Intelligent Parallel Optimization Method and System for Data Storage

Abstract

This invention relates to the field of data storage technology, and particularly to an intelligent parallel optimization method and system for data storage. The method includes: encrypting the data to be stored from each data source, storing the encryption key locally, and uploading the encrypted ciphertext data; generating seed parameters based on the ciphertext data from each data source and uploading them to construct a storage simulation model and generate emission parameters; performing mirror reflection propagation simulation based on all emission parameters, and interleaving the ciphertext bytes into physical storage units according to a preset grid occupancy rule to obtain an interleaved ciphertext storage volume. This invention forms a deeply obfuscated interleaved arrangement, ensuring that no single party can restore the writing order if the emission parameters from other data sources are missing. This guarantees the secure isolation of multi-party data joint storage. After storage is completed, relevant parameter data is erased, ensuring that third parties no longer possess any data to recover the propagation trajectory after writing.

Description

Technical Field

[0001] This invention belongs to the field of data storage technology, and in particular relates to a method and system for intelligent parallel optimization of data storage. Background Technology

[0002] In multi-source data co-storage scenarios, data from different data sources needs to be interleaved and stored by a third-party storage platform.

[0003] Existing technologies typically require data sources to submit the data to be stored and the write strategy to the platform in plaintext. This allows third parties to record and analyze the physical arrangement of the data during the write process, and even retain key parameters for later individual data recovery. Even with a strategy of deleting parameters afterward, it is difficult to mathematically prove that the third party did not retain a copy, posing a long-term trust dependency risk. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent parallel optimization method for data storage, which aims to solve the problem that existing federated storage solutions need to expose data or write strategies to third-party storage platforms, giving them the opportunity to retain key parameters for subsequent data recovery, which poses a long-term trust dependency risk that is difficult to eliminate.

[0005] This invention is implemented as follows: a smart parallel optimization method for data storage, the method comprising:

[0006] Each data source encrypts its own data to be stored, saves the encryption key locally, and then uploads the encrypted ciphertext data.

[0007] Seed parameters are generated based on encrypted data from various data sources and uploaded. A storage simulation model is constructed through a third-party storage platform, and emission parameters corresponding to each data source are generated based on the seed parameters. The storage simulation model is a convex polygon containing multiple grids, each grid corresponding to a physical storage unit.

[0008] Based on all emission parameters, the specular reflection propagation of multiple light rays is simulated in the storage simulation model. According to the preset grid occupancy rules, the ciphertext bytes of each data source are interleaved and written into the physical storage units mapped by each grid in the convex polygon to obtain the interleaved ciphertext storage body.

[0009] After the writing is complete, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage.

[0010] Preferably, the steps of generating seed parameters based on encrypted data from various data sources, uploading them, constructing a storage simulation model through a third-party storage platform, and generating transmission parameters corresponding to each data source based on the seed parameters include:

[0011] Each data source calculates and generates seed parameters based on its own encrypted data, and uploads the seed parameters to a third-party storage platform;

[0012] A storage simulation model is built using a third-party storage platform. The interior of the convex polygon is divided into multiple grids, and a unique mapping relationship is established between each grid and the physical storage unit. The total number of grids is equal to the total number of bytes of encrypted data from all data sources.

[0013] Based on the various sub-parameters received, a third-party storage platform generates corresponding emission parameters for each data source. The emission parameters include at least the initial position, initial direction, and propagation speed of the light source within the convex polygon.

[0014] Preferably, the step of simulating the specular reflection propagation of multiple light rays within a storage simulation model based on all emission parameters, and interleaving the ciphertext bytes of each data source into the physical storage units mapped by each grid within the convex polygon according to a preset grid occupancy rule to obtain the interleaved ciphertext storage volume includes:

[0015] Based on all transmission parameters, the specular reflection propagation of multiple light rays is synchronously simulated within a convex polygon, with each light ray carrying a ciphertext byte stream from its corresponding data source.

[0016] Monitor the propagation path of each ray. When any ray first enters a grid that has not yet been written to, write the ciphertext bytes corresponding to the ray to the physical storage unit mapped to that grid in sequence. If the grid has already been written to, skip that grid and continue propagating.

[0017] After the ciphertext data of any data source has been written, the ray corresponding to that data source is dynamically switched to an auxiliary ray to assist the remaining unfinished data sources in continuing to search and write to the empty grid until all ciphertext data has been written, thus obtaining an interwoven ciphertext storage body.

[0018] Preferably, after the writing is completed, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage body, including the following steps:

[0019] After confirming that all encrypted data has been written to the corresponding physical storage unit, the parameter erasure process is triggered.

[0020] Clear all launch parameters and related intermediate state data used during the storage simulation;

[0021] Traverse the parameter storage area, overwrite the memory space occupied by the emission parameters, and terminate the access process to the storage simulation model.

[0022] Preferably, the seed parameter is a 10-digit decimal string. Generating emission parameters for each data source based on the seed parameter specifically includes: taking the first four characters of the seed parameter, where the first two are used as the horizontal coordinate parameter and the last two as the vertical coordinate parameter; constructing a maximum inscribed square inside the convex polygon; normalizing the horizontal and vertical coordinate parameters and mapping them to the maximum inscribed square to obtain the initial position coordinates of the light source corresponding to the data source; taking the middle three characters of the seed parameter; calculating the ratio of the values ​​of the middle three characters corresponding to all data sources; dividing the 360-degree circle into angle intervals corresponding to each data source based on this ratio to determine the initial propagation direction angle of each data source; taking the last three characters of the seed parameter; calculating the ratio of the values ​​of the last three characters corresponding to all data sources; dividing the preset propagation speed interval into propagation speed values ​​corresponding to each data source based on this ratio.

[0023] Another object of the present invention is to provide an intelligent parallel optimization system for data storage, the system comprising:

[0024] The data encryption module is used to encrypt the data to be stored from each data source, save the encryption key locally, and then upload the encrypted ciphertext data.

[0025] The reflection parameter generation module is used to generate seed parameters based on encrypted data from various data sources and upload them. It constructs a storage simulation model through a third-party storage platform and generates emission parameters corresponding to each data source based on the seed parameters. The storage simulation model is a convex polygon containing multiple grids, each grid corresponding to a physical storage unit.

[0026] The storage simulation module is used to simulate the specular reflection propagation of multiple light rays within the storage simulation model based on all emission parameters. According to the preset grid occupancy rules, the ciphertext bytes of each data source are interleaved and written into the physical storage units mapped by each grid within the convex polygon to obtain the interleaved ciphertext storage body.

[0027] The ciphertext storage module is used so that after writing is completed, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage body.

[0028] Preferably, the reflection parameter generation module includes:

[0029] The seed parameter generation unit is used by each data source to calculate and generate seed parameters based on its own encrypted data, and upload the seed parameters to a third-party storage platform.

[0030] The storage mapping unit is used to build a storage simulation model through a third-party storage platform, divide the interior of the convex polygon into multiple grids, and establish a unique mapping relationship between each grid and the physical storage unit. The total number of grids is equal to the total number of bytes of encrypted data from all data sources.

[0031] The emission parameter calculation unit is used to generate corresponding emission parameters for each data source based on various received sub-parameters through a third-party storage platform. The emission parameters include at least the initial position, initial direction, and propagation speed of the light source within the convex polygon.

[0032] Preferably, the storage simulation module includes:

[0033] The simulated propagation unit is used to synchronously simulate the specular reflection propagation of multiple light rays within a convex polygon based on all transmission parameters, with each light ray carrying a ciphertext byte stream from its corresponding data source.

[0034] The storage writing unit is used to monitor the propagation path of each ray. When any ray first enters a grid that has not yet been written to, the ciphertext bytes corresponding to the ray are written to the physical storage unit mapped to that grid in sequence. If the grid has already been written to, the ray is skipped and the propagation continues.

[0035] The auxiliary storage unit is used to dynamically switch the light corresponding to any data source to an auxiliary light source after the ciphertext data of any data source has been written, so as to assist the remaining unfinished data sources to continue searching and writing to the empty grid until all ciphertext data has been written, thus obtaining the interleaved ciphertext storage body.

[0036] Preferably, the encrypted storage module includes:

[0037] The erase start unit is used to trigger the parameter erase process after confirming that all encrypted data has been written to the corresponding physical storage unit;

[0038] The data erasure unit is used to erase all transmission parameters and related intermediate state data used in the storage simulation process.

[0039] The program termination unit is used to traverse the parameter storage area, perform an overwrite operation on the memory space occupied by the emission parameters, and terminate the access process to the storage simulation model.

[0040] Preferably, the seed parameter is a 10-digit decimal string. Generating emission parameters for each data source based on the seed parameter specifically includes: taking the first four characters of the seed parameter, where the first two are used as the horizontal coordinate parameter and the last two as the vertical coordinate parameter; constructing a maximum inscribed square inside the convex polygon; normalizing the horizontal and vertical coordinate parameters and mapping them to the maximum inscribed square to obtain the initial position coordinates of the light source corresponding to the data source; taking the middle three characters of the seed parameter; calculating the ratio of the values ​​of the middle three characters corresponding to all data sources; dividing the 360-degree circle into angle intervals corresponding to each data source based on this ratio to determine the initial propagation direction angle of each data source; taking the last three characters of the seed parameter; calculating the ratio of the values ​​of the last three characters corresponding to all data sources; dividing the preset propagation speed interval into propagation speed values ​​corresponding to each data source based on this ratio.

[0041] The intelligent parallel optimization method for data storage provided by this invention dynamically determines the storage sequence of data from different data sources through physical simulation, forming a deeply confused intertwined arrangement. No single party can restore the writing order if the emission parameters of other data sources are missing, thus ensuring the security and isolation of multi-party data joint storage. After storage is completed, the relevant parameter data is erased, so that the third party no longer holds any data to recover the propagation trajectory after writing. Attached Figure Description

[0042] Figure 1 A flowchart of the intelligent parallel optimization method for data storage provided in an embodiment of the present invention;

[0043] Figure 2 This is an architecture diagram of the intelligent parallel optimization system for data storage provided in an embodiment of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0045] like Figure 1 The diagram shows a flowchart of a data storage intelligent parallel optimization method provided in an embodiment of the present invention. The method includes:

[0046] S100 encrypts the data to be stored through each data source, saves the encryption key locally, and then uploads the encrypted ciphertext data.

[0047] In this step, before the joint storage begins, each data source independently encrypts its own raw data to be stored, generating ciphertext data. The encryption key is generated locally by the data source and kept by the data source itself, without being sent to external parties. Each data source uploads the encrypted ciphertext data to the third-party storage platform. Since the third-party storage platform does not have the key, it cannot decrypt the ciphertext data.

[0048] S200: Seed parameters are generated based on encrypted data from various data sources and uploaded. A storage simulation model is constructed through a third-party storage platform, and emission parameters corresponding to each data source are generated based on the seed parameters. The storage simulation model is a convex polygon containing multiple grids, each grid corresponding to a physical storage unit.

[0049] In this step, each data source generates seed parameters based on the encrypted data and uploads them to a third-party storage platform. The seed parameters are a sequence of numbers generated by each data source based on its own encrypted data, serving as a representation of the emission parameters. After the data source uploads the seed parameters, the third-party storage platform receives various sub-parameters and extracts different fields from the seed parameters according to preset conversion rules. These fields are then mapped to the initial coordinates, initial propagation angle, and propagation speed of the light rays from the corresponding data source within the convex polygon, thereby generating the emission parameters. The third-party storage platform determines the number of sides of the convex polygon based on the number of data sources participating in the joint storage and constructs a storage simulation model. The interior of this convex polygon is divided into multiple grids, with the total number of grids equal to the total number of bytes of encrypted data from all data sources. Each grid uniquely corresponds to a physical storage unit.

[0050] S300, based on all emission parameters, simulates the specular reflection propagation of multiple light rays within a storage simulation model. According to the preset grid occupancy rules, it interleaves the ciphertext bytes of each data source into the physical storage units mapped by each grid within the convex polygon, thus obtaining the interleaved ciphertext storage body.

[0051] In this step, the third-party storage platform synchronously simulates the specular reflection propagation process of multiple light rays within the storage simulation model based on all emission parameters. Each light ray carries the ciphertext byte stream of its corresponding data source. The light rays propagate along their initial directions inside the convex polygon, and specular reflection occurs when they encounter the boundary. According to the preset grid occupancy rules, when any light ray first propagates to a grid that has not yet been written to, the ciphertext byte corresponding to that light ray is written sequentially to the physical storage unit mapped to that grid. If the grid has already been written to by other light rays, the light ray is skipped and propagation continues until the next grid. When all the ciphertext data of a certain data source has been written, the light ray corresponding to that data source is switched to an auxiliary light ray to assist the remaining unfinished data sources in continuing to search for and write to empty grids. When all the ciphertext data has been written, an interleaved ciphertext storage body is obtained.

[0052] S400, after the writing is completed, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage.

[0053] In this step, after all the encrypted data has been written to the corresponding physical storage unit, the third-party storage platform triggers the parameter erasure process. First, it clears all the transmission parameters and related intermediate state data used in the storage simulation process. Second, it traverses the parameter storage area, performs an overwrite operation on the memory space occupied by the transmission parameters, terminates the access process to the storage simulation model, and releases the related computing resources.

[0054] In a preferred embodiment of the present invention, the steps of generating seed parameters based on encrypted data from various data sources and uploading them, constructing a storage simulation model through a third-party storage platform, and generating transmission parameters corresponding to each data source based on the seed parameters include:

[0055] S201, each data source calculates and generates seed parameters based on its own encrypted data, and uploads the seed parameters to a third-party storage platform.

[0056] In this step, each data source does not need to directly specify the initial position, propagation direction, or other specific emission parameters of the light within the convex polygon. Instead, it generates a compact sequence of numbers, i.e., seed parameters, based on its own encrypted data. For example, data source A performs a hash operation on its encrypted data and takes the first 10 decimal digits of the hash value as the seed parameter. Taking data source A as an example, assuming that the seed parameter obtained after processing its encrypted data is 3847291056, the seed parameter itself does not directly expose the initial coordinates or direction, but serves as a credential for generating emission parameters later.

[0057] S202 constructs a storage simulation model through a third-party storage platform, divides the interior of the convex polygon into multiple grids, and establishes a unique mapping relationship between each grid and the physical storage unit, where the total number of grids is equal to the total number of bytes of encrypted data from all data sources.

[0058] In this step, the third-party storage platform first determines the number of sides of the convex polygon based on the number of data sources participating in the federated storage. For example, when there are three data sources A, B, and C participating in the storage, the convex polygon is set as a triangle; when there are four data sources, it is set as a quadrilateral. The number of sides is related to the number of data sources. The platform divides the convex polygon into grids. The total number of grids is strictly equal to the total number of bytes of encrypted data from all data sources. For example, if the encrypted data from data source A is 200 bytes, the encrypted data from data source B is 300 bytes, and the encrypted data from data source C is 500 bytes, then the total number of bytes is 1000. The convex polygon is then divided into 1000 grids. Each grid is uniquely mapped to a physical storage unit by its number. For example, grid 1 corresponds to storage unit 1, grid 2 corresponds to storage unit 2, and so on.

[0059] S203, based on the various sub-parameters received, a third-party storage platform generates corresponding emission parameters for each data source. The emission parameters include at least the initial position, initial direction, and propagation speed of the light source within the convex polygon.

[0060] In this step, the third-party storage platform converts the 10-digit decimal seed parameters uploaded from each data source into specific transmission parameters according to preset conversion rules. Taking the seed parameters 3847291056, 5610384927, and 9201653841 from three data sources A, B, and C as an example, the conversion process is as follows:

[0061] First, the initial position coordinates are determined. The platform extracts the first four digits of various sub-parameters, and uses the first two and last two digits as the x-coordinate and y-coordinate parameters, respectively. For data source A, the first four digits 3847 correspond to x-coordinate 38 and y-coordinate 47. After mapping to the normalized coordinate range of the largest inscribed square of the convex polygon, the initial position (38, 47) is obtained. For data source B, the first four digits 5610 give the initial position (56, 10). For data source C, the first four digits 9201 give the initial position (92, 1).

[0062] Secondly, the initial propagation direction angle is determined. The platform extracts the middle three digits of various sub-parameters: data source A is 291, data source B is 038, and data source C is 165. The sum of the three is 494. The proportions of each data source are calculated to be 0.589, 0.077, and 0.334, respectively. The 360 ​​degrees are divided according to this proportion. Data source A corresponds to 212.0°, data source B corresponds to 27.7°, and data source C corresponds to 120.2°.

[0063] Finally, the propagation speed was determined by extracting the last three digits of various sub-parameters: data source A was 056, data source B was 927, and data source C was 841. The sum of the three was 1824, with each data source accounting for 0.031, 0.508, and 0.461, respectively. These data sources were distributed proportionally within the preset speed range [1, 5]. Data source A corresponded to a speed of 1.12, data source B to 3.03, and data source C to 2.84, with the unit being grids per millisecond (ms).

[0064] As a preferred embodiment of the present invention, the step of simulating the specular reflection propagation of multiple light rays in a storage simulation model based on all emission parameters, and interleaving the ciphertext bytes of each data source into the physical storage units mapped by each grid within the convex polygon according to a preset grid occupancy rule to obtain an interleaved ciphertext storage body includes:

[0065] S301, based on all transmission parameters, synchronously simulates the specular reflection propagation of multiple light rays within a convex polygon, with each light ray carrying a ciphertext byte stream from its corresponding data source.

[0066] In this step, the third-party storage platform synchronously initiates the propagation simulation of multiple light rays within the constructed convex polygon storage simulation model, based on the emission parameters of each set. Each light ray corresponds to a data source and carries the encrypted byte stream of that data source. For example, the light ray from data source A carries its 200-byte encrypted sequence, the light ray from data source B carries its 300-byte encrypted sequence, and the light ray from data source C carries its 500-byte encrypted sequence. Taking data source A as an example, its light ray starts from the starting point according to the initial position, initial direction, and propagation speed in the emission parameters and propagates in a straight line. When the light ray touches the boundary of the convex polygon, it follows the law of specular reflection, changing its propagation direction in such a way that the angle of incidence equals the angle of reflection, and continues to travel inside the convex polygon. The three light rays propagate synchronously in the simulation space without interfering with each other, each following its own reflection trajectory.

[0067] S302, monitor the propagation path of each ray. When any ray first enters a grid that has not yet been written to, write the ciphertext byte corresponding to the ray to the physical storage unit mapped by the grid in sequence. If the grid has already been written to, skip the grid and continue propagating.

[0068] In this step, the third-party storage platform monitors the propagation paths of multiple light beams in real time. Whenever a light beam enters a grid, the platform checks whether the grid has already been written to. If the grid has not been written to, the platform retrieves the ciphertext bytes corresponding to the current light beam and writes them sequentially to the physical storage unit mapped to the grid, marking the grid as occupied. If the grid has already been written to by another light beam, it is skipped, and the light beam continues to move along the propagation direction until it encounters the next empty grid before writing is performed. Taking a specific scenario as an example, suppose that at a certain moment after the simulation starts, the light beam from data source A enters grid number 37 first. At this time, the grid is empty, and the platform writes the first byte of the ciphertext stream from data source A into storage unit 37. Subsequently, the light beam from data source B also reaches grid 37 during its propagation, but because the grid is already occupied, the platform skips the grid, and the light beam from data source B continues to propagate forward until it enters the next empty grid, such as grid number 42, before writing its first byte.

[0069] S303: After the ciphertext data of any data source has been written, the light corresponding to that data source is dynamically switched to an auxiliary light source to assist the remaining unfinished data sources in continuing to search and write to the empty grid until all ciphertext data has been written, thus obtaining the interwoven ciphertext storage.

[0070] In this step, after all the encrypted data from a certain data source has been written, its corresponding ray will not be idle, but will immediately switch to an auxiliary ray to join the search for empty grids of the remaining unfinished data sources. Taking three data sources A, B, and C as an example, assuming that the encrypted data volume of data source A is 200 bytes, data source B is 300 bytes, and data source C is 500 bytes, during the simulated writing process, data source A has the smallest data volume, and its ray will complete the writing of all 200 bytes first. The ray of data source A will dynamically switch to an auxiliary ray. After that, every time the ray enters an empty grid, it will take one byte from the remaining encrypted stream of data source B or data source C and write it. As the 300 bytes of data source B are subsequently written, the ray of data source B will also switch to an auxiliary ray. At this time, there are three rays in the storage simulation model that are simultaneously searching for empty grids of the remaining encrypted data of data source C. When all the encrypted data from all data sources has been written, the physical storage units mapped by each grid have been filled with interleaved encrypted bytes, thus obtaining the interleaved encrypted storage volume.

[0071] In a preferred embodiment of the present invention, after the writing is completed, the step of the third-party storage platform deleting all received transmission parameters and retaining only the interleaved ciphertext storage body includes:

[0072] S401, after confirming that all encrypted data has been written to the corresponding physical storage unit, triggers the parameter erasure process.

[0073] In this step, the third-party storage platform continuously monitors the completion status of the optical simulation write process. When the encrypted byte streams of all data sources have been stored, the write task is completed, and the parameter erasure process is automatically triggered.

[0074] S402, clear all launch parameters and related intermediate state data used in the storage simulation process.

[0075] In this step, the third-party storage platform performs a system cleanup of all data generated during the storage simulation process. The cleanup targets include the emission parameters of each data source, the grid occupancy bitmap, and any temporary caches and logs generated during the simulation.

[0076] S403, traverse the parameter storage area, perform an overwrite operation on the memory space occupied by the emission parameters, and terminate the access process to the storage simulation model.

[0077] In this step, after performing logical deletion, the third-party storage platform further performs an overwrite operation on the physical memory space occupied by the launch parameters to prevent the recovery of deleted data through memory remnants. The platform traverses the parameter storage area and uses a preset overwrite sequence to write and overwrite each memory address that once stored launch parameters and intermediate state data multiple times. The platform terminates all access processes related to the storage simulation model, disconnects the memory mapping relationship between the process and the storage simulation model, and releases the computing resources occupied by the convex polygon mesh data structure.

[0078] like Figure 2 As shown, this embodiment of the invention also provides a data storage intelligent parallel optimization system, the system comprising:

[0079] The data encryption module 100 is used to encrypt the data to be stored from each data source, save the encryption key locally, and then upload the encrypted ciphertext data.

[0080] The reflection parameter generation module 200 is used to generate seed parameters based on encrypted data from various data sources and upload them, construct a storage simulation model through a third-party storage platform, and generate emission parameters corresponding to each data source based on the seed parameters. The storage simulation model is a convex polygon containing multiple grids, each grid corresponding to a physical storage unit.

[0081] In this system, the reflection parameter generation module 200 includes:

[0082] The seed parameter generation unit is used by each data source to calculate and generate seed parameters based on its own encrypted data, and upload the seed parameters to a third-party storage platform.

[0083] The storage mapping unit is used to build a storage simulation model through a third-party storage platform, divide the interior of the convex polygon into multiple grids, and establish a unique mapping relationship between each grid and the physical storage unit. The total number of grids is equal to the total number of bytes of encrypted data from all data sources.

[0084] The emission parameter calculation unit is used to generate corresponding emission parameters for each data source based on various received sub-parameters through a third-party storage platform. The emission parameters include at least the initial position, initial direction, and propagation speed of the light source within the convex polygon.

[0085] The storage simulation module 300 is used to simulate the specular reflection propagation of multiple light rays within the storage simulation model based on all emission parameters. According to the preset grid occupancy rules, the ciphertext bytes of each data source are interleaved and written into the physical storage units mapped by each grid within the convex polygon to obtain the interleaved ciphertext storage body.

[0086] In this system, the storage simulation module 300 includes:

[0087] The simulated propagation unit is used to synchronously simulate the specular reflection propagation of multiple light rays within a convex polygon based on all transmission parameters, with each light ray carrying a ciphertext byte stream from its corresponding data source.

[0088] The storage writing unit is used to monitor the propagation path of each ray. When any ray first enters a grid that has not yet been written to, the ciphertext bytes corresponding to the ray are written to the physical storage unit mapped to that grid in sequence. If the grid has already been written to, the ray is skipped and the propagation continues.

[0089] The auxiliary storage unit is used to dynamically switch the light corresponding to any data source to an auxiliary light source after the ciphertext data of any data source has been written, so as to assist the remaining unfinished data sources to continue searching and writing to the empty grid until all ciphertext data has been written, thus obtaining the interleaved ciphertext storage body.

[0090] The ciphertext storage module 400 is used so that after writing is completed, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage body.

[0091] In this system, the encrypted storage module 400 includes:

[0092] The erase start unit is used to trigger the parameter erase process after confirming that all encrypted data has been written to the corresponding physical storage unit;

[0093] The data erasure unit is used to erase all transmission parameters and related intermediate state data used in the storage simulation process.

[0094] The program termination unit is used to traverse the parameter storage area, perform an overwrite operation on the memory space occupied by the emission parameters, and terminate the access process to the storage simulation model.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart parallel optimization method for data storage, characterized in that, The method includes: Each data source encrypts its own data to be stored, saves the encryption key locally, and then uploads the encrypted ciphertext data. Seed parameters are generated based on encrypted data from various data sources and uploaded. A storage simulation model is constructed through a third-party storage platform, and emission parameters corresponding to each data source are generated based on the seed parameters. The storage simulation model is a convex polygon containing multiple grids, each grid corresponding to a physical storage unit. Based on all emission parameters, the specular reflection propagation of multiple light rays is simulated in the storage simulation model. According to the preset grid occupancy rules, the ciphertext bytes of each data source are interleaved and written into the physical storage units mapped by each grid in the convex polygon to obtain the interleaved ciphertext storage body. After the writing is complete, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage.

2. The intelligent parallel optimization method for data storage according to claim 1, characterized in that, The steps of generating seed parameters based on encrypted data from various data sources, uploading them, constructing a storage simulation model through a third-party storage platform, and generating transmission parameters corresponding to each data source based on the seed parameters include: Each data source calculates and generates seed parameters based on its own encrypted data, and uploads the seed parameters to a third-party storage platform; A storage simulation model is built using a third-party storage platform. The interior of the convex polygon is divided into multiple grids, and a unique mapping relationship is established between each grid and the physical storage unit. The total number of grids is equal to the total number of bytes of encrypted data from all data sources. Based on the various sub-parameters received, a third-party storage platform generates corresponding emission parameters for each data source. The emission parameters include at least the initial position, initial direction, and propagation speed of the light source within the convex polygon.

3. The intelligent parallel optimization method for data storage according to claim 1, characterized in that, The step of simulating the specular reflection propagation of multiple light rays within a storage simulation model based on all emission parameters, and interleaving the ciphertext bytes of each data source into the physical storage units mapped by each grid within the convex polygon according to a preset grid occupancy rule to obtain the interleaved ciphertext storage volume includes: Based on all transmission parameters, the specular reflection propagation of multiple light rays is synchronously simulated within a convex polygon, with each light ray carrying a ciphertext byte stream from its corresponding data source. Monitor the propagation path of each ray. When any ray first enters a grid that has not yet been written to, write the ciphertext bytes corresponding to the ray to the physical storage unit mapped to that grid in order. If the grid has already been written to, skip that grid and continue propagating. After the ciphertext data of any data source has been written, the ray corresponding to that data source is dynamically switched to an auxiliary ray to assist the remaining unfinished data sources in continuing to search and write to the empty grid until all ciphertext data has been written, thus obtaining an interwoven ciphertext storage body.

4. The intelligent parallel optimization method for data storage according to claim 1, characterized in that, After the writing is completed, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage body, including: After confirming that all encrypted data has been written to the corresponding physical storage unit, the parameter erasure process is triggered. Clear all launch parameters and related intermediate state data used during the storage simulation; Traverse the parameter storage area, overwrite the memory space occupied by the emission parameters, and terminate the access process to the storage simulation model.

5. The intelligent parallel optimization method for data storage according to claim 2, characterized in that, The seed parameter is a 10-digit decimal string. Based on the seed parameter, emission parameters corresponding to each data source are generated. Specifically, this includes: taking the first four characters of the seed parameter, where the first two are used as the x-coordinate and the last two as the y-coordinate; constructing a maximum inscribed square inside the convex polygon; normalizing the x-coordinate and y-coordinate parameters and mapping them to the maximum inscribed square to obtain the initial position coordinates of the light source corresponding to that data source; taking the middle three characters of the seed parameter; calculating the ratio of the values ​​of the middle three characters corresponding to all data sources; dividing the 360-degree circle into angle intervals corresponding to each data source based on this ratio to determine the initial propagation direction angle of each data source; and taking the last three characters of the seed parameter; calculating the ratio of the values ​​of the last three characters corresponding to all data sources; dividing the preset propagation speed interval into propagation speed values ​​corresponding to each data source based on this ratio.

6. A data storage intelligent parallel optimization system, characterized in that, The system includes: The data encryption module is used to encrypt the data to be stored from each data source, save the encryption key locally, and then upload the encrypted ciphertext data. The reflection parameter generation module is used to generate seed parameters based on encrypted data from various data sources and upload them. It constructs a storage simulation model through a third-party storage platform and generates emission parameters corresponding to each data source based on the seed parameters. The storage simulation model is a convex polygon containing multiple grids, each grid corresponding to a physical storage unit. The storage simulation module is used to simulate the specular reflection propagation of multiple light rays within the storage simulation model based on all emission parameters. According to the preset grid occupancy rules, the ciphertext bytes of each data source are interleaved and written into the physical storage units mapped by each grid within the convex polygon to obtain the interleaved ciphertext storage body. The ciphertext storage module is used so that after writing is completed, the third-party storage platform deletes all received transmission parameters and retains only the interleaved ciphertext storage body.

7. The intelligent parallel optimization system for data storage according to claim 6, characterized in that, The reflection parameter generation module includes: The seed parameter generation unit is used by each data source to calculate and generate seed parameters based on its own encrypted data, and upload the seed parameters to a third-party storage platform. The storage mapping unit is used to build a storage simulation model through a third-party storage platform, divide the interior of the convex polygon into multiple grids, and establish a unique mapping relationship between each grid and the physical storage unit. The total number of grids is equal to the total number of bytes of encrypted data from all data sources. The emission parameter calculation unit is used to generate corresponding emission parameters for each data source based on various received sub-parameters through a third-party storage platform. The emission parameters include at least the initial position, initial direction, and propagation speed of the light source within the convex polygon.

8. The intelligent parallel optimization system for data storage according to claim 6, characterized in that, The storage simulation module includes: The simulated propagation unit is used to synchronously simulate the specular reflection propagation of multiple light rays within a convex polygon based on all transmission parameters, with each light ray carrying a ciphertext byte stream from its corresponding data source. The storage writing unit is used to monitor the propagation path of each ray. When any ray first enters a grid that has not yet been written to, the ciphertext bytes corresponding to the ray are written to the physical storage unit mapped to that grid in sequence. If the grid has already been written to, the ray is skipped and the propagation continues. The auxiliary storage unit is used to dynamically switch the light corresponding to any data source to an auxiliary light source after the ciphertext data of any data source has been written, so as to assist the remaining unfinished data sources to continue searching and writing to the empty grid until all ciphertext data has been written, thus obtaining the interleaved ciphertext storage body.

9. The intelligent parallel optimization system for data storage according to claim 6, characterized in that, The encrypted storage module includes: The erase start unit is used to trigger the parameter erase process after confirming that all encrypted data has been written to the corresponding physical storage unit; The data erasure unit is used to erase all transmission parameters and related intermediate state data used in the storage simulation process. The program termination unit is used to traverse the parameter storage area, perform an overwrite operation on the memory space occupied by the emission parameters, and terminate the access process to the storage simulation model.

10. The intelligent parallel optimization system for data storage according to claim 7, characterized in that, The seed parameter is a 10-digit decimal string. Based on the seed parameter, emission parameters corresponding to each data source are generated. Specifically, this includes: taking the first four characters of the seed parameter, where the first two are used as the x-coordinate and the last two as the y-coordinate; constructing a maximum inscribed square inside the convex polygon; normalizing the x-coordinate and y-coordinate parameters and mapping them to the maximum inscribed square to obtain the initial position coordinates of the light source corresponding to that data source; taking the middle three characters of the seed parameter; calculating the ratio of the values ​​of the middle three characters corresponding to all data sources; dividing the 360-degree circle into angle intervals corresponding to each data source based on this ratio to determine the initial propagation direction angle of each data source; and taking the last three characters of the seed parameter; calculating the ratio of the values ​​of the last three characters corresponding to all data sources; dividing the preset propagation speed interval into propagation speed values ​​corresponding to each data source based on this ratio.

Images