Method and device for recovering data UNC in solid state disk and server

Abstract

The invention provides a method and device for recovering data UNC in a solid state disk and a server, and relates to the technical field of inter-disk data recovery, the method comprises the following steps: obtaining preset recovery time of the UNC data in the disk, and receiving a read command issued by a host end, the UNC data being original data of single reading using default voltage and being uncorrectable; when it is monitored that the UNC data appears in the read command, in-disk repair processing of the SSD disk is carried out on the UNC data within preset recovery time; and if the in-disk repair fails, sending the address of the UNC to a host end so as to carry out system layer RAID repair on the UNC data to obtain correct target data corresponding to the UNC data. According to the method, the recovery efficiency of the UNC data can be remarkably improved.

Description

Technical Field

[0001] This invention relates to the technical field of data recovery between disks, and in particular to a method, apparatus, and server for recovering UNC data within a solid-state drive. Background Technology

[0002] Enterprise-level solid-state storage devices employ multi-level UNC data error correction and recovery strategies. During the response process of a Read command initiated by the Host to the SSD controller, if UNC errors are encountered, the SSD will attempt to recover the data by trying different methods at each level. Currently, related technologies suggest that data recovery can be attempted within the disk first, until the data error correction is successful or all methods have been tried before returning to the Host. This allows for data recovery at the system level via RAID. However, since the time required for in-disk repair is much longer than that for system-level repair, latency may occur, affecting the efficiency of data recovery. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method, apparatus and server for recovering UNC data in a solid-state drive, which can significantly improve the recovery efficiency of UNC data.

[0004] In a first aspect, embodiments of the present invention provide a method for recovering UNC data within a solid-state drive (SSD). The method includes: obtaining a preset recovery time for the UNC data within the drive and receiving a read command from the host, wherein the UNC data is raw, uncorrectable data from a single read using the default voltage; when UNC data is detected in the read command, performing SSD in-disk repair processing on the UNC data within the preset recovery time; if the in-disk repair fails, sending the UNC address to the host to perform system-level RAID repair on the UNC data to obtain the correct target data corresponding to the UNC data.

[0005] In one implementation, the step of performing on-disk repair processing on UNC data within a preset recovery time includes: rereading the UNC data, performing stripe fine-tuning scan processing, and performing RAID repair processing on the internal firmware of the SSD.

[0006] In one implementation, after the steps of rereading UNC data, stripe fine-tuning scan processing, and RAID repair processing of the SSD internal firmware, the method includes: if the repair time exceeds the preset recovery time, or the number of attempts exceeds the default number of recovery attempts, then the disk repair is determined to have failed.

[0007] In one implementation, after performing rereading of UNC data, stripe fine-tuning scanning, and RAID repair processing of the SSD internal firmware, the process further includes: if the disk repair is successful within a preset recovery time and the correct target data corresponding to the UNC data is obtained, then the target data is fed back to the host to end the data recovery process.

[0008] In one implementation, the step of performing system-level RAID repair on UNC data to obtain the correct target data corresponding to the UNC data includes: receiving the LBA logical address of the UNC data through the part of the host responsible for RAID logic, and using the SSD disk where UNC does not appear to perform verification calculation on the LBA logical address where UNC appears, and calculating the target data.

[0009] In one implementation, after obtaining the correct target data corresponding to the UNC data, the process includes: rewriting the target data to an SSD disk to eliminate SSD disk errors and complete the data recovery of the UNC data.

[0010] In one implementation, after the step of performing system-level RAID repair on UNC data, the method includes: if the target data is not obtained, extending the preset recovery time by a preset duration or increasing the default recovery attempt count by a preset number of times, and then re-performing the disk repair.

[0011] Secondly, embodiments of the present invention also provide a device for recovering UNC data within a solid-state drive (SSD). The device includes: a data acquisition module, which acquires a preset recovery time for the UNC data within the drive and receives a read command from the host, wherein the UNC data is raw, uncorrectable data from a single read using the default voltage; an in-drive repair module, which performs in-drive repair processing on the SSD for the UNC data within the preset recovery time when UNC data is detected in the read command; and a system repair module, which sends the address of the UNC data to the host if the in-drive repair fails, to perform system-level RAID repair on the UNC data and obtain the correct target data corresponding to the UNC data.

[0012] Thirdly, embodiments of the present invention also provide a server, including a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement any of the methods provided in the first aspect.

[0013] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement any of the methods provided in the first aspect.

[0014] The embodiments of the present invention bring the following beneficial effects: This invention provides a method, apparatus, and server for recovering UNC data within a solid-state drive (SSD). The method acquires a preset recovery time for the UNC data within the drive and receives a read command from the host. It then detects the read command. When UNC data is detected in the read command, it performs SSD in-disk repair processing on the UNC data within the preset recovery time. If the in-disk repair fails, the UNC address is sent to the host for system-level RAID repair of the UNC data, obtaining the correct target data corresponding to the UNC data. This invention significantly improves the efficiency of UNC data recovery.

[0015] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating a method for recovering UNC data within a solid-state drive, provided in an embodiment of the present invention; Figure 2 A schematic diagram illustrating the specific process of a method for recovering UNC data within a solid-state drive, provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a system for recovering UNC data within a solid-state drive, provided in an embodiment of the present invention. Figure 4 This is a schematic diagram of a device for recovering UNC data within a solid-state drive, provided in an embodiment of the present invention. Figure 5 This is a schematic diagram of the structure of a server provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Currently, enterprise-level solid-state storage devices (all-flash arrays, enterprise-level SSDs) have multi-level UNC data error correction and recovery strategies, such as Nand Retry, FSP, and RAID within the SSD, and RAID and off-site backup at the storage device system level. During the response process of the Read command initiated by the host to the SSD controller, if UNC is encountered, the SSD attempts to recover the data by trying each level, which takes a long time in the case of the longest path.

[0021] Since UNC data is generated within the SSD, SSD manufacturers typically try various methods to recover the data until the data is successfully corrected or all methods have been tried before returning the data to the host. Trying all measures often requires multiple or even dozens of reads on the NAND flash memory, which can take more than 1ms. In system-level storage solutions, multiple SSDs are often protected by RAID or off-site backup for disaster recovery. Data recovery at the system level via RAID is estimated to take about 100us, and the success rate is relatively high. Since the longest path within an SSD may have a latency in the millisecond range, prioritizing data recovery within the SSD is not the optimal approach from a time perspective. Therefore, the UNC data recovery method provided in this invention can significantly improve the efficiency of UNC data repair.

[0022] See Figure 1 The diagram shows a method for significantly improving the repair efficiency of UNC data. The method mainly includes the following steps S102 to S106: Step S102: Obtain the preset recovery time for UNC data in the disk and receive the read command issued by the host. The UNC data is the raw, uncorrectable data from a single read using the default voltage. In one implementation, the host can configure the UNC data recovery attempt time or / and number of attempts for all SSDs to 500us (the default value for SSD power-on is 100ms or even longer) and the number of attempts to 3. The UNC data recovery attempt time or / and number of attempts can be dynamically configured according to the actual situation or determined based on the average response time of the SSD read command in the system. In addition, when limiting the repair time of the disk repair, either a time limit or a number of attempts limit can be selected.

[0023] Step S104: When UNC data is detected in the read command, the SSD performs in-disk repair processing for the UNC data within a preset recovery time. In one implementation, when the SSD encounters UNC data in a subsequent Read command issued by the Host, it adds a check to determine whether it is within the configured time. If the threshold is not exceeded, operations such as Retry, FSP, and RAID are initiated. If the threshold is exceeded, UNC is returned. Retry is used to read the same NAND again and try to correct the error by changing the voltage and the reference level.

[0024] Step S106: If the disk fails to repair within a limited time or number of attempts, the address of the UNC is sent to the host to perform system-level RAID repair on the UNC data, thereby obtaining the correct target data corresponding to the UNC data. In one implementation, when the host receives the Read Response returned by the SSD as UNC, it starts RAID recovery through the RAID Controller, thereby obtaining the data needed by the host in a shorter time. The host then rewrites the data recovered by RAID back, improving the reliability of the data on the disk. In other words, after receiving the UNC data returned by the SSD, the host / RAID Controller can immediately use system-level RAID (inter-disk / strip redundancy) to calculate the missing data and directly return the calculated correct data (i.e., the target data) to the service (read command completed).

[0025] The UNC data recovery method provided in this embodiment of the invention can significantly improve the efficiency of UNC data recovery.

[0026] See Figure 2 The diagram illustrates a specific process for recovering UNC data within a solid-state drive (SSD). This embodiment also provides an implementation method for recovering UNC data within a SSD with dynamically configurable maximum latency. This method is applicable to applications such as... Figure 3 The system shown is a recovery system for UNC data within a solid-state drive. See (A) to (B) below for details: (A) In-disk repair: When the SSD encounters UNC data in a subsequent Read command issued by the Host, it can initiate operations such as Retry, FSP, and RAID within the configured time. If the data recovery attempt time configured by the Host is exceeded and the data is still not successfully recovered, the Read Response is immediately returned as UNC. That is, the UNC data is reread (i.e., Retry), stripe fine-tuning scan (i.e., FSP), and RAID repair processing of the SSD's internal firmware (i.e., RAID). Among them, FSP is a relatively complex process, requiring about 10 reread operations after voltage adjustment. If the repair time exceeds the preset recovery time, the in-disk repair is determined to have failed. If the in-disk repair is successful within the preset recovery time and the correct target data corresponding to the UNC data is obtained, the target data is fed back to the host to end the data recovery process. In one implementation, if the host does not configure the recovery time for in-disk UNC, the default recovery time of the SSD is used, and the repair time of in-disk repair is limited by the default number of recovery attempts.

[0027] In addition, during in-disk repair, UNC data can be reread, or data recovery can be attempted by using progressive data recovery methods such as HardDecode, SoftDecode, other methods provided by Nand, and RAID repair within the SSD. The preset recovery time for in-disk repair can be preset or temporarily configured by the host.

[0028] (B) System-level repair: After the in-disk repair fails, the SSD can continue to attempt long-path data recovery. If the subsequent recovery is successful, the off-site write method will be started to prevent the data from becoming truly irreparable. See (1) to (3) below for details: (1) After receiving the Read Response as UNC from the SSD, the Host / Raid Controller starts the RAID recovery operation at the system layer. It receives the LBA logical address of the UNC data through the part of the host responsible for RAID logic, and uses the SSD disk that does not have UNC to perform verification calculation on the LBA logical address that has UNC to calculate the target data. The Host / Raid Controller is the hardware responsible for RAID logic on the host side. The Host is the host side. The Raid Controller can be an independent PCIe RAID card (e.g., Broadcom MegaRAID, Microsemi Adaptec) or RAID software running in the Host (e.g., Linux mdadm, Windows Storage Spaces, SPDK RAID bdev). It is used to form a redundant array (RAID 5 / 6 / 10 …) of several SSDs / HDDs and present a virtual logical disk to the upper layer. All read and write operations must go through it to perform striping, XOR / RS verification calculation and bad disk reconstruction.

[0029] (2) If the RAID recovery at the system level is successful, in addition to returning to the host, the data can also be written to the SSD disk where UNC appeared before. That is, the target data is rewritten to the SSD disk to eliminate the SSD disk error and complete the data recovery of UNC data. In other words, the host writes the target data back to the new physical page of the original SSD or other backup disk. In this way, the next time the LBA logical address where UNC appears is read, the data UNC will not be prompted. If the write-back is not performed, the data UNC will still appear when reading the same address again. The data can only be recalculated by RAID again.

[0030] (3) If the system-level RAID recovery fails, the data will be read from the original SSD again. That is, if the target data is not obtained, the preset recovery time will be extended by a preset duration or the default number of recovery attempts will be increased by a preset number before the disk repair is performed again.

[0031] In summary, this invention typically requires only about 200µs to recover data between disks, while existing technologies may require more than 1ms or even longer to recover data within an SSD. Therefore, this invention can significantly improve the latency of the Read command when the original data in the flash memory is in UNC format on a solid-state drive, resulting in substantial benefits for improving the service quality of SSDs.

[0032] Meanwhile, if the RAID data between disks cannot be recovered successfully, the maximum recovery latency of the UNC can be dynamically adjusted to a larger value until the data is recovered or all data recovery measures are completed before exiting. In addition to prioritizing the startup of RAID at the system level, the RAID, Retry, and FSP within the SSD can also be implemented before the deep Retry and FSP to achieve better latency.

[0033] Regarding the method for recovering UNC data within a solid-state drive (SSD) provided in the foregoing embodiments, this invention provides a device for recovering UNC data within a SSD. (See attached image.) Figure 4 The diagram shows a structural schematic of a device for recovering UNC data within a solid-state drive. The device includes the following components: The data acquisition module 402 obtains the preset recovery time of the UNC data in the disk and receives the read command sent by the host. The UNC data is the raw data of a single read using the default voltage and is not editable. The in-disk repair module 404 performs in-disk repair processing on the SSD disk when UNC data is detected in the read command, within a preset recovery time. If the in-disk repair fails, the system repair module 406 sends the UNC address to the host to perform system-level RAID repair on the UNC data and obtain the correct target data corresponding to the UNC data.

[0034] The UNC data recovery device provided in this application embodiment can significantly improve the efficiency of UNC data recovery.

[0035] In one embodiment, when performing the step of in-disk repair processing of UNC data on the SSD within a preset recovery time, the in-disk repair module 404 is further used for: rereading UNC data, stripe fine-tuning scan processing, and RAID repair processing of the SSD internal firmware.

[0036] In one embodiment, after performing the steps of rereading UNC data, stripe fine-tuning scanning, and RAID repair of the SSD internal firmware, the disk repair module 404 is further configured to: determine that the disk repair has failed if the repair time exceeds the preset recovery time or the number of attempts exceeds the default number of recovery attempts.

[0037] In one embodiment, after performing the steps of rereading UNC data, stripe fine-tuning scanning, and RAID repair of the SSD internal firmware, the disk repair module 404 is further configured to: if the disk repair is successful within the preset recovery time and the correct target data corresponding to the UNC data is obtained, then feed the target data back to the host to end the data recovery process.

[0038] In one embodiment, when performing system-level RAID repair on UNC data to obtain the correct target data corresponding to the UNC data, the system repair module 406 is further configured to: receive the LBA logical address of the UNC data through the part of the host responsible for RAID logic, and use the SSD disk where UNC does not appear to perform verification calculation on the LBA logical address where UNC appears, and calculate the target data.

[0039] In one implementation, after obtaining the correct target data corresponding to the UNC data, the system repair module 406 is further configured to: rewrite the target data to the SSD disk to eliminate the SSD disk error and complete the data recovery of the UNC data.

[0040] In one embodiment, after performing system-level RAID repair on the UNC data, the system repair module 406 is further configured to: if the target data is not obtained, extend the preset recovery time by a preset duration or increase the default recovery attempt count by a preset number of times, and then perform disk repair again.

[0041] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0042] This invention provides a server, specifically, the server includes a processor and a storage device; the storage device stores a computer program, which, when run by the processor, executes the method described in any of the above embodiments.

[0043] Figure 5 This is a schematic diagram of the structure of a server provided in an embodiment of the present invention. The server 100 includes: a processor 50, a memory 51, a bus 52 and a communication interface 53. The processor 50, the communication interface 53 and the memory 51 are connected through the bus 52. The processor 50 is used to execute executable modules, such as computer programs, stored in the memory 51.

[0044] The memory 51 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 53 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0045] Bus 52 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0046] The memory 51 is used to store programs. After receiving an execution instruction, the processor 50 executes the programs. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 50 or implemented by the processor 50.

[0047] Processor 50 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 50 or by instructions in software form. Processor 50 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 51. The processor 50 reads the information in memory 51 and, in conjunction with its hardware, completes the steps of the above method.

[0048] The computer program product of the readable storage medium provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the foregoing method embodiments. For specific implementation, please refer to the foregoing method embodiments, which will not be repeated here.

[0049] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0050] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for recovering UNC data within a solid-state drive, characterized in that, The method includes: The preset recovery time for UNC data in the disk is obtained, and a read command is received from the host. The UNC data is the raw, uncorrectable data from a single read using the default voltage. When UNC data is detected in the read command, the SSD disk will be repaired within the preset recovery time. If the in-disk repair fails, the UNC address is sent to the host to perform system-level RAID repair on the UNC data, thereby obtaining the correct target data corresponding to the UNC data.

2. The method for recovering UNC data within a solid-state drive according to claim 1, characterized in that, The step of performing on-disk repair processing on the UNC data within the preset recovery time includes: The UNC data undergoes rereading, stripe fine-tuning scanning, and RAID repair processing of the SSD's internal firmware.

3. The method for recovering UNC data within a solid-state drive according to claim 2, characterized in that, After the steps of rereading the UNC data, stripe fine-tuning scan processing, and RAID repair processing of the SSD internal firmware, the following steps are included: If the repair time exceeds the preset recovery time, or the number of attempts exceeds the default number of recovery attempts, then the in-disk repair is determined to have failed.

4. The method for recovering UNC data within a solid-state drive according to claim 2, characterized in that, After the steps of rereading the UNC data, stripe fine-tuning scan processing, and RAID repair processing of the SSD internal firmware, the method further includes: If the disk repair is successful within the preset recovery time and the correct target data corresponding to the UNC data is obtained, the target data will be fed back to the host to end the data recovery process.

5. The method for recovering UNC data within a solid-state drive according to claim 1, characterized in that, The step of performing system-level RAID repair on the UNC data to obtain the correct target data corresponding to the UNC data includes: The host-side RAID logic receives the LBA logical address of the UNC data and uses an SSD disk without UNC to perform verification calculations on the LBA logical address of the UNC, thereby calculating the target data.

6. The method for recovering UNC data within a solid-state drive according to claim 1, characterized in that, After the step of obtaining the correct target data corresponding to the UNC data, the following is included: The target data is rewritten to the SSD disk to eliminate the SSD disk error and complete the data recovery of the UNC data.

7. The method for recovering UNC data within a solid-state drive according to claim 1, characterized in that, After performing system-level RAID repair on the UNC data, the following steps are included: If the target data is not obtained, the preset recovery time will be extended by a preset duration or the default number of recovery attempts will be increased by a preset number before resuming disk repair.

8. A device for recovering UNC data within a solid-state drive, characterized in that, The device includes: The data acquisition module obtains the preset recovery time of the UNC data in the disk and receives the read command issued by the host. The UNC data is the raw data of a single read using the default voltage and is uncorrectable. The in-disk repair module, when detecting UNC data in the read command, performs in-disk repair processing on the UNC data within the preset recovery time. If the system repair module fails to repair the disk, it sends the address of the UNC to the host to perform system-level RAID repair on the UNC data and obtain the correct target data corresponding to the UNC data.

9. A server, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

Images