Cloud-based safe destruction system for archive data

Abstract

The invention relates to the technical field of data destruction, in particular to a cloud-based safe destruction system for archival data, which comprises a fingerprint chain construction module, a compression boundary analysis module, a mapping path checking module, a disturbance destruction control module and a boundary fracture identification module. According to the method, through joint processing of each field label, the sequence index and the byte length in the field structure set, the static mapping path is screened and extracted by utilizing double conditions that the pointing frequency is zero and the path span number is consistent in the path identification, so that repeated writing or invalid data residue is avoided; in the destroying process, through content disturbance and logic offset writing of a middle pointer number, visual evaluation of a closed-loop effect of a destroying chain and combined propulsion of multiple structural level operations and a path control mechanism are realized, and structural clearness, path breakage and link irreversibility of a data destroying target are realized; and the irrecoverability and the security integrity of the data termination process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data destruction, and in particular to a cloud-based file data security destruction system. Background Art

[0002] The field of data destruction technology includes various technical means to completely delete and irreversibly process electronic information, ensuring that sensitive data in digital carriers cannot be illegally recovered or used after the end of their service life. The core content of this technical field is to operate the physical structure and logical structure of data in the storage medium, and to achieve permanent invalidation of information through overwriting, zeroing, rewriting, etc. Data destruction can be applied to local storage devices such as hard disks and solid-state drives, as well as network platforms and cloud storage systems. In particular, with the popularization of cloud computing and remote work, the demand for data destruction in distributed environments is increasing, prompting this field to cover technical directions such as remote command control, destruction strategy formulation, multi-tenant permission management, and data lifecycle end control. It also collaborates with modules such as identity authentication, log tracking, and permission auditing in the system security framework to build a complete data termination mechanism.

[0003] Among them, the cloud-based archival data security destruction system refers to the destruction control of electronic archival data achieved through the cloud platform, especially the unified data termination processing solution in a multi-terminal access and multi-tenant environment. The technical matters targeted by this patent subject include: first, the logical positioning and retrieval of archival files in cloud storage; second, the completion of data overwriting operations on target files through instruction scheduling; third, the use of distributed storage management strategies to determine destruction permissions and operation nodes; fourth, multiple write operations on storage blocks according to the set destruction strategy; and fifth, the use of a logging mechanism to irreversibly audit the destruction process. The system issues destruction tasks through the control interface of the cloud platform, executes data rewriting commands in storage nodes, and combines preset destruction rules to achieve structural reconstruction of file data at the physical layer, making archival data irrecoverable.

[0004] In the existing technology, logical positioning and rewriting instructions are commonly used to perform destruction operations. Although they can perform surface coverage of file contents, in an environment with complex compression structures and path mappings, it is difficult to penetrate the deep logical relationships of the data structure layer, resulting in some compressed segments remaining without being accurately hit, forming a hidden danger of data residue. In terms of path management, distributed nodes are relied upon to execute rewrite commands. There is no effective identification mechanism for the multiple pointing relationships existing in the logical path, which easily leads to redundant path writes or path breakpoint backflow, thereby increasing the probability of data recovery. For data change records, common systems fail to establish a mapping logic between structural versions and field changes, ignoring the impact of structural differences on the direction of the destruction chain, resulting in unclear boundaries, overlapping or missing rewrite ranges during the destruction of multiple versions. For example, in a cloud platform, different tenants may modify the same field structure. If there is a lack of field-level difference tracking, mixed instructions of the old and new versions will not be fully covered. The data will be mistakenly identified as destroyed at the logical layer, but in fact, remnants will still remain, causing data security risks and affecting the destruction compliance assessment. Summary of the Invention

[0005] In order to solve the problem that the existing technology can cover the surface of the file content, but in the environment of complex compression structure and path mapping, it is difficult to penetrate the deep logical relationship of the data structure layer, resulting in some compressed segments remaining without being accurately hit, forming a hidden danger of data residue. In terms of path management, it relies on distributed nodes to execute rewrite commands. There is no effective identification mechanism for the multiple pointing relationships in the logical path, which easily leads to redundant path writing or path breakpoint reflux, thereby increasing the probability of data recovery. For data change records, common systems fail to establish a mapping logic between structure versions and field changes, and ignore the impact of structural differences on the direction of the destruction chain, resulting in unclear boundaries, overlapping or omissions in the destruction process of multiple versions. For example, in a cloud platform, different tenants have modification operations on the same field structure. If there is a lack of field-level difference tracking, there will be a situation where the old and new versions are mixed and the instructions are not fully covered. The data is mistakenly identified as destroyed at the logical layer, but in fact there are still traces left, causing data security risks and affecting the destruction compliance assessment. The embodiment of the present invention provides a cloud-based archival data security destruction system. The technical solution is as follows: In one aspect, a cloud-based archival material security destruction system is provided, comprising: The fingerprint chain construction module processes the field sequence based on the field structure set in the cloud archive data to generate a structural fingerprint. It then compares the field quantity change value, field rearrangement displacement, and field addition and deletion cross-mark value of the structural fingerprint to generate the start and end positioning segments of the destruction chain. The compression boundary parsing module extracts the compression start address, logical offset bit and compression ratio tag value of the fragment in the corresponding segment compression index based on the start and end positioning segments of the destruction chain, performs distance landing point attribution judgment, and generates a compression segment offset compensation structure; The mapping path checking module calls the compressed section offset compensation structure, extracts the starting address number, the number of times pointed to, and the path span position value, filters and marks them as mapping static paths, and generates a path chain static mapping mark table; The disturbance destruction control module calls the random content record number and the archive data destruction scheduling period based on the static mapping mark table of the path chain, writes the content to the intermediate node and shifts the address offset to the logical number, removes the record number of the address before the offset in the mapping path table, and generates a path disturbance break tracking log.

[0006] As a further solution of the present invention, the start and end positioning segments of the destruction chain include the structural change boundary position, the field change behavior mapping index, and the version addition and deletion cross-record group; the compression segment offset compensation structure includes the offset distance attribution identifier, the compression rate correspondence mapping, and the logical address offset threshold positioning item; the path chain static mapping mark table includes the path stability identification number, the unchanged pointing structure set, and the path span equivalence mark item; the path disturbance fracture tracking log includes the fracture path record number, the disturbance write content record number, and the logical offset migration record information.

[0007] As a further solution of the present invention, the fingerprint chain construction module includes: The field extraction submodule extracts the field level label, sequence index and field name byte length based on the field structure set in the cloud archive data, aligns the field level label and sequence index, filters and sorts the field name byte length, calls the field structure sequence and the level label set for sequence comparison, and generates the field sequence value. The structure identification submodule calls the field arrangement sequence value and the field level label set, combines the field unit identifier according to the field arrangement order and the level segmentation relationship, performs aggregation processing on continuous segments, divides the field segment boundary mark value and generates an aggregated segment sequence, and generates a structure fingerprint density; The change identification submodule extracts the field quantity change value and the field rearrangement displacement based on the structure fingerprint density and the version evolution data of the field structure set, determines whether the addition of fields and the deletion of fields exist simultaneously in the same change group, filters the version index values that meet the conditions, and generates the start and end segment values of the destruction chain.

[0008] As a further solution of the present invention, the compression boundary parsing module includes: The fragment extraction submodule extracts the fragment compression start address and logical offset bit in the corresponding compression index based on the start and end segment values of the destruction chain, calls the absolute distance between the compression start address and the logical offset bit, evaluates the displacement mapping relationship within the compression segment, and generates the compression segment logical offset distance value; The threshold mapping submodule calls the compression segment logical offset distance value and the compression ratio label value corresponding to the compression index segment, obtains the offset threshold interval corresponding to the compression ratio label, compares the landing position of the distance value within the offset threshold interval, determines the belonging segment, and generates a distance belonging flag value; The sequence adjustment submodule extracts the assigned segment flag index and the logical offset bit according to the distance assigned flag value and the compressed segment logical offset distance value, calls the compressed start address to rebuild the offset position table within the segment, calculates the offset position compensation value, and arranges the compensation order to generate a compressed segment offset compensation structure.

[0009] As a further solution of the present invention, the formula for calculating the offset position compensation value is: ; in, Represents the offset position compensation value, Represents the total number of segments calculated, Representative The segment's belonging flag value, Representative The logical offset distance value of the segment, Representative The compensation coefficient of the segment, Representative The reconstruction coefficient of the segment, Representative The starting position value of the segment, Representative The end position value of the segment.

[0010] As a further solution of the present invention, the mapping path checking module includes: The path extraction submodule extracts the starting address number, the pointing count value, and the path span position value based on the compressed segment offset compensation structure, calculates the offset value between the starting address number and the path span position value, detects whether the pointing count value is zero, combines the offset value and the pointing count value, and generates a path positioning offset; The path identification submodule calls the path positioning offset, performs number verification on the data item with the pointing count value of zero, compares the path span position value with the difference between the starting address number, determines whether the difference is consistent with the span number, and generates a consistency identification sequence value; The static marking submodule numbers and classifies the path span position value set according to the consistency identification sequence value, assigns static path marks to path items with the same number, allocates a mark index value to each group of static path mark items, and establishes a path chain static mapping mark table.

[0011] As a further solution of the present invention, the disturbance destruction control module includes: The node selection submodule locates the logical number of the middle position of each path chain based on the middle pointer number of the static path chain in the path chain static mapping mark table, obtains the corresponding address number in the path chain static mapping mark, selects the path set whose middle pointer number is within the scheduling period, calculates the score value of the path set, and generates the middle node address set; The random write submodule calls the intermediate node address set, generates a random write value sequence for each intermediate node according to the random content record number, detects the match between each write value and the logical number of the node address offset segment, determines that the write position is within a range of the backward address offset, obtains the new position number after writing the random value content to the node, and generates a disturbed write position value; The path removal submodule extracts the corresponding logical number original address according to the disturbance write position value, locates the record number corresponding to the address, removes the record number and reconstructs the mapping path table index sequence, analyzes the path break trend and the logical number mapping relationship under the scheduling cycle, and establishes a path disturbance break tracking log.

[0012] As a further solution of the present invention, the formula for calculating the score value of the path set is: ; in, Represents the score value of the path set, Represents the path chain The logical number of the node, Representative The target position number of each node in the scheduling period, Representative The weight value of the path chain, Representative The start time of the path chain, Representative The end time of the path chain, is the number of paths.

[0013] As a further solution of the present invention, the system further includes a boundary fracture identification module: The boundary fracture identification module extracts the fracture node number from the path disturbance fracture tracking log and compares it with the continuous index value in the start and end positioning section of the archive destruction chain, determines the logical segment index not covered by the fracture number in the index, marks it as a chain coverage fault, and generates a destruction chain fracture coverage status mapping table; The destruction chain break coverage status mapping table includes uncovered logical paragraph index groups, broken node coverage mapping conditions, and structure coverage integrity comparison results.

[0014] As a further solution of the present invention, the boundary fracture identification module includes: The node extraction submodule extracts the set of broken node numbers based on the path disturbance fracture tracking log, extracts the continuous index values in the start and end positioning segments of the archive data destruction chain, arranges the broken node numbers and continuous index value lists in order and classifies the mapping relationship, and generates a broken node coverage index group; The position comparison submodule calls the broken node coverage index group, performs a node number matching operation on the continuous index value list, identifies the index position not matched by the broken node number, extracts the logical number as the residual index item, and obtains the segment uncovered number set; The fault marking submodule determines the belonging segments in the destroyed chain location area according to the chain segment uncovered number set and the continuous logical number sequence, filters the logical segments including the uncovered numbers in the belonging segments, marks them as chain coverage faults, and generates a chain segment fault number sequence; The state identification submodule calls the segment fault number sequence, evaluates the mapping index between the segment number and the fault state flag value, adds the corresponding fault state attribute to each segment logical number, arranges them in a logical order, and generates a destruction chain fracture coverage state mapping table.

[0015] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: By jointly processing the field labels, sequential indexes, and byte lengths of each field structure set, a fingerprint chain capable of sensing structural changes is constructed. Combined with cross-marking of field additions and deletions, this system identifies version differences, significantly improving destruction targeting accuracy under dynamic data structure changes. The distance between the compressed segment logical offset and the compression start address is quantified and matched to compression ratio thresholds. This system then constructs a compressed segment attribution identification system with a compensation mechanism, effectively improving the control of physical address mapping errors during the destruction process. In path identification, a dual-conditional screening process, where the number of pointers is zero and the path span number is consistent, is used to extract static mapping paths, preventing duplicate writes or residual invalid data. During the destruction process, structural perturbations are created by perturbing the content of intermediate pointer numbers and writing logical offsets, removing path mapping record numbers and making path chain break locations traceable, thereby reducing the technical barriers to data recovery. By comparing the coverage of path break nodes with logical segment indexes, uncovered indexes are selected as chain break mapping items. A status table of chain coverage integrity is constructed, enabling a visual assessment of the effectiveness of the destruction chain closure. The joint promotion of multiple structural-level operations and path control mechanisms can achieve structural clarification, path breaking and link irreversibility of data destruction targets in mixed scenarios with multiple versions, multiple paths and multiple compression segments, thereby improving the irreversibility and security integrity of the data termination process. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 Schematic diagram of a cloud-based archival data security destruction system provided by an embodiment of the present invention; Figure 2 Schematic diagram of the system framework of the present invention; Figure 3 This is a flowchart of the fingerprint chain construction module in the present invention; Figure 4 This is a flow chart of the compression boundary analysis module in the present invention; Figure 5 This is a flowchart of the mapping path troubleshooting module in the present invention; Figure 6 This is a flow chart of the disturbance destruction control module in the present invention; Figure 7 This is a flow chart of the boundary fracture identification module in the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0019] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0020] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0021] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0022] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0023] The embodiment of the present invention provides a cloud-based file data security destruction system, such as Figure 1-2 The schematic diagram of the cloud-based archival data security destruction system shown in the figure includes: The fingerprint chain construction module is based on the field structure set in the cloud archive data. It calls the field level label, sequence index and field name byte length, processes the field sequence to generate a structural fingerprint, compares the field quantity change value, field rearrangement displacement and field addition and deletion cross-mark value of the structural fingerprint, and lists the version index with both addition and deletion in the change group into the change boundary list to generate the start and end positioning segments of the destruction chain. The compression boundary parsing module extracts the compression start address, logical offset, and compression ratio tag value of the fragment in the corresponding segment compression index based on the start and end positioning segments of the destruction chain. It uses the distance value between the logical offset and the fragment write address, combined with the offset threshold interval corresponding to the compression ratio tag, to determine the distance landing point and generate a compression segment offset compensation structure. The mapping path checking module calls the compressed segment offset compensation structure to extract the starting address number, the number of points, and the path span position value. It uses the consistency judgment method of the number of points being zero and the span number to identify the static path chain, selects and marks it as a mapped static path, and generates a path chain static mapping mark table. The disturbance destruction control module uses the intermediate pointer number of the static path chain in the path chain static mapping mark table, calls the random content record number and the archive data destruction scheduling period, writes the content to the intermediate node, shifts the address offset by one segment to the logical number, removes the record number of the address before the offset in the mapping path table, and generates a path disturbance fracture tracking log; The boundary fracture identification module extracts the fracture node number from the path disturbance fracture tracking log and compares it with the continuous index value in the start and end positioning section of the archive destruction chain. It determines the logical segment index not covered by the fracture number in the index, marks it as a chain coverage fault, and generates a destruction chain fracture coverage status mapping table. The start and end positioning sections of the destruction chain include the structural change boundary position, the field change behavior mapping index, and the version addition and deletion cross-record group. The compression section offset compensation structure includes the offset distance attribution identifier, the compression rate correspondence mapping, and the logical address offset threshold positioning item. The path chain static mapping mark table includes the path stability identification number, the unchanged pointing structure set, and the path span equivalence mark item. The path disturbance fracture tracking log includes the fracture path record number, the disturbance write content record number, and the logical offset migration record information. The destruction chain fracture coverage status mapping table includes the uncovered logical segment index group, the fracture node coverage mapping status, and the structural coverage integrity comparison result.

[0024] Specifically, if Figure 2 、 3 As shown, the fingerprint chain building blocks include: The field extraction submodule extracts the field level label, sequence index and field name byte length based on the field structure set in the cloud archive data, aligns the field level label and sequence index, filters and sorts the field name byte length, calls the field structure sequence and the level label set for sequence comparison, and generates the field sequence value. Based on the field structure set loaded in the cloud archive data, it is necessary to read the structure definition content one by one, extract the hierarchical label of each field node, and set it in the structured metadata. The hierarchical label of a field is identified as "2.1.4". This type of information is nested in XML, JSON or custom structure description files. The structure parsing script can be used to traverse the entire structure, extract the hierarchical label of each field as a hierarchical array, and then extract its sequential index value from the same structure. For example, if its physical position number in the document structure is "7", it is recorded as sorting index 7, and the byte length corresponding to the field name is obtained. In this step, the encoding analysis function is called to count the UTF-8 or GBK byte length of the field name, and the field name "customer code" is set to use The length after encoding in UTF-8 is 9 bytes, so this length is recorded. After the initial extraction is completed, the fields are calibrated according to their hierarchical labels and sequential indexes. The purpose of the calibration is to identify fields that are not arranged in order at the same level and implement position correction suggestions. The byte lengths of all field names are taken out and formed into an array, which is sorted. The lengths are compared and the order is exchanged through a double-layer traversal method. After obtaining the sorted results, they are compared one by one with the original field structure sequence to identify the degree of position matching between the fields. A reference field structure sequence is constructed based on the hierarchical label set and the original order. The byte length sorting result is then compared with the reference sequence, and the order-changed fields are marked one by one to generate a field arrangement sequence value.

[0025] The structure recognition submodule calls the field arrangement sequence value and the field level label set, combines the field unit identifier according to the field arrangement order and the level segmentation relationship, performs aggregation processing on continuous segments, divides the field segment boundary mark value and generates an aggregated segment sequence to generate the structure fingerprint density; According to the field arrangement sequence value and the field level label set, the structural combination operation is performed on each field node. Each field node, combined with its position value and level value, constitutes a unique field identifier, which is used to determine its belonging relationship in the overall structure and the continuity of adjacent fields. If a field arrangement sequence is set to "1011" and the level label is "1.1.2", the field is positioned in the second sub-item under the first level of the first layer through combination. All fields are sorted from small to large according to the arrangement sequence, and compared and classified according to the level label. Fields with the same level and continuous arrangement in the structure are identified and merged into an aggregation unit. During the processing, the current field area needs to be set. The starting and ending positions of the segment. Once a jump or inconsistent level change is found in the arrangement sequence during the traversal process, the current aggregation segment will be terminated immediately, and the segment boundary position will be recorded to start a new round of aggregation segment processing. When all fields are traversed, multiple field aggregation segments can be obtained. Each segment consists of its starting and ending positions. The aggregation segments are then numbered and recorded, organized into a structural division map, and an aggregation segment sequence is formed. It is further marked as a structural feature, which comes from the density ratio between the length of each aggregation segment and the total number of fields. The structural distribution analysis is performed through the number of aggregation segments and their distribution frequency to generate the structural fingerprint density.

[0026] The change identification submodule extracts the field quantity change value and field rearrangement displacement based on the structural fingerprint density and the version evolution data of the field structure set, determines whether field additions and field deletions coexist in the same change group, selects the version index values that meet the conditions, and generates the start and end segment values of the destruction chain; Perform differential extraction on the field structures of different versions, determine the tightness of each version structure by the density of the structural fingerprint, and then extract the number of fields in each version for comparison based on the version evolution data of the field structure set. Set version A to have 120 fields and version B to have 126 fields, and identify that the total number of fields has changed to 6. By extracting the permutation sequence list of the two version fields and comparing the position changes of the fields in the structure, identify the rearrangement of the fields, and record the displacement distance of the fields with the largest position changes. Then, based on the original sequence value and the new sequence value of each field, identify whether it is in the same hierarchical segment. If the displacement is accompanied by a hierarchical jump, it is recorded as a rearranged field. ,For the identification of newly added and deleted fields, it is completed by checking the existence of field identifiers. If a field identifier exists in version A but is missing in version B, it is a deleted field. Otherwise, it is a new field. The newly added and deleted fields are then combined and analyzed to identify whether they belong to the same change group. If both fields disappear and fields are added in the same paragraph, it is marked as a co-occurrence change, and the start and end positions of the segment are recorded. The destruction chain range is generated in combination with the field segment information. During the screening process, it is also necessary to confirm whether the change meets the density judgment condition. When the density is higher than a fixed value and the number of field changes exceeds the threshold, the index value of the version is included in the candidate set to generate the start and end segment values of the destruction chain.

[0027] Specifically, if Figure 2 、 4 As shown, the compression boundary parsing module includes: The fragment extraction submodule extracts the fragment compression start address and logical offset bits in the corresponding compression index based on the start and end segment values of the destruction chain, calls the absolute distance between the compression start address and the logical offset bits, evaluates the displacement mapping relationship within the compression segment, and generates the compression segment logical offset distance value; Perform positioning operations in the compressed index, and retrieve the compressed segment index items that match the start and end segment values of the destruction chain one by one. Set the start segment value to 3056 and the end segment value to 3082. Then, extract the compressed segment information corresponding to these two segment values in the compressed index structure. The compressed segment contains the compressed start address and the logical offset bit. For example, if the compressed start address is "0x00400000", the corresponding logical offset bit is "128KB". The absolute distance within the compressed segment is calculated by the numerical difference between the start address and the offset bit to form the compressed segment logical offset distance value. Set the start address of a compressed segment to "0x00400000" and the logical offset to "128KB". If the point is "0x00420000", it means that the logical data span of the compressed segment is "128KB". This type of distance is identified and summarized for multiple fragments to form a compressed segment logical offset distance sequence covering the destruction chain interval. During the distance value summary process, the displacement mapping relationship evaluation operation is performed to determine whether the offset distance shows linear growth or there are intra-segment jumps. By analyzing the regularity of the address and offset bits in each segment, areas with uneven compression rates can be found. The absolute distance values within these segments are constructed into a logical offset distance structure, which is used to describe the integrity and offset trend of the mapping relationship between each field in the compressed segment, and generate the compressed segment logical offset distance value.

[0028] The threshold mapping submodule calls the compression segment logical offset distance value and the compression ratio label value corresponding to the compression index segment to obtain the offset threshold interval corresponding to the compression ratio label, compares the landing position of the distance value within the offset threshold interval, determines the belonging segment, and generates the distance belonging flag value; Add a compression ratio label to each compressed segment, set the compression ratio of a segment to "3:1" or "5:1", which reflects the ratio of the content length before and after compression. Preset the logical offset threshold intervals corresponding to several compression ratio labels. Set the offset interval corresponding to the compression ratio "3:1" to "100KB~150KB", and the interval corresponding to the compression ratio "5:1" to "60KB~90KB". Then, perform interval identification for the offset distance value of each segment. If the offset distance value of a compressed segment is "120KB" and its matching compression ratio is "3:1", it falls between "100KB~150KB". , it is judged to be an internal value of the interval and is assigned to the corresponding segment of the compression ratio. During the comparison process, the offset values used are all expanded in the form of actual bytes or KB values to ensure consistency with the physical unit of the threshold interval. After interval judgment of all fragments, each fragment is assigned a distance attribution flag value, which indicates the compression level segment to which the compressed segment belongs in the offset space. For example, the flag value "1" represents a normal match, and "0" represents an unmatched or abnormal attribution. The mapping relationship between the offset distance value and the compression ratio label is evaluated, and a clear segment attribution basis is provided for subsequent sequence adjustment to generate a distance attribution flag value.

[0029] The sequence adjustment submodule extracts the assigned segment flag index and the logical offset bit according to the distance assigned flag value and the compressed segment logical offset distance value, calls the compressed start address to rebuild the offset position table within the segment, calculates the offset position compensation value, arranges the compensation order, and generates the compressed segment offset compensation structure; The formula for calculating the offset position compensation value is: ; in, Represents the offset position compensation value, Represents the total number of segments calculated, Representative The segment's belonging flag value, Representative The logical offset distance value of the segment, Representative The compensation coefficient of the segment, Representative The reconstruction coefficient of the segment, Representative The starting position value of the segment, Representative The end position value of the segment; The core objective of the formula is to calculate the offset compensation value generated during the reconstruction of multi-segment compressed structures. This is used to correct for spatial misalignment caused by differences between the compression and logical displacement of the assigned segments. The formula starts with the absolute difference between the assigned marker value and the logical offset distance of each processed segment, reflecting the positional deviation between the segment's original assignment and the actual logical adjustment. Half the compensation coefficient length is superimposed to fine-tune the compensation scale. This difference is multiplied by the square of the reconstruction coefficient, nonlinearly amplifying the influence weight of each segment in the structural reconstruction to highlight the dominant role of key segments in the offset adjustment. The distance range of the segment's start and end positions is then introduced. By taking the square root, the sensitivity of the spatial span to the compensation amplitude is balanced, avoiding disproportionate offset fluctuations caused by long segments. A comprehensive offset correction output is generated by summing the processed values of all segments. The overall calculation process integrates four variables: structural position difference, intra-segment adjustment scale, reconstruction influence coefficient, and physical spatial range. This allows for precise measurement and comprehensive adjustment of intra-segment misalignment behavior, enabling global compensation control for positional alignment of compressed data in the decompression mapping. Parameter description and acquisition method: : The total number of segments involved in the calculation, indicating the number of segments that need to be processed during the reconstruction process. This value is obtained through analysis or data structure; : No. The segment's ownership flag value indicates the segment's ownership information before compression. This value is obtained through management and is an integer value, indicating the segment's ownership type or category. : No. The logical offset distance value of the segment indicates the offset of the segment in the logical address space. This value is obtained through the memory management unit or the segment table. : No. The compensation coefficient of a segment represents the adjustment coefficient for the offset of the segment during the reconstruction process. This value is determined by design or algorithm optimization and is a floating value that reflects the impact of the segment characteristics on the offset. : No. The reconstruction coefficient of a segment indicates the reconstruction ratio of the offset of the segment during the reconstruction process. This value is determined by design or algorithm optimization and is a floating value. The characteristics of the segment affect the reconstruction ratio. : No. The starting position value of the segment indicates the starting position of the segment in the physical address space. This value is obtained through the segment table or the memory management unit; : No. The end position value of the segment indicates the end position of the segment in the physical address space. This value is obtained through the segment table or the memory management unit. Example description: There are three sections in the settings ( ), and its corresponding parameter values are as follows: Paragraph 1: , , , , , ; Paragraph 2: , , , , , ; Paragraph 3: , , , , , ; Substitute the above parameters into the formula for calculation: ; For paragraph 1: ; For paragraph 2: ; For paragraph 3: ; Adding the above results: ; The results show that when reconstructing the intra-segment offset position table, an offset compensation of 12186.125 bytes is required. The offset position compensation value is used to characterize the overall adjustment degree of the position offset in the reconstruction of the compressed segment. It is a quantitative indicator formed by comprehensive calculation of multiple segment parameters. This value reflects the intensity and range of the position correction required for each segment due to logical adjustment. It is used to support the repositioning accuracy of the compressed structure to ensure the correctness of the intra-segment offset position.

[0030] Specifically, if Figure 2 、 5 As shown, the mapping path troubleshooting module includes: The path extraction submodule extracts the starting address number, the number of pointing times, and the path span position value based on the compressed segment offset compensation structure, calculates the offset value between the starting address number and the path span position value, detects whether the number of pointing times is zero, combines the offset value and the number of pointing times, and generates the path positioning offset; Parse the starting address number of each data item, set the corresponding address number of a compression field to "ADDR001", read the corresponding pointing count value of the field during the compression process, set the pointing count value to "3" if the field is used as the mapping target of the remaining fields multiple times, and set the value to "0" if it is not referenced by the remaining fields, extract the path span position value of the field, set its relative position number in the compression logic path to "SP007", calculate the numerical offset difference between the starting address number and the path span position value, if ADDR00 1 is "1000", SP007 is "1080", then the offset value is "80", which reflects the extension of the field in the path. It detects whether the number of points of the field is zero. If it is zero, it is marked as the end node of the path, which is suitable for subsequent path identification and tail proofreading. The above offset value is then combined with the number of points to construct the path positioning offset information. This information is used to reflect the starting position, span and call frequency of each path clue. It is an important data source for path tracking and static path marking, and outputs the path positioning offset.

[0031] The path identification submodule calls the path positioning offset, performs number verification on the data items with the pointing count value of zero, compares the path span position value with the difference between the starting address number, determines whether the difference is consistent with the span number, and generates a consistency identification sequence value; Identify the fields in the path that are not further referenced, indicating the terminal nodes of the path. During proofreading, find the difference between the path span position value and the original field number. If the starting address number of a data item is "1200", the path span position value is "1280", and the difference is "80", then compare whether the span number identified by the path item is also "80". If they are consistent, it can be confirmed that the path span is consistent with the calculated difference, indicating that there is no structural error in the path. If they are inconsistent, it indicates that the path has a pointing anomaly or a cross-segment error. Execute the above judgment logic on the data items and generate consistency identification results one by one. Output "1" for each data item to indicate consistency and "0" for inconsistency. This can be used to assist in identifying structural jumps and inter-segment association breaks in the compressed path. In the process of generating sequence values, the position number of the error field can also be recorded, and the segment it is in and the nature of the error can be marked for subsequent marking and correction processing to ensure the overall coherence of the path chain and generate a consistency identification sequence value.

[0032] The static tag submodule identifies the sequence value based on consistency, numbers and classifies the path span position value set, assigns static path tags to path items with the same number, assigns a tag index value to each group of static path tag items, and establishes a path chain static mapping tag table; Perform a numbering and classification operation on the path span position value set. Scan the entire set and group items with the same path number. Set items with path number "SP010" to form path group A, and all items with path number "SP011" to form path group B. Generate a static path tag for each group of path items. This tag is based on the path number and adds a unified prefix to generate the static path tag value. Set "STATICPATHA" to represent path group A. Assign a tag index value to each group of static path tag items. Set the corresponding index value of Group A to "0x01" and that of Group B to "0x02". The tag index values of the path groups must be unique and arranged in order. List each static path tag with its corresponding path span set, starting address number, and classification index value. This allows for quick query of the group to which the path belongs and its structural position relationship. The static path index configuration as a compressed structure is subsequently referenced for path-level comparison and identification during path mapping, retrieval optimization, and structural analysis. Establish a path chain static mapping tag table.

[0033] Specifically, if Figure 2 、 6 As shown, the disturbance destruction control module includes: The node selection submodule locates the logical number of the middle position of each path chain based on the middle pointer number of the static path chain in the path chain static mapping mark table, obtains the corresponding address number in the path chain static mapping mark, selects the path set whose middle pointer number is within the scheduling period, calculates the score value of the path set, and generates the middle node address set; The formula for calculating the score value of the path set is: ; in, Represents the score value of the path set, Represents the path chain The logical number of the node, Representative The target position number of each node in the scheduling period, Representative The weight value of the path chain, Representative The start time of the path chain, Representative The end time of the path chain, is the number of paths; The formula aims to quantify the overall degree of deviation of a multi-path chain set in static mapping and scheduling matching. The absolute value of the deviation between the logical number and the target position number of each path node is used to measure the degree of deviation from the target. The score is then constructed by combining the weight of the path chain and the influence of the time span. The logical number reflects the static position of the path node in the original path structure, while the target number reflects its expected return to the position in the current scheduling cycle. The larger the difference between the two, the more significant the scheduling deviation. The offset value of each node is multiplied by the path chain weight to strengthen the influence of the scheduling critical path. At the same time, this value is divided by the square root of the start and end times of the path chain to form a time span adjustment factor, which weakens the amplification effect of long-term paths on the results and balances the time influence between nodes. By coupling the multi-factor of deviation, path weight, and time adjustment, this formula constructs a scoring system that comprehensively reflects the degree of path chain scheduling deviation. It is suitable for control strategies such as scheduling accuracy assessment, path chain screening, and key node identification, providing a well-structured and computationally controllable multi-segment path evaluation method. Parameter meaning and formula calculation derivation process: Represents the path chain The logical number of each node is determined by the position of the node in the path chain through the path chain static mapping tag table. In specific applications, the value comes from the logical position number of the path chain and is extracted from the static mapping table according to the path setting; Representative The target position number of each node in the scheduling cycle. This value is the target number assigned to each node, which is dynamically calculated or scheduled according to the current scheduling cycle. In specific applications, the target task position from the scheduling cycle can be obtained through the task scheduling table, based on the arrangement of the node position in the current scheduling cycle; Representative The weight of each path chain is calculated based on a comprehensive evaluation of parameters such as the path chain's priority, bandwidth, and path stability. The weight is calculated using a comprehensive scoring method based on path reliability, path load, and data transmission efficiency. The weight changes in real time as path chain conditions (such as bandwidth and latency) change. Representative The start time of the path chain, It is calculated based on the timestamp in the actual scheduled task and is determined by the time point when the path chain starts executing. The specific value of is obtained through the path chain start time recorded in the scheduling table; Representative The end time of the path chain, It is calculated based on the task completion time or the end time of the path chain, and the value comes from the end time of the task scheduling; Calculation example: There are three path chains. The nodes on each path chain have different numbers and scheduling information. The following are their corresponding parameter values: Path chain 1: P1=5, D1=10, A1=3.2, B1=8, C1=15; Path chain 2: P2=7, D2=12, A2=2.8, B2=10, C2=18; Path chain 3: P3=4, D3=9, A3=4.1, B3=7, C3=14; Substitute the parameters into the formula for calculation: ; Calculate each item: Item 1: ; Item 2: ; Item 3: ; Add the results to calculate: ; The result indicates that within the scheduling cycle, the path set has a comprehensive score of 10.46. The path set score is used to assess the overall match between the logical positions of nodes in the path chain and the target positions, reflecting the strength of path scheduling deviations. This value integrates node deviations, path importance, and time span factors to provide a quantitative basis for path scheduling evaluation, reflecting the comprehensive results of the time and scheduling conditions of each path in the current path chain selection.

[0034] The random write submodule calls the intermediate node address set, generates a random write value sequence for each intermediate node based on the random content record number, detects the logical number match between each write value and the node address offset segment, determines that the write position is within the range of the backward address offset, obtains the new position number after writing the random value content to the node, and generates the perturbation write position value; Generate a corresponding random write value sequence for each intermediate node. The write value is generated based on the random content record number set internally. Set the record number "RNUM23" to generate the write sequence "VAL83, VAL42, VAL76". Match a random value to each intermediate node and check whether the write logical number of the random value matches the logical number of the current node address offset segment. Set the current address number of a node to "ADDR1204" and the offset segment logical number range to "LOGIC1204–LOGIC1210". If the write logical number is "LOGIC1207", it is determined to be an offset match. Confirm that the write position is three steps backward from the current address. After the match, write the random value to the specified node logical position and record the new logical number. If the node number changes to "LOGIC1207" after writing, record the new number as the perturbation write position value. After all address set operations are performed, identify the dynamically changed position of the write operation in the path structure, provide basic address change information for the path removal operation, and generate a perturbation write position value set.

[0035] The path removal submodule extracts the corresponding logical number original address based on the disturbance write position value, locates the record number corresponding to the address, removes the record number and reconstructs the mapping path table index sequence, analyzes the path break trend and the logical number mapping relationship under the scheduling cycle, and establishes a path disturbance break tracking log; Extract the original address of the logical number corresponding to each item, and use the logical number to reversely query its original address number record in the path mapping structure. Set the disturbance write location "LOGIC1207" to correspond to the original address "ADDR1204". Locate the record number of the address in the path chain structure, such as "REC91", and perform a removal operation on the record number. Mark its entry in the mapping path table as deleted or transferred. After the removal is completed, reconstruct the mapping path table index sequence to ensure the continuity of the overall path structure and the correctness of the address reference. During the reconstruction process, the path interruption situation is simultaneously analyzed. If multiple path disturbance writes are concentrated on a certain path chain, the path integrity of the chain will be reduced. Based on this, the path break trend indicator is calculated. Combined with the current scheduling cycle number, the displacement trajectory and gap area of the logical number in the path structure in this cycle are tracked. The node displacement caused by each disturbance, the record number removal result and its relative impact in the path chain are listed in detail. As an important basis for evaluating the stability of the path structure, a path disturbance break tracking log is established.

[0036] Specifically, if Figure 2 、 7 As shown, the boundary fracture identification module includes: The node extraction submodule extracts the broken node number set based on the path disturbance fracture tracking log, extracts the continuous index values in the start and end positioning segments of the archive destruction chain, arranges the broken node numbers and continuous index value lists in order and classifies the mapping relationship to generate a broken node coverage index group; Extract the node numbers of the broken nodes. The numbers are composed of the logical number positions recorded in the log to form a broken node number set. Set the broken nodes marked in the log to be "LOGIC1302", "LOGIC1305", "LOGIC1311", etc. Read the start and end positioning segment range of the archival material destruction chain. For example, the destruction chain starts with "IDX1300" and ends with "IDX1320". Extract the continuous index values within this range to form a list, that is, [IDX1300, IDX1301, IDX1320]. Compare the broken node numbers with the continuous index value list in sequence, and establish a classification mapping relationship between the two. Set "LOGIC1302" to correspond to "IDX1302". Combine the successfully matched entries into a broken node coverage index group to form a structured set. This is used to describe the specific index position distribution of the break event within the destruction chain segment range, which can be used as the basis for subsequent identification of uncovered positions in the chain structure. At the same time, this index group also provides a logical number backtracking channel in the subsequent processing stage to generate a broken node coverage index group.

[0037] The position comparison submodule calls the broken node coverage index group, performs node number matching on the continuous index value list, identifies the index positions that are not matched by the broken node number, extracts the logical number as the residual index item, and obtains the segment uncovered number set; Perform matching analysis on the associated continuous index value list, and compare each item in the index list one by one to see if it has been covered by the broken node number. If "IDX1300" or "IDX1301" does not appear in the broken node number set, mark it as an unmatched item. Extract all index positions corresponding to the unbroken node identifiers and record them as residual index items. Organize their logical numbering form and set "LOGIC1300", "LOGIC1301", "LOGIC1306" and so on as uncovered items to identify areas in the segment structure that are not affected by the path disturbance. Through the numbering items, further operations such as fault analysis and structural residue identification can be performed to obtain the segment uncovered number set.

[0038] The fault marking submodule determines the belonging segments in the destroyed chain location area according to the chain segment uncovered number set and the continuous logical number sequence, filters the logical segments including uncovered numbers in the belonging segments, marks them as chain coverage faults, and generates a chain segment fault number sequence; Execute the number attribution judgment operation, serialize and arrange the numbers in the logical order, identify the continuous relationship and segmentation trend between the numbers, and correspond them to the positioning area of the destruction chain. If the number "LOGIC1300-LOGIC1303" belongs to the destruction chain segment "SEGA", then SEGA contains uncovered logical number items and needs to be marked as a fault segment. In each segment, check whether it contains the chain segment uncovered number one by one. If it does, mark the segment as "chain covered fault". The marking content is constructed by combining the logical number segment and the fault flag value, such as [SEGA:FAULT, SEGC:FAULT]. It is one of the core components in the fault identification structure, providing accurate breakpoint identification for path status identification and generating a chain segment fault number sequence.

[0039] The state identification submodule calls the chain segment fault number sequence, evaluates the mapping index between the chain segment number and the fault state flag value, adds the corresponding fault state attribute to each chain segment logical number, arranges them in a logical order, and generates a destruction chain fracture coverage state mapping table; Based on the segment fault number sequence, a state injection operation is performed on the segment logical number. The logical number list of each segment is read one by one and matched one by one with the fault status mark value. If a segment "SEGA" has been marked as a fault state in the previous sequence, all its subordinate logical numbers will be additionally marked with the attribute "FAULT". If a segment is not marked, it will maintain the "NORMAL" state. All logical numbers are arranged in numerical order and the state value is added to form an ordered structure list. It contains the identification result of each logical number and whether it is in a fault in the path chain. It can be used as a reference for path integrity in subsequent path recovery, node reconstruction or structure repair. The output is a destruction chain fracture coverage state mapping table.

[0040] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A cloud-based archival data security destruction system, characterized in that: The system comprises: The fingerprint chain construction module processes the field sequence based on the field structure set in the cloud archive data to generate a structural fingerprint. It then compares the field quantity change value, field rearrangement displacement, and field addition and deletion cross-mark value of the structural fingerprint to generate the start and end positioning segments of the destruction chain. The compression boundary parsing module extracts the compression start address, logical offset bit and compression ratio tag value of the fragment in the corresponding segment compression index based on the start and end positioning segments of the destruction chain, performs distance landing point attribution judgment, and generates a compression segment offset compensation structure; The mapping path checking module calls the compressed section offset compensation structure, extracts the starting address number, the number of times pointed to, and the path span position value, filters and marks them as mapping static paths, and generates a path chain static mapping mark table; The disturbance destruction control module calls the random content record number and the archive data destruction scheduling period based on the static mapping mark table of the path chain, writes the content to the intermediate node and shifts the address offset to the logical number, removes the record number of the address before the offset in the mapping path table, and generates a path disturbance break tracking log.

2. The cloud-based archival data security destruction system according to claim 1 is characterized in that: The destruction chain start and end positioning segments include the structural change boundary position, the field change behavior mapping index, and the version addition and deletion cross-record group; the compression segment offset compensation structure includes the offset distance attribution identifier, the compression rate correspondence mapping, and the logical address offset threshold positioning item; the path chain static mapping mark table includes the path stability identification number, the unchanged pointing structure set, and the path span equivalence mark item; the path disturbance fracture tracking log includes the fracture path record number, the disturbance write content record number, and the logical offset migration record information.

3. The cloud-based archival data security destruction system according to claim 1, characterized in that: The fingerprint chain building module includes: The field extraction submodule extracts the field level label, sequence index and field name byte length based on the field structure set in the cloud archive data, aligns the field level label and sequence index, filters and sorts the field name byte length, calls the field structure sequence and the level label set for sequence comparison, and generates the field sequence value. The structure identification submodule calls the field arrangement sequence value and the field level label set, combines the field unit identifier according to the field arrangement order and the level segmentation relationship, performs aggregation processing on continuous segments, divides the field segment boundary mark value and generates an aggregated segment sequence, and generates a structure fingerprint density; The change identification submodule extracts the field quantity change value and the field rearrangement displacement based on the structure fingerprint density and the version evolution data of the field structure set, determines whether the addition of fields and the deletion of fields exist simultaneously in the same change group, filters the version index values that meet the conditions, and generates the start and end segment values of the destruction chain.

4. The cloud-based archival data security destruction system according to claim 3 is characterized in that: The compression boundary parsing module includes: The fragment extraction submodule extracts the fragment compression start address and logical offset bit in the corresponding compression index based on the start and end segment values of the destruction chain, calls the absolute distance between the compression start address and the logical offset bit, evaluates the displacement mapping relationship within the compression segment, and generates the compression segment logical offset distance value; The threshold mapping submodule calls the compression segment logical offset distance value and the compression ratio label value corresponding to the compression index segment, obtains the offset threshold interval corresponding to the compression ratio label, compares the landing position of the distance value within the offset threshold interval, determines the belonging segment, and generates a distance belonging flag value; The sequence adjustment submodule extracts the assigned segment flag index and the logical offset bit according to the distance assigned flag value and the compressed segment logical offset distance value, calls the compressed start address to rebuild the offset position table within the segment, calculates the offset position compensation value, and arranges the compensation order to generate a compressed segment offset compensation structure.

5. The cloud-based archival data security destruction system according to claim 4 is characterized in that: The formula for calculating the offset position compensation value is: ； in, Represents the offset position compensation value, Represents the total number of segments calculated, Representative The segment's belonging flag value, Representative The logical offset distance value of the segment, Representative The compensation coefficient of the segment, Representative The reconstruction coefficient of the segment, Representative The starting position value of the segment, Representative The end position value of the segment.

6. The cloud-based archival data security destruction system according to claim 4, characterized in that: The mapping path checking module includes: The path extraction submodule extracts the starting address number, the pointing count value, and the path span position value based on the compressed segment offset compensation structure, calculates the offset value between the starting address number and the path span position value, detects whether the pointing count value is zero, combines the offset value and the pointing count value, and generates a path positioning offset; The path identification submodule calls the path positioning offset, performs number verification on the data item with the pointing count value of zero, compares the path span position value with the difference between the starting address number, determines whether the difference is consistent with the span number, and generates a consistency identification sequence value; The static marking submodule numbers and classifies the path span position value set according to the consistency identification sequence value, assigns static path marks to path items with the same number, allocates a mark index value to each group of static path mark items, and establishes a path chain static mapping mark table.

7. The cloud-based archival data security destruction system according to claim 6, characterized in that: The disturbance destruction control module includes: The node selection submodule locates the logical number of the middle position of each path chain based on the middle pointer number of the static path chain in the path chain static mapping mark table, obtains the corresponding address number in the path chain static mapping mark, selects the path set whose middle pointer number is within the scheduling period, calculates the score value of the path set, and generates the middle node address set; The random write submodule calls the intermediate node address set, generates a random write value sequence for each intermediate node according to the random content record number, detects the match between each write value and the logical number of the node address offset segment, determines that the write position is within a range of the backward address offset, obtains the new position number after writing the random value content to the node, and generates a disturbed write position value; The path removal submodule extracts the corresponding logical number original address according to the disturbance write position value, locates the record number corresponding to the address, removes the record number and reconstructs the mapping path table index sequence, analyzes the path break trend and the logical number mapping relationship under the scheduling cycle, and establishes a path disturbance break tracking log.

8. The cloud-based archival data security destruction system according to claim 7, characterized in that: The formula for calculating the score value of the path set is: ； in, Represents the score value of the path set, Represents the path chain The logical number of the node, Representative The target position number of each node in the scheduling period, Representative The weight value of the path chain, Representative The start time of the path chain, Representative The end time of the path chain, is the number of paths.

9. The cloud-based archival data security destruction system according to claim 1, characterized in that: The system also includes a boundary fracture identification module: The boundary fracture identification module extracts the fracture node number from the path disturbance fracture tracking log and compares it with the continuous index value in the start and end positioning section of the archive destruction chain, determines the logical segment index not covered by the fracture number in the index, marks it as a chain coverage fault, and generates a destruction chain fracture coverage status mapping table; The destruction chain break coverage status mapping table includes uncovered logical paragraph index groups, broken node coverage mapping conditions, and structure coverage integrity comparison results.

10. The cloud-based archival data security destruction system according to claim 9, characterized in that: The boundary fracture identification module includes: The node extraction submodule extracts the set of broken node numbers based on the path disturbance fracture tracking log, extracts the continuous index values in the start and end positioning segments of the archive data destruction chain, arranges the broken node numbers and the continuous index value lists in order and classifies the mapping relationship, and generates a broken node coverage index group; The position comparison submodule calls the broken node coverage index group, performs a node number matching operation on the continuous index value list, identifies the index position not matched by the broken node number, extracts the logical number as the residual index item, and obtains the segment uncovered number set; The fault marking submodule determines the belonging segments in the destroyed chain location area according to the chain segment uncovered number set and the continuous logical number sequence, filters the logical segments including the uncovered numbers in the belonging segments, marks them as chain coverage faults, and generates a chain segment fault number sequence; The state identification submodule calls the segment fault number sequence, evaluates the mapping index between the segment number and the fault state flag value, adds the corresponding fault state attribute to each segment logical number, arranges them in a logical order, and generates a destruction chain fracture coverage state mapping table.

Images