A method and device for RAID reconstruction based on spaceborne solid-state storage

Abstract

The application discloses a spaceborne configurable solid-state storage system supporting RAID reconstruction and a high-speed data parallel processing method, and relates to the technical field of spaceborne data network of a spacecraft. In order to enhance the flexibility of spaceborne data storage, reduce the complexity of file system interaction, and reduce the high dependence of a data network on the integrity of a solid-state storage system, the application is based on a solid-state storage system, solves the recording and playback problems of multi-channel gigabit interface data, and considers the chip-level reliability difference in the solid-state storage system. The application first proposes and completes a spaceborne flexible configurable design of high-reliability data storage, increases the design of flexible partitioning of a data network and RAID strip reconstruction, and optimizes storage partitioning, increases control information and the mask disk configuration of the solid-state storage system, so that the parallel recording and playback capability of multi-channel gigabit data is realized. Meanwhile, the solid-state storage system has the capability of on-orbit RAID strip reconstruction under the condition that the storage performance of the system is not affected, so as to reduce the interference of a space environment on the spaceborne data network.

Description

Technical Field

[0001] This invention belongs to the field of spacecraft onboard computer system technology, and particularly relates to RAID reconstruction, high-speed parallel processing methods and devices based on onboard solid-state storage. Background Technology

[0002] With the rapid development of technologies such as remote sensing satellites and space science exploration, the amount of data acquired by spacecraft in orbit is growing exponentially, posing unprecedented challenges to the capacity, storage rate, playback rate, and reliability of onboard data storage systems. Traditional onboard storage systems mostly use dedicated storage modules, but in practical applications, they face many bottlenecks such as cost, speed, and size. First, the space environment is extremely harsh, with a large amount of high-energy particle radiation, which can easily cause single-event upsets (SEUs) or even single-event latch-ups (SELs) in storage chips, resulting in data errors or permanent hardware damage. Traditional systems lack effective on-orbit fault tolerance and autonomous repair mechanisms. Once a storage unit fails, it can easily lead to the loss of mission data or even system crashes, seriously affecting the success of space missions. Second, to meet the demand for high-speed data throughput, the system needs to support parallel data recording and playback of multiple high-speed interfaces (such as 10 Gigabit Ethernet, SpaceFibre, etc.). Existing solutions usually use complex multi-level caching and frequent file system interactions to schedule data. This not only introduces huge software overhead and complexity, but also makes the metadata of the file system a single point of failure, resulting in the entire data network being overly dependent on the integrity of the storage system, thus reducing system robustness. Furthermore, on-board missions are multimodal, with different payloads and mission phases dynamically changing the requirements for storage system speed, capacity, and reliability. For example, imaging payloads need to record raw data at ultra-high speeds in data transmission dead zones, while high-speed playback of compressed data is required during data transmission windows. Traditional fixed-configuration storage systems cannot flexibly adapt to these dynamically changing needs, resulting in low resource utilization efficiency.

[0003] While existing technologies employ RAID (Redundant Array of Independent Disks) to improve reliability, these solutions are primarily designed for terrestrial applications. In the space environment, directly applying traditional RAID schemes faces challenges such as long rebuild times, high computational resource consumption, a sharp performance degradation during rebuilding, and an inability to effectively handle diverse concurrent errors caused by space radiation. Furthermore, spaceborne systems have strict limitations on power consumption, size, and mass, making it difficult to deploy high-performance general-purpose computing units for complex RAID calculations. Summary of the Invention

[0004] To address the aforementioned issues, this invention proposes a RAID reconstruction and high-speed parallel processing method and apparatus based on spaceborne solid-state storage. The apparatus includes several hardened disk modules and a main control FPGA module, enabling the recording and playback of multi-channel 10 Gigabit interface data. Considering the chip-level reliability differences within the solid-state storage system, it innovatively proposes a flexible and configurable design for high-reliability spaceborne data storage. This invention achieves parallel recording and playback of multi-channel 10 Gigabit data by designing flexible data network partitioning and RAID stripe reconstruction. Through optimized storage partitioning, increased control information, and mask configuration of the solid-state storage system, it ensures that the system's storage performance is not affected while possessing the ability to rebuild RAID stripes in orbit, thereby reducing interference from the space environment on the spaceborne data network.

[0005] A first aspect of the present invention provides a RAID reconstruction and high-speed parallel processing apparatus based on spaceborne solid-state storage, comprising:

[0006] The main control FPGA module is used to complete high-speed parallel data processing, storage control, build RAID mode of solid-state storage system and configure and rebuild the RAID in orbit.

[0007] Several hardened disk modules are used to process the service data sent by the main control processing chip, and to process the recorded transmission data and then play it back to the main control processing chip.

[0008] Preferably, the hardened disk module uses a high-speed SATA3.0 interface and is based on particle device-level differential lifetime vector storage performance data. The performance data includes at least wear leveling data, address translation data, and bad block management data.

[0009] Preferably, the main control processing FPGA module has a number of 10 Gigabit data recording capabilities, and the bandwidth of the main control processing FPGA module is greater than 80 Gbps.

[0010] A second aspect of the present invention provides a RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage, applied to the RAID reconstruction and high-speed parallel processing apparatus based on spaceborne solid-state storage described in any one of the preceding claims, comprising:

[0011] By optimizing storage partitions, adding control information, and configuring masking for solid-state storage systems, parallel recording and playback of multi-channel 10 Gigabit interface data is achieved, and on-orbit RAID stripe reconstruction is supported.

[0012] Based on the control information, a mask is configured for each reinforcement disk module.

[0013] Preferably, the process also includes the step of dynamically rebuilding the RAID in orbit, specifically including:

[0014] HEADER data is generated based on the transmitted data, and the HEADER data is written into the stripe of the current RAID. The stripe-level parity data is calculated and written to the parity disk.

[0015] The main control FPGA module is driven by playback instructions to read each RAID stripe in a preset order, and starts RAID reconstruction or obtains status data based on the HEADER data.

[0016] Preferably, the steps of initiating RAID reconstruction or obtaining status data based on the HEADER data specifically include:

[0017] During on-orbit operation, the main control FPGA module scans the HEADER data of each RAID stripe and verifies the HEADER data to obtain the verification result. If the verification result is determined to be abnormal, the HEADER data is used to locate the failed data block and drive RAID reconstruction. During the RAID reconstruction process, the main control FPGA module writes the reconstruction data to the log area to obtain status data, and constructs RAID control parameters based on the HEADER data and the status data.

[0018] Preferably, the step of obtaining the corresponding spaces for the record area, rebuild cache area, log area, and spare area based on the mapping data and using a partitioning strategy further includes:

[0019] Based on high-speed and variable-length business data, intermediate application data is obtained by unpacking and framed to obtain fixed-length data frames. The intermediate application data may have different lengths for different users.

[0020] Based on the fixed-length data frame, the storage space is aligned and divided into units that are integer multiples of the smallest storage block of the solid-state storage system.

[0021] Preferably, the control information includes at least status data, HEADER data, and control parameters during the data writing and playback process.

[0022] A third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage as described in any of the preceding claims.

[0023] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage as described in any of the preceding claims.

[0024] Due to the adoption of the above technical solutions, this invention has the following advantages and positive effects compared with existing technologies: High reliability: Online RAID reconstruction and configurable disk masking strategies implemented through FPGA hardware can effectively cope with storage unit failures caused by spatial radiation, achieve data self-healing, and greatly improve the task success rate. High performance and low latency: By adopting a preset length frame direct mapping and multi-channel parallel access architecture, the software overhead of traditional file systems is avoided, achieving stable and ultra-high read / write bandwidth and deterministic latency. Attached Figure Description

[0025] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, wherein:

[0026] Figure 1 One embodiment of the storage space partitioning of the device in this invention;

[0027] Figure 2 This is an embodiment of the mapping between physical and logical addresses in this invention;

[0028] Figure 3 This is an illustration of one embodiment of the present invention based on framing transmitted data;

[0029] Figure 4 This is an illustration of one embodiment of the present invention based on the unpacking of transmitted data. Detailed Implementation

[0030] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer from the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise ratios, and are only used to facilitate and clarify the illustration of the embodiments of the present invention.

[0031] 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.

[0032] First Embodiment

[0033] See Figure 1 and Figure 2 The first aspect of the present invention provides a RAID reconstruction and high-speed parallel processing apparatus based on spaceborne solid-state storage, comprising:

[0034] The main control FPGA module is used to complete high-speed parallel data processing, storage control, build RAID mode of solid-state storage system and configure RAID, and rebuild in orbit.

[0035] Several hardened disk modules are used to process the business data sent by the main control processing chip, and to process the recorded transmission data and then play it back to the main control processing chip.

[0036] The ruggedized disk module is the basic storage medium of this device, and it is a storage module independently constructed using industrial-grade or military-grade NAND Flash chips. Each ruggedized disk module uses a SATA 3.0 high-speed interface with a speed of up to 6Gbps. The SATA 3.0 high-speed interface of the ruggedized disk module is used to transmit business data and acquire status data during transmission. The status data includes at least bad block information, lifetime status, temperature, and read / write error counts. The above process realizes data interaction and status monitoring. By working in parallel with the ruggedized disk modules, a large-capacity storage pool is realized.

[0037] The main control FPGA module uses industrial-grade or higher chips with high-speed serial transceivers. Internally, it implements the following key functional engines through hardware logic programming: responsible for processing, buffering, and routing multiple incoming 10 Gigabit data streams (4 streams of 10Gbps). The main control FPGA module has a number of 10 Gigabit data recording capabilities, and its bandwidth is greater than 80Gbps. The main control FPGA module includes a storage control engine for fine-grained management of the solid-state storage system, including physical address mapping, read / write command issuance, wear leveling, and bad block management. The main control FPGA module also includes a RAID management and rebuild engine: this is the core of the invention. The RAID management and rebuild engine implements RAID stripe assembly and data storage within the main control FPGA module. When a failure is detected in any hardened disk module through status feedback, the RAID management and rebuild engine drives the failure disk to rebuild the RAID. Because data is unpacked and framed according to the storage frame format at the entry point, and processed in 8KB data frames with 1MB solid-state storage granularity at the back end, the internal mapping relationships at each level are tightly coupled with the RAID configuration. It supports arbitrary adjustment of RAID configuration, enabling on-orbit recovery of storage system functions. The main control FPGA module is also equipped with a management engine: it receives configuration commands from programmable / ground commands and dynamically adjusts the system's operating mode, such as mask configuration updates, RAID switching, and storage partition size adjustments, thereby achieving flexible and configurable system resources.

[0038] The device in this embodiment can be configured as follows: Main control FPGA module: A Xilinx Virtex-7 series XC7VX690T-FFG1761 chip is selected. This chip has abundant logic resources and 36 high-speed serial transceivers, meeting the requirements of aerospace applications. Rugged disk module: Eight ruggedized SATA disks are used. Each disk consists of multiple NAND Flash chips and is connected to the main control FPGA module via a SATA 3.0 interface controller. The total storage capacity is designed to be 16 Tbit (effective user capacity). Interface: The main control FPGA module provides four 10 Gigabit Ethernet (10Gbps) interfaces for payload data reception and playback. This embodiment does not impose configuration limitations and other configurations are also possible.

[0039] Preferably, the hardened disk module adopts a high-speed SATA3.0 interface and is based on the differential lifetime vector storage performance data of the component device level. The performance data includes at least wear leveling data, address translation data, and bad block management data.

[0040] The ruggedized disk module adopts a high-speed SATA 3.0 interface, significantly shortening the data read / write cycle. Especially during RAID reconstruction, it can quickly complete cross-disk data migration and verification, meeting the high-speed access requirements of onboard equipment for massive amounts of data. Based on the standardized protocol of the SATA 3.0 interface, it supports flexible cascading of multiple ruggedized disk modules, facilitating the construction of highly redundant RAID arrays (such as RAID 5 / 6), and is compatible with existing storage architectures. It provides a stable bus management foundation for the main control FPGA module, reducing system integration complexity. The low transmission latency of the SATA 3.0 interface, combined with the parallel processing capabilities of the main control FPGA module, can efficiently support the immediate injection and backhaul of urgent data in onboard missions, while accelerating the synchronous writing of verification data during RAID reconstruction, ensuring the timeliness of critical operations.

[0041] Preferably, the main control FPGA module has a number of 10 Gigabit data recording capabilities and a bandwidth greater than 80 Gbps.

[0042] The main control FPGA module has a 10-gigabit-level data recording capability and a bandwidth of >80Gbps, which meets the real-time acquisition and storage requirements of spaceborne equipment for large-scale data such as remote sensing images and scientific experiments.

[0043] Second Embodiment

[0044] A second aspect of the present invention provides a RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage, and an apparatus for RAID reconstruction and high-speed parallel processing based on spaceborne solid-state storage applied to any one of the above, comprising:

[0045] By optimizing storage partitions, adding control information, and configuring masking for solid-state storage systems, parallel recording and playback of multi-channel 10 Gigabit interface data is achieved, and on-orbit RAID stripe reconstruction is supported.

[0046] Based on control information, mask configuration is performed on each hardened disk module.

[0047] This embodiment describes the process based on a hardened disk module and a main control FPGA module: Data from four 10 Gigabit cameras (totaling 40 Gbps) is input to the main control FPGA module. The FPGA module converts the data into data frames of a preset length. These data frames have headers and are striped and written in parallel to the corresponding hardened disk modules. In this embodiment, the data frame length is 8KB, and there are eight hardened disk modules. Other parameters can also be used; this embodiment is not limited. During on-orbit monitoring, if an abnormal link status is detected in a hardened disk module, a masking and reconstruction process is triggered. Specifically, the main control FPGA module transmits telemetry data back to the satellite computer. The satellite computer receives the masking instruction and performs the following operation: logically masking the parameters corresponding to the hardened disk module with the abnormal status. The hardened disk is enabled for recording and playback tasks. When the satellite flies over the ground station, it retrieves the data from the reconstruction buffer and transmits the data to the data transmission system at high speed via the 10 Gigabit interface. Through this embodiment, the present invention successfully achieves reliable storage, autonomous management, and rapid fault recovery of high-speed data in harsh space environments, meeting the stringent requirements of modern aerospace missions for onboard storage systems. The system is composed of a hardware platform and firmware logic working together, with its core being the organic integration of a high-performance main control FPGA and multiple ruggedized disk units.

[0048] Preferably, the process also includes the step of dynamically rebuilding the RAID in orbit, specifically including:

[0049] HEADER data is generated based on the transmitted data and written to the current RAID stripe. Band-level parity data is calculated and written to the parity disk.

[0050] The FPGA module, driven by playback instructions, reads each RAID stripe in a preset order and starts rebuilding the RAID or obtains status data based on the HEADER data.

[0051] By utilizing HEADER data (including address mapping and verification identifiers), the main control FPGA module can automatically identify the logical structure of RAID stripes, accurately locate failed data blocks, and trigger the reconstruction process without ground intervention, significantly improving the autonomous operation and maintenance capabilities of onboard equipment. Writing stripe-level verification data independently to a verification disk creates a dual-insurance mechanism: it verifies data correctness through verification values ​​and can fully recover original information in the event of a single disk failure, effectively addressing bit flipping or media aging issues caused by space radiation. A striped reading strategy driven by playback instructions allows the FPGA to parse RAID data stripe by stripe in a preset order. Guided by HEADER data, it can quickly skip damaged areas and prioritize the reconstruction of critical data segments, shortening system downtime. Through the association and binding of HEADER data and status data, the ground station can obtain RAID health status, reconstruction progress, and verification results in real time, achieving end-to-end visual monitoring and providing complete traceability for fault diagnosis. This solution does not rely on specific disk numbers or fixed topologies; it reconstructs the RAID logical view by dynamically parsing HEADER data, facilitating automatic adjustment of array configuration after future disk expansion or replacement of failed disks, adapting to the needs of long-term on-orbit missions.

[0052] Preferably, the steps for initiating RAID reconstruction or obtaining status data based on HEADER data specifically include:

[0053] During on-orbit operation, the main control FPGA module scans the HEADER data of each RAID stripe and verifies the HEADER data to obtain the verification result. If the verification result is determined to be abnormal, the HEADER data is used to locate the failed data block and drive the RAID reconstruction. During the RAID reconstruction process, the main control FPGA module writes the reconstruction data to the log area to obtain status data, and constructs RAID control parameters based on the HEADER data and status data.

[0054] By scanning and verifying the HEADER data strip by strip, the specific location of failed data blocks can be accurately identified, avoiding blind reconstruction of the entire system. Only damaged areas are repaired, significantly reducing the amount of invalid data copied and shortening reconstruction time. During reconstruction, temporary data is written to a dedicated log area (partitioning strategy as described in claim 7), achieving the following dual protections: if reconstruction is interrupted, it can be restored to a consistent state through log replay; the reconstruction path and intermediate data are fully recorded, facilitating ground station review and verification of data accuracy. Based on the joint analysis of HEADER metadata and real-time status data, the main control FPGA can dynamically adjust RAID configuration parameters to achieve an adaptive performance-reliability balance. The log area is used to temporarily store reconstruction data, avoiding frequent direct erasure and rewriting of the business data area; data distribution is intelligently scheduled based on the historical access frequency recorded by the HEADER, mitigating NAND flash wear; and by statistically analyzing HEADER verification failure modes, early warnings are given for storage units nearing the end of their lifespan.

[0055] See Figure 3 and Figure 4 Preferably, the step of obtaining the corresponding spaces for the record area, rebuild cache area, log area, and spare area based on the mapping data and using a partitioning strategy further includes:

[0056] Based on high-speed and variable-length business data, intermediate application data is obtained by unpacking and framed to obtain fixed-length data frames. There are cases where the length of intermediate application data is inconsistent among users.

[0057] Storage space is aligned and divided based on fixed-length data frames, using integer multiples of the smallest storage block size in the solid-state storage system as units.

[0058] Business data is broken down into data frames of preset length, achieving normalized processing of data units. This design enables the main control FPGA to perform parallel operations at a fixed granularity, avoiding the parsing complexity of traditional variable-length data packets and significantly improving data flow efficiency. A storage alignment mechanism forces the physical address of each data frame to be strictly aligned with the boundary of the smallest storage block, eliminating space waste caused by cross-block storage. Especially in block devices such as NAND flash memory, this design can reduce the generation of invalid pages and increase the actual usable storage capacity. Write amplification effect suppression: Aligned storage avoids additional erase operations caused by cross-block writes, slowing down the P / E cycle consumption of NAND flash memory; Garbage collection friendliness: The regular data block distribution simplifies the migration workload during garbage collection, further reducing unnecessary erase / write cycles. Atomicity write guarantee: Each data frame is completely contained within a single storage block, preventing data tearing caused by unexpected power outages; see [reference in this embodiment] Figure 4Format: MPDU header, 2 bytes, indicating the offset position of the EPP encapsulation packet header within the current VCDU data area: if there is no EPP packet header in the current frame, the MPDU header is 0x7FF; if the current frame's EPP packet header immediately follows the MPDU, the MPDU is 0. The EPP encapsulation packet consists of EPP_HDR + DATA. EPP_HDR, 8 bytes, contains the following: EA_00_len(2 bytes)_00_00_00_43. The encapsulation packet length is the entire IP packet length + 8 bytes. The packet identifier is of two types: 0xEA when the encapsulation packet contains valid data; 0xE0 when the encapsulation packet contains padding data. The encapsulation packet data content is the received Ethernet packet (excluding the 16 bytes of the MAC header: DST_MAC(6 bytes) + SRC_MAC(6 bytes) + IP_TYPE(2 bytes)).

[0059] Preferably, the control information includes at least the status data, HEADER data, and control parameters during the data writing and playback process.

[0060] By integrating recorded / replay status data, HEADER metadata, and control parameters, the entire data flow is observable, providing ground stations with complete storage behavior traceability capabilities. Anomaly alarms are triggered based on status data, and failure areas are precisely located using HEADER data. This drives control parameters to automatically perform partial reconstruction or redundancy switching, shortening system recovery time. Historical status statistics are linked with control parameters to dynamically adjust partition resource allocation, verification strength, and load balancing strategies to adapt to storage needs at different task stages. Transaction-level atomicity guarantees, cross-cycle data comparison, and HEADER version verification ensure consistency of multi-disk synchronous writes, reducing the risk of data corruption under spatial radiation environments. Remote configuration distribution and automated diagnostic report generation are supported, reducing the need for manual intervention. Furthermore, control parameters constrain illegal operations, improving system reliability.

[0061] Third Embodiment

[0062] A third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage as described above.

[0063] The processor is dedicated to executing critical algorithms such as RAID verification and header parsing, working in conjunction with tiered storage scheduling to achieve high-density data throughput and low-latency response. Program-based protection, ECC error correction, and a watchdog mechanism mitigate space radiation interference, ensuring the safety of control parameters and core logic during on-orbit missions. Software-defined storage strategies allow for flexible adjustments to RAID levels, partition ratios, and verification strength to meet the data management requirements of different scientific payloads. Online firmware upgrades, self-diagnosis of health status, and redundant cold backup activation are supported, extending equipment lifespan and reducing ground station maintenance complexity.

[0064] Fourth embodiment

[0065] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage as described above.

[0066] Computer-readable storage media ensures that core commands such as RAID rebuilds are not lost under extreme environments, guaranteeing the safe startup and execution of basic functions of the onboard system. By updating the instruction set in the storage media, RAID algorithms or partitioning strategies can be dynamically adjusted without modifying the hardware, adapting to changes in mission requirements.

[0067] In the description of this application, it should be noted that the terms "inner" and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use. They are used only for the convenience of describing this application and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0068] It should also be noted that, unless otherwise explicitly specified and limited, the terms "setup" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0069] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific identification content executed by the system and device described above can be referred to the corresponding process in the foregoing method embodiments.

[0070] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they shall still fall within the protection scope of the present invention.

Claims

1. A RAID reconstruction and high-speed parallel processing device based on spaceborne solid-state storage, characterized in that, include: The main control FPGA module is used to perform high-speed parallel data processing, storage control, RAID mode construction of solid-state storage systems, RAID configuration, and on-orbit reconstruction. The RAID management and reconstruction engine in the main control FPGA module implements RAID stripe construction and data storage. When any hardened disk module fails, the RAID management and reconstruction engine drives the masking of the failed disk and rebuilds the RAID. The main control FPGA module also configures a management engine: receiving configuration commands from programmable commands / ground commands, and dynamically adjusting the system's operating mode, including any one of masking disk configuration updates, RAID switching, and storage partition size adjustments. Several hardened disk modules are used to process the service data sent by the main control processing chip, and to process the recorded transmission data and then play it back to the main control processing chip.

2. The RAID reconstruction and high-speed parallel processing device based on spaceborne solid-state storage according to claim 1, characterized in that, The hardened disk module uses a high-speed SATA3.0 interface and is based on particle device-level differential lifetime vector storage performance data. The performance data includes at least wear leveling data, address translation data, and bad block management data.

3. The RAID reconstruction and high-speed parallel processing device based on spaceborne solid-state storage according to claim 1, characterized in that, The main control FPGA module has a number of 10 Gigabit data recording capabilities, and the bandwidth of the main control FPGA module is greater than 80 Gbps.

4. A RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage, applied to the RAID reconstruction and high-speed parallel processing apparatus based on spaceborne solid-state storage as described in any one of claims 1-3, characterized in that, include: By optimizing storage partitions, adding control information, and configuring masking for solid-state storage systems, parallel recording and playback of multi-channel 10 Gigabit interface data is achieved, and on-orbit RAID stripe reconstruction is supported. Based on the control information, a mask is configured for each reinforcement disk module.

5. The RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage according to claim 4, characterized in that, This also includes the steps for dynamically rebuilding the RAID in orbit, specifically: HEADER data is generated based on the transmitted data, and the HEADER data is written into the stripe of the current RAID. The stripe-level parity data is calculated and written to the parity disk. The main control FPGA module is driven by playback instructions to read each RAID stripe in a preset order, and starts RAID reconstruction or obtains status data based on the HEADER data.

6. The RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage according to claim 5, characterized in that, The specific steps for initiating RAID reconstruction or obtaining status data based on the HEADER data include: During on-orbit operation, the main control FPGA module scans the HEADER data of each RAID stripe and verifies the HEADER data to obtain the verification result. If the verification result is determined to be abnormal, the HEADER data is used to locate the failed data block and drive RAID reconstruction. During the RAID reconstruction process, the main control FPGA module writes the reconstruction data to the log area to obtain status data, and constructs RAID control parameters based on the HEADER data and the status data.

7. The RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage according to claim 5, characterized in that, The steps of obtaining the corresponding spaces for the record area, rebuild cache area, log area, and spare area based on the mapping data and using a partitioning strategy further include: Based on high-speed and variable-length business data, intermediate application data is obtained by unpacking and framed to obtain fixed-length data frames. The intermediate application data may have different lengths for different users. Based on the fixed-length data frame, the storage space is aligned and divided into units that are integer multiples of the smallest storage block of the solid-state storage system.

8. The RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage according to claim 4, characterized in that, The control information includes at least the status data, HEADER data, and control parameters during the data writing and playback processes.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage as described in any one of claims 4-8.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements the RAID reconstruction and high-speed parallel processing method based on spaceborne solid-state storage as described in any one of claims 4-8.

Images