure 5 , Figure 5 A flowchart illustrating the process of replacing a memory cell to be released is provided, specifically including the following steps:

[0178] S501, at each query time, selects storage units marked with a degradation flag from the created storage units that are to be released.

[0179] The query time refers to a preset scheduled query time, such as every hour or every half day. The storage units to be released are storage units marked with a degradation indicator that need to be replaced.

[0180] For example, at each query time, the business module can automatically trigger the query process, traverse all created storage units, and filter out the storage units marked with the degradation flag as storage units to be released.

[0181] S502, release the storage unit to be released and request the corresponding storage unit that is not marked with a downgrade indicator.

[0182] Among them, the corresponding storage unit is a storage unit that stores the same data as the storage unit to be released, is not marked with a degradation flag, and meets the current maximum redundancy.

[0183] For example, the business module can release all storage units to be released and apply for a new storage unit corresponding to each storage unit to be released based on the currently available hard disks and the maximum redundancy.

[0184] S503, writes the data copy stored in the current storage unit to the corresponding storage unit.

[0185] For example, a copy of the data stored in the storage unit to be released can be written into each kObj of the corresponding storage unit to complete the replacement.

[0186] For example, the preset query time can be 0:00 every day. The business module can trigger the query at 0:00, traverse all storage units, and select three storage units with degradation flags to be released: kSeg-012, kSeg-015, and sysSeg-008. The business module can release these three units and create three corresponding storage units: kSeg-026, kSeg-027, and sysSeg-010, based on the currently available hard drives (4 data disks and 4 system disks). The data copies from the original units are then written to the new units to complete the replacement.

[0187] In the above embodiments, the timed query mechanism can proactively replace the storage units to be released with reliable units that provide the maximum redundancy that the storage system can offer, thereby improving data security.

[0188] In some alternative implementations, see [link to relevant documentation]. Figure 6 , Figure 6 A flowchart illustrating another method for managing storage areas is provided, which includes the following steps:

[0189] Step 1: Trigger the Mgr's traversal and filtering.

[0190] When the system experiences a frame redundancy recovery (the storage pool recovers from a frame redundancy failure) or an increase in available disks (the number of data disks meets the preset redundancy), the Mgr will traverse the kSeg set and filter out all kSegs with "frame redundancy failure flag" and "insufficient disk replica flag".

[0191] The kSeg set is a collection of all ordinary storage units. Each kSeg has different anomaly markers, such as a frame over-redundancy fault marker, which indicates that the storage pool is in a redundant anomaly state of "frame-level redundancy fault" (such as insufficient disk frames) when the kSeg is created; and a disk copy insufficiency marker, which indicates that the number of data disks is less than the preset redundancy when the kSeg is created.

[0192] Step 2: Notification of Degradation and Tag Writing.

[0193] After the Mgr filters out the kSegs to be processed, it sends a "notify downgrade" instruction to the Agent. After receiving the instruction, the Agent adds a downgrade identifier for these kSegs in the reserved memory for kSeg metadata.

[0194] Step 3: Subsequent management of kSeg.

[0195] When a business module uses kSeg, that is, after receiving a business request, it will refer to the state of the kSeg metadata reserved in memory (check for a degradation flag) and replace kSeg as needed.

[0196] The business module can also perform periodic checks. Based on the maintained kSeg set module, the business module periodically checks the status of kSeg and proactively handles kSegs with degradation flags to prevent degradation kSegs from occupying resources for a long time.

[0197] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0198] Based on the same inventive concept, this application also provides a storage region management device for implementing the storage region management method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more storage region management device embodiments provided below can be found in the limitations of the storage region management method described above, and will not be repeated here.

[0199] In one exemplary embodiment, such as Figure 7 As shown, a storage area management device is provided, comprising:

[0200] The filtering module 10 is used to filter the storage units to be degraded from each created storage unit when a preset degrade trigger event is detected; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy.

[0201] Add module 20 to add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0202] The above solution, on the one hand, filters storage units to be degraded by detecting preset degradation trigger events and adds degradation tags, which can trigger storage unit replacement operations based on the current maximum number of data replicas, ensuring that data replicas always maintain the maximum redundancy that the system can provide. Compared with traditional technologies, which directly remove the corresponding data replicas when the storage carrier fails or changes, this avoids the problem of reduced data security caused by replica removal. On the other hand, by using a tag-trigger method, there is no need to replace storage units in real time, reducing the immediate impact on system operation and balancing data security and system operating efficiency.

[0203] In one embodiment, the detection of a preset degradation trigger event includes any of the following:

[0204] Redundancy anomaly recovery events were detected; these include redundancy levels being restored from disk-level redundancy to frame-level redundancy, or increasing the number of disks when the number of disks is less than the preset redundancy; disk failure events were detected; and disk object reclamation events were detected.

[0205] In one embodiment, the filtering module 10 is specifically used for:

[0206] In the event of a redundancy anomaly recovery event, storage units corresponding to the redundancy anomaly event are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk failure event, storage units including hard disk objects from the failed hard disk are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk object reclamation event, a preset number of storage units including hard disk objects to be reclaimed are selected from all created storage units to obtain storage units to be downgraded; wherein, the preset number is the number of storage units that meet the reclamation requirements of the hard disk object reclamation event.

[0207] In one embodiment, the device further includes a creation module for:

[0208] In response to a data write request, a storage unit is created based on the preset redundancy and the currently available hard disks, and a data copy of the data corresponding to the data write request is written to each hard disk object of the created storage unit; if a redundancy exception event is detected, the created storage unit and the redundancy exception event are recorded accordingly.

[0209] In one embodiment, when the redundancy exception event is that the number of hard disks is less than a preset redundancy level, the creation module is specifically used for:

[0210] When the created storage unit is a first-type storage unit, an event identifier for a redundancy exception event is added to the metadata of the created storage unit; wherein, the metadata is stored in the preset memory of the controller to which the created storage unit belongs; the first-type storage unit is a storage unit that only includes a data disk; when the created storage unit is a second-type storage unit, the created storage unit is recorded in the storage unit list corresponding to the redundancy exception event; wherein, different storage unit lists are used to record created storage units whose actual number of data disks is less than different first numbers and created storage units whose actual number of system disks is less than different second numbers; wherein, the first number is not greater than the target number of data disks corresponding to the preset redundancy, the second number is not greater than the target number of system disks corresponding to the preset redundancy, and the second-type storage unit is a storage unit that at least includes a system disk.

[0211] In one embodiment, the device further includes a deletion module for:

[0212] In response to a disk deletion command, the system selects storage units from the existing storage units that contain the target disk object in the disk to be deleted as indicated by the disk deletion command, to obtain storage units to be processed; if the number of disk objects in the storage unit to be processed is greater than a preset number threshold, the target disk objects in the storage unit to be processed are released; wherein the preset number threshold is not less than 2; if the number of disk objects in the storage unit to be processed is not greater than the preset number threshold, a degradation flag is added to the storage unit to be processed.

[0213] In one embodiment, the device further includes an execution module for:

[0214] Upon receiving a business request, the system determines the current storage unit corresponding to the business request from among the existing storage units. If the current storage unit is marked with a degradation flag, the system releases the current storage unit and requests a target storage unit that is not marked with a degradation flag. The system writes a copy of the business data stored in the current storage unit to each hard disk object in the target storage unit and executes the business request based on the target storage unit.

[0215] In one embodiment, the device further includes a writing module for:

[0216] At each query time, select storage units marked with a degradation flag from the existing storage units to be released; release the storage units to be released and request corresponding storage units without a degradation flag; write the data copy stored in the current storage unit to the corresponding storage unit.

[0217] Each module in the aforementioned storage area management device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0218] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores copies of data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a memory area management method.

[0219] The staff can understand. Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0220] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0221] Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0222] Add a degradation flag to each storage unit to be degraded; the degradation flag is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0223] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0224] Redundancy anomaly recovery events were detected; these include redundancy levels being restored from disk-level redundancy to frame-level redundancy, or increasing the number of disks when the number of disks is less than the preset redundancy; disk failure events were detected; and disk object reclamation events were detected.

[0225] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0226] In the event of a redundancy anomaly recovery event, storage units corresponding to the redundancy anomaly event are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk failure event, storage units including hard disk objects from the failed hard disk are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk object reclamation event, a preset number of storage units including hard disk objects to be reclaimed are selected from all created storage units to obtain storage units to be downgraded; wherein, the preset number is the number of storage units that meet the reclamation requirements of the hard disk object reclamation event.

[0227] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0228] In response to a data write request, a storage unit is created based on the preset redundancy and the currently available hard disks, and a data copy of the data corresponding to the data write request is written to each hard disk object of the created storage unit; if a redundancy exception event is detected, the created storage unit and the redundancy exception event are recorded accordingly.

[0229] In one embodiment, when the redundancy exception event is that the number of hard disks is less than a preset redundancy level, the processor further performs the following steps when executing the computer program:

[0230] When the created storage unit is a first-type storage unit, an event identifier for a redundancy exception event is added to the metadata of the created storage unit; wherein, the metadata is stored in the preset memory of the controller to which the created storage unit belongs; the first-type storage unit is a storage unit that only includes a data disk; when the created storage unit is a second-type storage unit, the created storage unit is recorded in the storage unit list corresponding to the redundancy exception event; wherein, different storage unit lists are used to record created storage units whose actual number of data disks is less than different first numbers and created storage units whose actual number of system disks is less than different second numbers; wherein, the first number is not greater than the target number of data disks corresponding to the preset redundancy, the second number is not greater than the target number of system disks corresponding to the preset redundancy, and the second-type storage unit is a storage unit that at least includes a system disk.

[0231] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0232] In response to a disk deletion command, the system selects storage units from the existing storage units that contain the target disk object in the disk to be deleted as indicated by the disk deletion command, to obtain storage units to be processed; if the number of disk objects in the storage unit to be processed is greater than a preset number threshold, the target disk objects in the storage unit to be processed are released; wherein the preset number threshold is not less than 2; if the number of disk objects in the storage unit to be processed is not greater than the preset number threshold, a degradation flag is added to the storage unit to be processed.

[0233] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0234] Upon receiving a business request, the system determines the current storage unit corresponding to the business request from among the existing storage units. If the current storage unit is marked with a degradation flag, the system releases the current storage unit and requests a target storage unit that is not marked with a degradation flag. The system writes a copy of the business data stored in the current storage unit to each hard disk object in the target storage unit and executes the business request based on the target storage unit.

[0235] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0236] At each query time, select storage units marked with a degradation flag from the existing storage units to be released; release the storage units to be released and request corresponding storage units without a degradation flag; write the data copy stored in the current storage unit to the corresponding storage unit.

[0237] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0238] Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0239] Add a degradation flag to each storage unit to be degraded; the degradation flag is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0240] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0241] Redundancy anomaly recovery events were detected; these include redundancy levels being restored from disk-level redundancy to frame-level redundancy, or increasing the number of disks when the number of disks is less than the preset redundancy; disk failure events were detected; and disk object reclamation events were detected.

[0242] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0243] In the event of a redundancy anomaly recovery event, storage units corresponding to the redundancy anomaly event are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk failure event, storage units including hard disk objects from the failed hard disk are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk object reclamation event, a preset number of storage units including hard disk objects to be reclaimed are selected from all created storage units to obtain storage units to be downgraded; wherein, the preset number is the number of storage units that meet the reclamation requirements of the hard disk object reclamation event.

[0244] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0245] In response to a data write request, a storage unit is created based on the preset redundancy and the currently available hard disks, and a data copy of the data corresponding to the data write request is written to each hard disk object of the created storage unit; if a redundancy exception event is detected, the created storage unit and the redundancy exception event are recorded accordingly.

[0246] In one embodiment, when the redundancy exception event is that the number of hard disks is less than a preset redundancy level, the computer program, when executed by the processor, further performs the following steps:

[0247] When the created storage unit is a first-type storage unit, an event identifier for a redundancy exception event is added to the metadata of the created storage unit; wherein, the metadata is stored in the preset memory of the controller to which the created storage unit belongs; the first-type storage unit is a storage unit that only includes a data disk; when the created storage unit is a second-type storage unit, the created storage unit is recorded in the storage unit list corresponding to the redundancy exception event; wherein, different storage unit lists are used to record created storage units whose actual number of data disks is less than different first numbers and created storage units whose actual number of system disks is less than different second numbers; wherein, the first number is not greater than the target number of data disks corresponding to the preset redundancy, the second number is not greater than the target number of system disks corresponding to the preset redundancy, and the second-type storage unit is a storage unit that at least includes a system disk.

[0248] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0249] In response to a disk deletion command, the system selects storage units from the existing storage units that contain the target disk object in the disk to be deleted as indicated by the disk deletion command, to obtain storage units to be processed; if the number of disk objects in the storage unit to be processed is greater than a preset number threshold, the target disk objects in the storage unit to be processed are released; wherein the preset number threshold is not less than 2; if the number of disk objects in the storage unit to be processed is not greater than the preset number threshold, a degradation flag is added to the storage unit to be processed.

[0250] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0251] Upon receiving a business request, the system determines the current storage unit corresponding to the business request from among the existing storage units. If the current storage unit is marked with a degradation flag, the system releases the current storage unit and requests a target storage unit that is not marked with a degradation flag. The system writes a copy of the business data stored in the current storage unit to each hard disk object in the target storage unit and executes the business request based on the target storage unit.

[0252] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0253] At each query time, select storage units marked with a degradation flag from the existing storage units to be released; release the storage units to be released and request corresponding storage units without a degradation flag; write the data copy stored in the current storage unit to the corresponding storage unit.

[0254] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0255] Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy;

[0256] Add a degradation flag to each storage unit to be degraded; the degradation flag is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

[0257] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0258] Redundancy anomaly recovery events were detected; these include redundancy levels being restored from disk-level redundancy to frame-level redundancy, or increasing the number of disks when the number of disks is less than the preset redundancy; disk failure events were detected; and disk object reclamation events were detected.

[0259] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0260] In the event of a redundancy anomaly recovery event, storage units corresponding to the redundancy anomaly event are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk failure event, storage units including hard disk objects from the failed hard disk are selected from all created storage units to obtain storage units to be downgraded; in the event of a hard disk object reclamation event, a preset number of storage units including hard disk objects to be reclaimed are selected from all created storage units to obtain storage units to be downgraded; wherein, the preset number is the number of storage units that meet the reclamation requirements of the hard disk object reclamation event.

[0261] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0262] In response to a data write request, a storage unit is created based on the preset redundancy and the currently available hard disks, and a data copy of the data corresponding to the data write request is written to each hard disk object of the created storage unit; if a redundancy exception event is detected, the created storage unit and the redundancy exception event are recorded accordingly.

[0263] In one embodiment, when the redundancy exception event is that the number of hard disks is less than a preset redundancy level, the computer program, when executed by the processor, further performs the following steps:

[0264] When the created storage unit is a first-type storage unit, an event identifier for a redundancy exception event is added to the metadata of the created storage unit; wherein, the metadata is stored in the preset memory of the controller to which the created storage unit belongs; the first-type storage unit is a storage unit that only includes a data disk; when the created storage unit is a second-type storage unit, the created storage unit is recorded in the storage unit list corresponding to the redundancy exception event; wherein, different storage unit lists are used to record created storage units whose actual number of data disks is less than different first numbers and created storage units whose actual number of system disks is less than different second numbers; wherein, the first number is not greater than the target number of data disks corresponding to the preset redundancy, the second number is not greater than the target number of system disks corresponding to the preset redundancy, and the second-type storage unit is a storage unit that at least includes a system disk.

[0265] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0266] In response to a disk deletion command, the system selects storage units from the existing storage units that contain the target disk object in the disk to be deleted as indicated by the disk deletion command, to obtain storage units to be processed; if the number of disk objects in the storage unit to be processed is greater than a preset number threshold, the target disk objects in the storage unit to be processed are released; wherein the preset number threshold is not less than 2; if the number of disk objects in the storage unit to be processed is not greater than the preset number threshold, a degradation flag is added to the storage unit to be processed.

[0267] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0268] Upon receiving a business request, the system determines the current storage unit corresponding to the business request from among the existing storage units. If the current storage unit is marked with a degradation flag, the system releases the current storage unit and requests a target storage unit that is not marked with a degradation flag. The system writes a copy of the business data stored in the current storage unit to each hard disk object in the target storage unit and executes the business request based on the target storage unit.

[0269] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0270] At each query time, select storage units marked with a degradation flag from the existing storage units to be released; release the storage units to be released and request corresponding storage units without a degradation flag; write the data copy stored in the current storage unit to the corresponding storage unit.

[0271] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0272] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0273] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0274] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A storage area management method, characterized in that, The method includes: Upon detecting a preset degradation trigger event, the storage units to be degraded corresponding to the preset degradation trigger event are selected from each created storage unit; wherein, each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy; Add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

2. The method according to claim 1, characterized in that, The condition of detecting a preset degradation trigger event includes any of the following: A redundancy anomaly recovery event was detected; wherein, the redundancy anomaly recovery event includes the redundancy level being restored from disk-level redundancy to frame-level redundancy, or the number of hard drives being increased when the number of hard drives is less than a preset redundancy level; A hard drive failure event was detected; A disk object reclamation event was detected.

3. The method according to claim 2, characterized in that, The step of filtering the storage units to be downgraded corresponding to the preset downgrade trigger event from each created storage unit when a preset downgrade trigger event is detected includes any one of the following: In the event of a redundancy anomaly recovery event, the storage units corresponding to the redundancy anomaly events are selected from the existing storage units to obtain the storage units to be downgraded. In the event of a hard disk failure, storage units that include hard disk objects from the failed hard disk are selected from all created storage units to obtain the storage units to be degraded. Upon detecting a disk object reclamation event, a preset number of storage units, including the disk objects to be reclaimed, are selected from all created storage units to obtain storage units to be downgraded; wherein, the preset number is the number of storage units that meet the reclamation requirements of the disk object reclamation event.

4. The method according to claim 2, characterized in that, The method further includes: In response to a data write request, a storage unit is created based on a preset redundancy and currently available hard disks, and a data copy of the data corresponding to the data write request is written to each hard disk object of the created storage unit; If a redundant abnormal event is detected, the corresponding storage unit created and the redundant abnormal event are recorded.

5. The method according to claim 4, characterized in that, When the redundancy anomaly event is that the number of hard disks is less than the preset redundancy, the storage unit created for the corresponding record and the redundancy anomaly event include any one of the following: When the created storage unit is a first-type storage unit, an event identifier for the redundant exception event is added to the metadata of the created storage unit; wherein, the metadata is stored in the preset memory of the controller to which the created storage unit belongs; the first-type storage unit is a storage unit that only includes a data disk; When the created storage unit is a second type of storage unit, the created storage unit is recorded in the storage unit list corresponding to the redundancy exception event; wherein, the different storage unit lists are used to record created storage units with an actual data disk number less than a different first number and created storage units with an actual system disk number less than a different second number; wherein, the first number is not greater than the target data disk number corresponding to the preset redundancy, the second number is not greater than the target system disk number corresponding to the preset redundancy, and the second type of storage unit is a storage unit that includes at least a system disk.

6. The method according to claim 1, characterized in that, The method further includes: In response to a hard disk deletion command, storage units containing the target hard disk object in the hard disk to be deleted indicated by the hard disk deletion command are selected from each created storage unit to obtain the storage unit to be processed; If the number of hard disk objects in the storage unit to be processed is greater than a preset number threshold, the target hard disk objects in the storage unit to be processed are released; wherein, the preset number threshold is not less than 2; If the number of hard disk objects in the storage unit to be processed is not greater than a preset threshold, the degradation flag is added to the storage unit to be processed.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: Upon receiving a service request, the current storage unit corresponding to the service request is determined from among the existing storage units. If the current storage unit is marked with the degradation flag, release the current storage unit and request a target storage unit that is not marked with the degradation flag; Write a copy of the business data stored in the current storage unit to each hard disk object of the target storage unit, and execute the business request based on the target storage unit.

8. The method according to any one of claims 1-6, characterized in that, The method further includes: At each query time, select the storage units marked with the degradation flag from the created storage units to be released; Release the storage unit to be released, and request a corresponding storage unit that is not marked with the downgrade identifier; Write a copy of the data stored in the current storage unit to the corresponding storage unit.

9. A storage area management device, characterized in that, The device includes: The filtering module is used to filter the storage units to be degraded from each created storage unit when a preset degradation trigger event is detected; wherein each storage unit includes each hard disk object belonging to different hard disks, the different hard disks include data disks and / or system disks, the data disks belong to the same storage pool, the system disks include at least one system disk of each available controller, and each hard disk object stores the same data copy; An add module is used to add a degradation identifier to each storage unit to be degraded; wherein, the degradation identifier is used to trigger a response to a business request, release the storage unit to be degraded, and apply for a new storage unit based on the maximum number of data replicas that can be achieved at present.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images