Data error correction and performance optimization method based on multiple optical disks

Abstract

The technical scheme provides a data error correction and performance optimization method based on multiple optical disks, comprising the following steps: S1, generating a logical volume which is in a temporary data state before data recording and comprises a data block and a verification block; s2, converting the logical volume into a physical volume through a recombination algorithm, wherein the physical volume is a mirror image file which can be directly recorded to an optical disc; s3, respectively burning the physical volume to a plurality of corresponding optical disks, and synchronously mounting the optical disks through a plurality of CD drivers to realize concurrent read-write operation on the physical volume; and S4, when the data block in any optical disc is damaged, reversely calculating the position of the damaged data block in the logical volume through the recombination algorithm, calling undamaged data blocks and verification blocks corresponding to the damaged data block in other optical discs, and recovering the damaged data block based on an erasure code algorithm. According to the technical scheme, the reading performance of the large file can be optimized, and data protection can be provided.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of optical disc storage technology and data redundancy, and particularly relates to a data error correction and performance optimization method based on multiple optical discs. BACKGROUND

[0002] With the increasing demand for data archiving, Blu-ray discs have become an important medium for static data archiving due to their large capacity (25GB per single-layer disc, 50GB per double-layer disc), low cost, long-term storage stability, and other advantages. However, Blu-ray discs have significant performance defects: their read-write bandwidth is usually only 16MB / s-54MB / s (commonly 26MB / s), which is much lower than that of mechanical hard drives (100-200MB / s), and single-drive reading and writing can severely affect work efficiency; at the same time, Blu-ray discs are physically fragile and susceptible to scratches, moisture, and high temperatures, which can lead to data loss, requiring data protection mechanisms.

[0003] In the prior art, to improve storage performance and security, RAID5 or RAID6 schemes (such as 3+1 RAID5: 3 data discs + 1 parity disc) are often used to achieve redundancy through parallel reading and writing of multiple discs and checking data. However, such schemes have obvious limitations: first, RAID schemes are designed based on the random access characteristics of hard drives, and when adapted to the sequential access characteristics of Blu-ray discs, the performance optimization effect is limited; second, in traditional RAID5 / 6, the parity data is stored in a single disc, and only data discs can be concurrently read (e.g., 3+1 RAID5 only 3 drives concurrently), which cannot achieve full concurrency of all drives, resulting in low bandwidth utilization; third, Blu-ray discs lack an efficient data recovery mechanism adapted to their characteristics, and data reconstruction efficiency is low after physical damage.

[0004] Therefore, there is an urgent need for a technical solution that is tailored to the characteristics of Blu-ray discs to improve read-write performance and data security. SUMMARY

[0005] To solve the above problems, the present application provides a data error correction and performance optimization method based on multiple optical discs, and the specific technical solution is as follows: A data error correction and performance optimization method based on multiple optical discs, comprising the following steps: S1: generating a logical volume, the logical volume being a temporary data state before data is burned, including data blocks and parity blocks; wherein the data blocks are the smallest data units formed by allocating the data to be stored to each data disc according to a predetermined disc capacity and then splitting them according to a predetermined strip size; and the parity blocks are generated based on an erasure code algorithm for all data blocks and allocated to a predetermined parity disc; S2: converting the logical volume into a physical volume by a reorganization algorithm, the physical volume being an image file that can be directly burned to an optical disc; the reorganization algorithm establishes a mapping relationship between each minimum data unit in the logical volume and a storage location in the physical volume through a mathematical formula, so that data blocks and check blocks in the physical volume are dispersedly stored in a plurality of optical discs, and each optical disc contains data blocks and check blocks; S3: burning the physical volume to a corresponding plurality of optical discs respectively, synchronously mounting the plurality of optical discs by a plurality of optical drives, and realizing concurrent read-write operations on the physical volume; S4: when a data block in any of the optical discs is damaged, reversely calculating the position of the damaged data block in the logical volume by the reorganization algorithm, calling undamaged data blocks and check blocks corresponding to the damaged data block in other optical discs, and recovering the damaged data block based on an erasure code algorithm.

[0006] Further, in step S1, the preset strip size is 2KB; and the generation process of the logical volume includes: S11: obtaining to-be-stored data and N optical discs, setting N-1 optical discs as data optical discs and 1 optical disc as a check optical disc, and setting the capacity of each optical disc as 10KB; S12: sequentially allocating the to-be-stored data to each data optical disc according to the capacity of the data optical disc, to form a continuous data segment; S13: calculating a check data segment based on an erasure code algorithm, and allocating the check data segment to the check optical disc; S14: splitting the continuous data segment of each data optical disc and the check data segment of the check optical disc according to a 2KB strip size, to obtain a plurality of minimum data units, marked as “data block n.m” or “check block n.m”, wherein n is the optical disc number and m is the serial number of the minimum data unit in the optical disc, to form a logical volume.

[0007] Further, in step S2, the mathematical formula of the reorganization algorithm includes a conversion formula from the logical volume to the physical volume and a conversion formula from the physical volume to the logical volume; The conversion formula from the logical volume to the physical volume is: PhysicOff = (LogicOff / BandSize + LogicIndexBandNum) / DiscNum *BandSize PhysicIndex = (LogicOff / BandSize + LogicIndexBandNum) % DiscNum The conversion formula from the physical volume to the logical volume is: LogicOff = (((PhysicOff / BandSize * DiscNum) + PhysicIndex) % BandNum)* BandSize LogicIndex = ((PhysicOff / BandSize * DiscNum) + PhysicIndex) / BandNum Wherein, LogicIndex is the index of the optical disc in the logical volume (starting from 0), LogicOff is the offset of the smallest data unit in the logical volume, DiscNum is the total number of optical discs, BandNum is the number of stripes on a single optical disc, BandSize is the preset stripe size, PhysicalIndex is the index of the optical disc in the physical volume (starting from 0), and PhysicalOff is the offset of the smallest data unit in the physical volume.

[0008] Furthermore, in step S3, the concurrent read / write operation is implemented as follows: when reading target data, all optical drives synchronously read the smallest data unit of the physical volume in their respective mounted optical discs, and convert the read smallest data unit into a continuous data segment of the logical volume through a reorganization algorithm to complete the target data reading; the reading time of the target data is equivalent to the time of a single optical drive reading the maximum value among the number of the smallest data units corresponding to the target data.

[0009] Furthermore, in step S4, the specific steps for data recovery are as follows: S41: Determine the location parameters (PhysicIndex, PhysicOff) of the corrupted data block in the physical volume. S42: Substitute the physical volume to logical volume conversion formula to calculate the location parameters (LogicIndex, LogicOff) of the corrupted data block in the logical volume. S43: Extract other undamaged data blocks and corresponding check blocks in the logical volume that are in the same group as the damaged data block; S44: Based on the erasure coding algorithm, perform operations on the extracted undamaged data blocks and check blocks to generate the original data of the damaged data blocks and complete the recovery.

[0010] Furthermore, the optical disc is a Blu-ray disc, and the read / write bandwidth of the Blu-ray disc is 16MB / s-54MB / s; the number of the plurality of optical drives is 4-12.

[0011] Furthermore, it also includes RAID metadata management steps: the RAID metadata is based on the MDADM superblock design of the Linux system, and the storage location of the RAID metadata in the logical volume and physical volume is consistent, and it does not participate in the conversion of the reassembly algorithm.

[0012] Compared with the prior art, the beneficial effects of this invention patent are as follows: Significant performance improvement: The read / write bandwidth of a single optical drive Blu-ray disc is approximately 26MB / s, and the theoretical peak bandwidth reaches 312MB / s when 12 optical drives are running concurrently, comparable to mechanical hard drives, solving the problem of "low speed of single drive"; Higher concurrency efficiency: Compared to the traditional 3+1 RAID5 which only allows 3 optical drives to operate concurrently, this solution enables all optical drives to operate concurrently, reducing the time to read the same data by 25%-50%; Strong data security: Data recovery from single disk failure is achieved based on erasure coding, and data blocks and parity blocks are stored separately in the physical volume, reducing the risk of "multiple disks failing at the same time"; Good scenario adaptability: Designed specifically for Blu-ray disc archiving scenarios, it is compatible with Blu-ray's advantages of large capacity and low cost, and is suitable for storing static data (such as historical videos and offline backup data) in enterprises / archives; Flexible deployment: Supports flexible expansion of 4-12 optical drives, and the number of optical discs and optical drives can be adjusted according to archiving needs, with controllable hardware costs. Attached Figure Description

[0013] Figure 1 illustrates the data optimization and performance optimization steps of this invention.

[0014] Figure 2 illustrates the steps for generating a logical volume in this invention.

[0015] Figure 3 illustrates the data recovery steps in this invention.

[0016] Figure 4 is one of the schematic diagrams of the logical volume generation process in an embodiment of the present invention.

[0017] Figure 5 is a second schematic diagram of the logical volume generation process in an embodiment of the present invention.

[0018] Figure 6 is a schematic diagram of the logical volume generation process in an embodiment of the present invention.

[0019] In the diagram: Optical discs 1-3 are data discs, storing 0-10KB, 10-20KB, and 20-30KB of data respectively; optical disc 4 is a parity disc, storing parity data generated based on the data discs; all optical discs are divided into 5 smallest data units in 2KB stripes, labeled as "data block nm" or "parity block nm".

[0020] Figure 7 is one of the schematic diagrams of logical volume to physical volume mapping in an embodiment of the present invention.

[0021] Figure 8 is a second schematic diagram of the logical volume to physical volume mapping in an embodiment of the present invention.

[0022] Figure 9 is a third schematic diagram of the logical volume to physical volume mapping in an embodiment of the present invention.

[0023] In the diagram: the left side shows the logical volume structure (4 optical discs, with data blocks and check blocks stored in a continuous order), and the right side shows the physical volume structure (4 optical discs, with data blocks and check blocks stored in a distributed manner using a reassembly algorithm). The arrows indicate the mapping relationship between the smallest data unit of the logical volume and the physical volume.

[0024] Figure 10 This is a schematic diagram of concurrent reading from multiple optical drives in an embodiment of the present invention; In the diagram: four optical drives are mounted with physical volumes - optical discs 1-4 respectively, and synchronously read two smallest data units (a total of eight units) from their respective optical discs to splice the target data "0-20KB". The dashed box indicates the synchronous operation of concurrent reading.

[0025] Figure 11 is a schematic diagram of the data recovery steps in an embodiment of the present invention; In the diagram: the damaged data block is 2.1 of physical volume - optical disc 2. After reconstructing its logical location by reverse calculation using the reconstruction algorithm, the three blocks 1.1, 3.1 and 4.1 in the logical volume are called, and the data of 2.1 is recovered based on the XOR operation. Detailed Implementation

[0026] To make the technical problems solved, the technical solutions, and the beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0027] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0028] Furthermore, the use of terms such as "first" and "second" in the embodiments of the present invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated.

[0029] This solution uses the "3+1 mode (4 x 10KB Blu-ray discs, 30KB of data to be stored)" as an example to explain the technical process in detail: Step 1: Create a logical volume Logical volume generation is the foundation for data pre-allocation and redundancy calculation, specifically including: 1.1 Data Allocation: Allocate the 30KB of data to be stored to three data optical discs according to their capacity (10KB per disc): Optical disc 1 (data disc): Stores 0-10KB of data; Optical disc 2 (data disc): Stores 10-20KB of data; Optical disc 3 (data disc): Stores 20-30KB of data; CD 4 (verification disc): Temporarily reserved for storing verification data.

[0030] 1.2 Verification Data Generation: Based on erasure coding algorithms (such as XOR operation), verification data is calculated for each of the 10KB data segments on optical discs 1-3, generating three 2KB verification blocks (6KB in total, with 4KB of reserved redundant space), which are then allocated to optical disc 4. CD 4 (Check Disc): Stores 0-2KB (corresponding to 0-2KB check of CD 1), 2-4KB (corresponding to 2-4KB check of CD 1), 4-6KB (corresponding to 10-12KB check of CD 2), 6-8KB (corresponding to 12-14KB check of CD 2), and 8-10KB (corresponding to 20-22KB check of CD 3).

[0031] 1.3 Striping: All continuous data segments on the optical disc are split into 2KB stripes to obtain the smallest data units and mark them. CD-ROM 1: Data blocks 1.1 (0-2KB), 1.2 (2-4KB), 1.3 (4-6KB), 1.4 (6-8KB), 1.5 (8-10KB); CD-ROM 2: Data blocks 2.1 (10-12KB), 2.2 (12-14KB), 2.3 (14-16KB), 2.4 (16-18KB), 2.5 (18-20KB); CD-ROM 3: Data blocks 3.1 (20-22KB), 3.2 (22-24KB), 3.3 (24-26KB), 3.4 (26-28KB), 3.5 (28-30KB); CD 4: Checksum blocks 4.1 (0-2KB), 4.2 (2-4KB), 4.3 (4-6KB), 4.4 (6-8KB), 4.5 (8-10KB); At this point, the logical volume generation is complete, and its structure is shown in Figure 1 (a schematic diagram of the logical volume generation process).

[0032] Step 2: Generate physical volumes using a recombination algorithm. The reorganization algorithm is the core of this solution. It uses mathematical formulas to "break down" the smallest data unit of the logical volume into multiple optical discs on the physical volume, ensuring that all optical drives can concurrently read and write. Specifically, it includes: 2.1 Application of the mapping formula: Using the conversion formula from logical volume to physical volume, the physical storage location of each smallest data unit is calculated (taking data block 2.1 as an example): Known parameters: LogicIndex=1 (index of CD 2, starting from 0), LogicOff=0 (offset of data block 2.1 in the logical volume), BandSize=2KB, BandNum=5 (5 stripes per disc), DiscNum=4 (4 discs). Calculate PhysicOff: (0 / 2 + 1*5) / 4 *2 = (5) / 4 *2 = 2.5 → rounded to 2KB (optimized for integer operations in practical applications); Calculate PhysicIndex: (0 / 2 + 1*5) %4 = 5%4 = 1; Result: Data block 2.1 is stored in the physical volume at location 2-4KB (PhysicOff=2KB) on optical disc 2 (PhysicIndex=1).

[0033] 2.2 Batch Mapping and Physical Volume Formation: Following the method described above, batch calculations are performed on the smallest data unit of all logical volumes to obtain the storage structure of the physical volumes: Physical Volume - CD-ROM 1: Data blocks 1.1 (0-2KB), 1.5 (8-10KB), 2.4 (16-18KB), 3.3 (24-26KB), 4.2 (2-4KB); Physical Volume - CD-ROM 2: Data blocks 1.2 (2-4KB), 2.1 (10-12KB), 2.5 (18-20KB), 3.4 (26-28KB), 4.3 (4-6KB); Physical Volume - CD-ROM 3: Data blocks 1.3 (4-6KB), 2.2 (12-14KB), 3.1 (20-22KB), 3.5 (28-30KB), 4.4 (6-8KB); Physical Volume - CD-ROM 4: Data blocks 1.4 (6-8KB), 2.3 (14-16KB), 3.2 (22-24KB), 4.1 (0-2KB), 4.5 (8-10KB); At this point, the physical volume generation is complete, and its structure is shown in Figure 2 (a schematic diagram of the mapping from logical volume to physical volume).

[0034] Step 3: Physical volume burning and concurrent read / write of multiple optical drives 3.1 Physical Volume Burning: Burn the four image files of the above physical volume to four Blu-ray discs (physical discs 1-4). 3.2 Multiple optical drive mounting: Four Blu-ray optical drives (optical drives 1-4) are used, each mounting a physical disc 1-4. All optical drives are connected to the host via SATA interfaces and are synchronously controlled by customized software. 3.3 Concurrent Read Example: Reading target data "0-20KB" (corresponding to 10 minimum data units of 2KB): Optical drive 1 synchronously reads 1.1 (0-2KB) and 1.5 (8-10KB) of physical optical disc 1. Optical drive 2 synchronously reads physical optical disc 2's 1.2 (2-4KB) and 2.1 (10-12KB) files. Optical drive 3 synchronously reads physical optical disc 3's 1.3 (4-6KB) and 2.2 (12-14KB) files. Optical drive 4 synchronously reads physical optical disc 4's 1.4 (6-8KB) and 2.3 (14-16KB) files. The remaining two data units (2.4 and 2.5) were read in a second round of concurrent access. Read time: It is only equivalent to the time it takes for a single optical drive to read 2 minimum data units (approximately 0.077 seconds, 2KB / 26MB / s), which is much lower than the read time of a traditional 3+1 RAID5 (approximately 0.15 seconds, with each of the 3 optical drives reading 3-4 units). The concurrent reading process is shown in Figure 3 (Schematic diagram of concurrent reading of multiple optical drives).

[0035] Step 4: Data Recovery Taking "data block 2.1 (10-12KB) of physical volume - optical disc 2 corrupted" as an example, the data recovery process is explained as follows: 4.1 Damage Location: Determine the parameters of the damaged data block in the physical volume: PhysicIndex=1, PhysicOff=0 (10-12KB corresponds to an offset of 0); 4.2 Reverse calculation of logical location: Substitute the conversion formula from physical volume to logical volume: LogicOff = (((0 / 2 *4) +1) %5)*2 = ((0+1)%5)*2=2KB (corresponding to an offset of 2KB in the logical volume); LogicIndex = ((0 / 2 * 4) + 1) / 5 = 1 / 5 = 0 (corresponding to logical volume - CD-ROM 1); Correction verification: Based on the data block marking rules, it is determined that the corrupt data block 2.1 corresponds to "10-12KB data block of CD 2" in the logical volume. It is necessary to call data blocks 1.1 (CD 1 0-2KB), 3.1 (CD 3 20-22KB) and check block 4.1 (CD 4 0-2KB) in the same group. 4.3 Recovery based on erasure coding: Using the XOR operation algorithm, the data blocks 1.1 and 3.1 are processed with the parity block 4.1: 1.1 XOR 3.1 XOR 4.1 = 2.1, generating the original data of the corrupted data block; 4.4 Data Writing: Re-burn the recovered data blocks from 2.1 to the corresponding location on physical volume - optical disc 2 to complete the recovery; The data recovery process is shown in Figure 4 (a schematic diagram of data recovery steps).

[0036] Step 5: RAID Metadata Management The RAID metadata is based on the MDADM superblock design of the Linux system, storing information such as the number of optical discs, stripe size, erasure coding type, and data block mapping table. The RAID metadata is stored in the same location in both the logical and physical volumes (e.g., the last 100KB area of ​​each optical disc) and does not participate in the reconstruction algorithm conversion, ensuring the stability and accessibility of the metadata.

[0037] Example 1 4. Optical Drive 3+1 Mode (Blu-ray Disc Archiving System) Hardware configuration: Blu-ray discs: 4 discs, 25GB per layer, read / write bandwidth 26MB / s; Blu-ray drives: 4 units, SATA interface, supporting 26MB / s read / write speeds; Host: CPU is Intel Xeon E3-1230, memory is 16GB, and storage controller supports multiple optical drives simultaneously; Software: Multi-CD drive control software developed based on the Linux system, integrating reconstruction algorithm and erasure coding module.

[0038] Technical parameter settings: Band size: 2KB; BandNum: 12,500,000 (25GB / 2KB); Erasure coding algorithm: XOR operation; RAID metadata storage location: 100KB area at the end of each optical disc.

[0039] Implementation steps: 3.1 Logical Volume Generation: Allocate 100GB of video data to be archived to 3 data optical discs (discs 1-3) in 25GB increments, generate 25GB of checksum data and allocate it to disc 4, and split it into 12,500,000 minimum data units in 2KB stripes; 3.2 Physical Volume Generation: By recombining the algorithm formula, the physical location of all the smallest data units is calculated, and four 25GB physical volume images are generated; 3.3 Burning and Mounting: Burn the four physical volume images to four Blu-ray discs, mount them on four optical drives and synchronize them with the software; 3.4 Concurrent Read Test: Reading 50GB of video data, with 4 optical drives simultaneously and concurrently, the actual read bandwidth reached 98MB / s (close to the theoretical value of 104MB / s), and the read time was approximately 8.7 minutes; 3.5 Data Recovery Test: One data block (2.1) in CD 2 was artificially damaged. After reverse calculation using the reconstruction algorithm, blocks 1.1, 3.1, and 4.1 were called and the recovery was completed in 0.5 seconds. The data integrity verification passed after recovery.

[0040] Example 2: 12-CD drive 11+1 mode (large-scale Blu-ray archiving system) Hardware configuration: 12 Blu-ray drives (26MB / s), 12 25GB Blu-ray discs, host configuration as above; Technical parameter settings: stripe size 2KB, erasure coding algorithm RS code (supports recovery from 2 damaged disks); Implementation results: The theoretical peak bandwidth of 12 optical drives for concurrent read and write is 312MB / s. The actual time to read 1TB of data is about 56 minutes, which is 13 times more efficient than a single optical drive (about 10 hours). During data recovery, it supports the recovery of two damaged optical discs at the same time, further improving security.

[0041] For those skilled in the art, various other corresponding changes and modifications can be obtained based on the structure and principles disclosed in this invention, and all such changes and modifications fall within the protection scope of this invention.

Claims

1. A method for data error correction and performance optimization based on multiple optical discs, characterized in that, Includes the following steps: S1: Generate a logical volume, which is a temporary data state before data is burned, including data blocks and check blocks; wherein, the data block is the smallest data unit formed by allocating the data to be stored to each data optical disc according to the preset optical disc capacity and then splitting it according to the preset stripe size; the check block is calculated and generated based on the erasure coding algorithm for all the data blocks and allocated to the preset check optical disc. S2: The logical volume is converted into a physical volume using a reorganization algorithm. The physical volume is an image file that can be directly burned to an optical disc. The reorganization algorithm establishes a mapping relationship between each smallest data unit in the logical volume and the storage location in the physical volume using mathematical formulas. This results in the data blocks and check blocks in the physical volume being distributed and stored on multiple optical discs, and each optical disc contains both data blocks and check blocks. S3: Burn the physical volume to multiple corresponding optical discs, and simultaneously mount the multiple optical discs through multiple optical drives to achieve concurrent read and write operations on the physical volume; S4: When any data block in the optical disc is damaged, the location of the damaged data block in the logical volume is calculated in reverse using the reconstruction algorithm. Undamaged data blocks and check blocks corresponding to the damaged data block in other optical discs are called, and the damaged data block is recovered based on the erasure coding algorithm.

2. The data error correction and performance optimization method based on multiple optical discs according to claim 1, characterized in that, In step S1, the preset stripe size is 2KB; the logical volume generation process includes: S11: Obtain the data to be stored and N optical discs, setting N-1 as data optical discs and 1 as a verification optical disc, with each optical disc having a capacity of 10KB; S12: Distribute the data to be stored sequentially to each data optical disc according to the capacity of the data optical disc, forming a continuous data segment; S13: Based on the erasure coding algorithm, calculate and generate a check data segment for consecutive data segments in all data optical discs, and allocate it to the check optical disc; S14: Split the continuous data segments of each data optical disc and the check data segments of the check optical disc into 2KB stripes to obtain multiple minimum data units, labeled as "data block nm" or "check block nm", where n is the optical disc number and m is the sequence number of the minimum data unit in the optical disc, forming a logical volume.

3. The data error correction and performance optimization method based on multiple optical discs according to claim 1, characterized in that, In step S2, the mathematical formula of the reorganization algorithm includes the conversion formula from logical volume to physical volume and the conversion formula from physical volume to logical volume. The formula for converting a logical volume to a physical volume is as follows: PhysicOff=(LogicOff / BandSize+LogicIndexBandNum) / DiscNum*BandSize PhysicIndex=(LogicOff / BandSize+LogicIndexBandNum)%DiscNum The formula for converting a physical volume to a logical volume is as follows: LogicOff=(((PhysicOff / BandSize*DiscNum)+PhysicIndex)%BandNum)*BandSize LogicIndex=((PhysicOff / BandSize*DiscNum)+PhysicIndex) / BandNum Wherein, LogicIndex is the index of the optical disc in the logical volume (starting from 0), LogicOff is the offset of the smallest data unit in the logical volume, DiscNum is the total number of optical discs, BandNum is the number of stripes on a single optical disc, BandSize is the preset stripe size, PhysicIndex is the index of the optical disc in the physical volume (starting from 0), and PhysicOff is the offset of the smallest data unit in the physical volume.

4. The data error correction and performance optimization method based on multiple optical discs according to claim 1, characterized in that, In step S3, the concurrent read / write operation is implemented as follows: when reading target data, all optical drives synchronously read the smallest data unit of the physical volume in their respective mounted optical discs, and convert the read smallest data unit into a continuous data segment of the logical volume through a reorganization algorithm to complete the target data reading; the reading time of the target data is equivalent to the time of a single optical drive reading the maximum value of the number of the smallest data units corresponding to the target data.

5. The data error correction and performance optimization method based on multiple optical discs according to claim 1, characterized in that, In step S4, the specific steps for data recovery are as follows: S41: Determine the location parameters (PhysicIndex, PhysicOff) of the corrupted data block in the physical volume; S42: Substitute the physical volume to logical volume conversion formula to calculate the location parameters (LogicIndex, LogicOff) of the corrupted data block in the logical volume; S43: Extract other undamaged data blocks and corresponding check blocks in the logical volume that are in the same group as the damaged data block; S44: Based on the erasure coding algorithm, perform operations on the extracted undamaged data blocks and check blocks to generate the original data of the damaged data blocks and complete the recovery.

6. A data error correction and performance optimization method based on multiple optical discs according to any one of claims 1-5, characterized in that, The optical disc is a Blu-ray disc, and the read / write bandwidth of the Blu-ray disc is 16MB / s-54MB / s; the number of the multiple optical drives is 4-12.

7. A data error correction and performance optimization method based on multiple optical discs according to any one of claims 1-5, characterized in that, It also includes RAID metadata management steps: the RAID metadata is based on the MDADMsuperblock design of the Linux system, and the storage location of the RAID metadata in the logical volume and physical volume is consistent, and it does not participate in the conversion of the reassembly algorithm.

Images