Object storage data encryption method based on large model

Abstract

The invention particularly relates to an object storage data encryption method based on a large model. According to the object storage data encryption method based on the large model, user data is partitioned and compressed, and a deep learning large model is trained by using historical data accumulated by a cloud storage provider and known security threats; according to a large model prediction result, selecting an optimal encryption algorithm and a key management strategy; encrypting the preprocessed data, and storing the encrypted data and metadata in a cloud storage system; and when the user needs to access the data, reading the metadata, obtaining the decryption key, and recovering the original data. The object storage data encryption method based on the large model is efficient, safe and flexible, the optimal encryption strategy can be intelligently selected according to changes of data characteristics and the safety environment, dynamic encryption is achieved, therefore, the safety and privacy of data are remarkably improved, and the requirements of a modern object storage system can be met.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing, and in particular to an object storage data encryption method based on a large model. Background Art

[0002] With the rapid development of information technology, cloud computing and big data have become the mainstream technologies for modern information processing and storage. In this context, object storage has been widely used as an efficient and scalable data storage solution. Object storage systems are usually used to store large amounts of unstructured data, such as pictures, videos, log files, backup data, etc. However, with the explosive growth of data volume, data security issues have become increasingly prominent.

[0003] The main security challenges facing current object storage systems include:

[0004] Data leakage risk: Data stored in the cloud is at risk of unauthorized access and leakage, especially when there is a lack of effective encryption protection during data transmission and storage, sensitive data is more vulnerable to attacks and theft.

[0005] Data integrity risk: Data may be tampered with or damaged during storage and transmission. Ensuring the integrity and authenticity of data is another major challenge facing object storage systems.

[0006] Complex and ever-changing security threats: With the continuous escalation of network attack methods, traditional encryption methods such as AES and RSA have shown certain limitations in dealing with complex and ever-changing security threats and are unable to provide sufficient security protection.

[0007] Selection of encryption strategy: Existing encryption methods usually rely on manual selection of encryption strategies, which is difficult to flexibly adjust according to changes in data characteristics and security environment, resulting in low encryption efficiency and insufficient security.

[0008] At present, although some solutions have proposed using traditional encryption algorithms to protect data in object storage, most of these methods have the following shortcomings:

[0009] Static encryption strategy: Traditional methods usually use fixed encryption algorithms and key management strategies, which cannot be dynamically adjusted according to actual needs and changes in the security environment, resulting in poor flexibility and adaptability of encryption strategies.

[0010] Low encryption efficiency: Traditional encryption methods often have problems with low encryption and decryption efficiency when processing large-scale data, and are unable to meet the needs of efficient processing in a big data environment.

[0011] Single encryption algorithm: Existing methods usually rely on a single encryption algorithm, which cannot provide multi-level and multi-dimensional security protection and is easily targeted by attacks.

[0012] In order to solve these problems, the present invention proposes an object storage data encryption method based on a large model. Summary of the invention

[0013] In order to make up for the defects of the prior art, the present invention provides a simple and efficient object storage data encryption method based on a large model.

[0014] The present invention is achieved through the following technical solutions:

[0015] A method for encrypting object storage data based on a large model comprises the following steps:

[0016] Step S1: Data preprocessing

[0017] Data block division: The original data to be stored is divided into several data blocks according to the predetermined size to facilitate parallel encryption processing and improve encryption efficiency;

[0018] Data compression: compress the divided data to reduce storage space and increase data transmission speed;

[0019] Step S2: Large model training

[0020] Data collection: Collect a large amount of historical data, including known security threat information and corresponding optimal encryption strategies, as a training data set;

[0021] Model construction: Build a large deep learning model. The model input includes data features and security threat features, and the output is the optimal encryption strategy.

[0022] Model training: Use the training data set to train the constructed deep learning large model so that it can accurately predict potential security threats and select the optimal encryption strategy;

[0023] In step S2, a deep learning large model is constructed based on the Transformer or RNN architecture.

[0024] Step S3: Encryption strategy selection

[0025] Strategy prediction: Using the trained deep learning model, the optimal encryption strategy is predicted and selected, including the most suitable encryption algorithm and key management strategy, based on the characteristics of the data to be stored (such as data type, importance, etc.) and the current security environment (such as currently known security threats);

[0026] In step S3, the encryption strategy includes a symmetric encryption (AES) algorithm, an asymmetric encryption (RSA) algorithm, and a combined encryption algorithm;

[0027] The combined encryption algorithm combines the symmetric encryption (AES) algorithm and the asymmetric encryption (RSA) algorithm to improve security.

[0028] In step S3, the key management strategy adopts a distributed key management solution to prevent single point failure and key leakage.

[0029] Policy adjustment: Use the trained deep learning model to dynamically adjust encryption policies based on real-time security threat changes and data usage to ensure that data is always in the best protection state;

[0030] Step S4: Data encryption

[0031] Encryption algorithm application: Encrypt the pre-processed data according to the optimal encryption strategy selected by the deep learning large model;

[0032] Key management: Generate, distribute and store encryption keys according to the selected key management strategy;

[0033] Step S5: Data storage

[0034] Encrypted data storage: Storing encrypted data in the object storage system, and generating corresponding metadata, including but not limited to encryption algorithm identifier, key identifier, and data block information, to facilitate subsequent data decryption and management;

[0035] Metadata protection: Encrypt and back up the generated metadata to ensure its security during transmission and storage; Step S6: Data decryption

[0036] Metadata reading: When data needs to be accessed, the stored metadata is first read to obtain the encryption algorithm identification and key identification information;

[0037] Key acquisition: According to the key identifier, obtain the corresponding decryption key from the key management system;

[0038] Data decryption: Use the obtained decryption key and encryption algorithm to decrypt the encrypted data and restore the original data.

[0039] An object storage data encryption device based on a large model, comprising:

[0040] Data preprocessing module: responsible for preprocessing the data to be stored, including data segmentation and data compression operations. Through preliminary processing of the data, the efficiency of subsequent encryption and decryption operations can be significantly improved;

[0041] Large model training module: responsible for building and training large deep learning models, training models based on historical data and known threats, and continuously optimizing model performance according to actual needs and changes in the security environment;

[0042] The large model training module builds a deep learning large model based on the Transformer or RNN architecture.

[0043] Encryption strategy selection module: responsible for selecting the optimal encryption strategy based on the prediction results of the deep learning large model, realizing the intelligentization and dynamic adjustment of encryption strategy to ensure the security and flexibility of data;

[0044] Data encryption module: responsible for performing data encryption operations and applying the selected encryption strategy to encrypt data to provide higher security;

[0045] The data encryption module supports symmetric encryption (AES) algorithm, asymmetric encryption (RSA) algorithm, and combined encryption algorithm;

[0046] The combined encryption algorithm combines the symmetric encryption (AES) algorithm and the asymmetric encryption (RSA) algorithm to improve security.

[0047] The data encryption module supports a distributed key management scheme to prevent single point failure and key leakage.

[0048] Data storage module: responsible for storing encrypted data in the object storage system, generating corresponding metadata, and encrypting and backing up the metadata to ensure its security;

[0049] Data decryption module: responsible for reading metadata and decrypting encrypted data to restore the original data to ensure that users can efficiently and securely access stored data when needed.

[0050] An object storage data encryption device based on a large model, characterized in that it includes a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the above-mentioned method steps when executing the computer program.

[0051] A readable storage medium, characterized in that: a computer program is stored on the readable storage medium, and the computer program implements the above method steps when executed by a processor.

[0052] The beneficial effects of the present invention are as follows: the object storage data encryption method based on the large model is efficient, secure and flexible, and can intelligently select the optimal encryption strategy and realize dynamic encryption according to changes in data characteristics and security environment, thereby significantly improving the security and privacy of data and meeting the needs of modern object storage systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Attached Figure 1 It is a schematic diagram of the object storage data encryption method based on a large model of the present invention. DETAILED DESCRIPTION

[0055] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0056] The object storage data encryption method based on the large model includes the following steps:

[0057] Step S1: Data preprocessing

[0058] Data block division: The original data to be stored is divided into several data blocks according to the predetermined size to facilitate parallel encryption processing and improve encryption efficiency;

[0059] Data compression: compress the divided data to reduce storage space and increase data transmission speed;

[0060] Step S2: Large model training

[0061] Data collection: Collect a large amount of historical data, including known security threat information and corresponding optimal encryption strategies, as a training data set;

[0062] Model construction: Build a large deep learning model. The model input includes data features and security threat features, and the output is the optimal encryption strategy.

[0063] Model training: Use the training data set to train the constructed deep learning large model so that it can accurately predict potential security threats and select the optimal encryption strategy;

[0064] In step S2, a deep learning large model is constructed based on the Transformer or RNN architecture.

[0065] Step S3: Encryption strategy selection

[0066] Strategy prediction: Using the trained deep learning model, the optimal encryption strategy is predicted and selected, including the most suitable encryption algorithm and key management strategy, based on the characteristics of the data to be stored (such as data type, importance, etc.) and the current security environment (such as currently known security threats);

[0067] In step S3, the encryption strategy includes a symmetric encryption (AES) algorithm, an asymmetric encryption (RSA) algorithm, and a combined encryption algorithm;

[0068] The combined encryption algorithm combines the symmetric encryption (AES) algorithm and the asymmetric encryption (RSA) algorithm to improve security.

[0069] In step S3, the key management strategy adopts a distributed key management solution to prevent single point failure and key leakage.

[0070] Policy adjustment: Use the trained deep learning model to dynamically adjust encryption policies based on real-time security threat changes and data usage to ensure that data is always in the best protection state;

[0071] Step S4: Data encryption

[0072] Encryption algorithm application: Encrypt the pre-processed data according to the optimal encryption strategy selected by the deep learning large model;

[0073] Key management: Generate, distribute and store encryption keys according to the selected key management strategy;

[0074] Step S5: Data storage

[0075] Encrypted data storage: Storing encrypted data in the object storage system, and generating corresponding metadata, including but not limited to encryption algorithm identifier, key identifier, and data block information, to facilitate subsequent data decryption and management;

[0076] Metadata protection: Encrypt and back up the generated metadata to ensure its security during transmission and storage; Step S6: Data decryption

[0077] Metadata reading: When data needs to be accessed, the stored metadata is first read to obtain the encryption algorithm identification and key identification information;

[0078] Key acquisition: According to the key identifier, obtain the corresponding decryption key from the key management system;

[0079] Data decryption: Use the obtained decryption key and encryption algorithm to decrypt the encrypted data and restore the original data.

[0080] The object storage data encryption device based on the large model includes:

[0081] Data preprocessing module: responsible for preprocessing the data to be stored, including data segmentation and data compression operations. Through preliminary processing of the data, the efficiency of subsequent encryption and decryption operations can be significantly improved;

[0082] Large model training module: responsible for building and training large deep learning models, training models based on historical data and known threats, and continuously optimizing model performance according to actual needs and changes in the security environment;

[0083] The large model training module builds a deep learning large model based on the Transformer or RNN architecture.

[0084] Encryption strategy selection module: responsible for selecting the optimal encryption strategy based on the prediction results of the deep learning large model, realizing the intelligentization and dynamic adjustment of encryption strategy to ensure the security and flexibility of data;

[0085] Data encryption module: responsible for performing data encryption operations and applying the selected encryption strategy to encrypt data to provide higher security;

[0086] The data encryption module supports symmetric encryption (AES) algorithm, asymmetric encryption (RSA) algorithm, and combined encryption algorithm;

[0087] The combined encryption algorithm combines the symmetric encryption (AES) algorithm and the asymmetric encryption (RSA) algorithm to improve security.

[0088] The data encryption module supports a distributed key management scheme to prevent single point failure and key leakage.

[0089] Data storage module: responsible for storing encrypted data in the object storage system, generating corresponding metadata, and encrypting and backing up the metadata to ensure its security;

[0090] Data decryption module: responsible for reading metadata and decrypting encrypted data to restore the original data to ensure that users can efficiently and securely access stored data when needed.

[0091] The object storage data encryption device based on a large model includes a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the above method steps when executing the computer program.

[0092] The readable storage medium stores a computer program, and when the computer program is executed by a processor, the above method steps are implemented.

[0093] Compared with the existing technology, the object storage data encryption method based on the large model has the following characteristics:

[0094] First, through data segmentation and compression preprocessing, the efficiency of data encryption and decryption is improved, which is suitable for large-scale data processing scenarios.

[0095] Second, through deep learning large models, the optimal encryption strategy can be intelligently selected based on data characteristics and security environment, avoiding the limitations of manual selection in traditional encryption methods.

[0096] Third, dynamic adjustment of encryption policies is achieved. Encryption policies can be adjusted at any time according to real-time security threat changes and data usage to ensure that data is always in the best protection state.

[0097] Fourth, the combination of multiple encryption algorithms and distributed key management solutions significantly improves data security and anti-attack capabilities.

[0098] The embodiment described above is only one specific implementation of the present invention. Common changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for encrypting object storage data based on a large model, characterized in that: The following steps are involved: Step S1: Data preprocessing Data block division: The original data to be stored is divided into several data blocks according to the predetermined size to facilitate parallel encryption processing and improve encryption efficiency; Data compression: compress the divided data to reduce storage space and increase data transmission speed; Step S2: Large model training Data collection: Collect historical data, including known security threat information and corresponding optimal encryption strategies, as training data sets; Model construction: Build a large deep learning model. The model input includes data features and security threat features, and the output is the optimal encryption strategy. Model training: Use the training data set to train the constructed deep learning large model so that it can accurately predict potential security threats and select the optimal encryption strategy; Step S3: Encryption strategy selection Strategy prediction: Use the trained deep learning model to predict and select the optimal encryption strategy, including the most suitable encryption algorithm and key management strategy, based on the characteristics of the data to be stored and the current security environment; Policy adjustment: Use the trained deep learning model to dynamically adjust encryption policies based on real-time security threat changes and data usage to ensure that data is always in the best protection state; Step S4: Data encryption Encryption algorithm application: Encrypt the pre-processed data according to the optimal encryption strategy selected by the deep learning large model; Key management: Generate, distribute and store encryption keys according to the selected key management strategy; Step S5: Data storage Encrypted data storage: Storing encrypted data in the object storage system, and generating corresponding metadata, including but not limited to encryption algorithm identifier, key identifier, and data block information, to facilitate subsequent data decryption and management; Metadata protection: Encrypt and back up the generated metadata to ensure its security during transmission and storage; Step S6: Data decryption Metadata reading: When data needs to be accessed, the stored metadata is first read to obtain the encryption algorithm identification and key identification information; Key acquisition: According to the key identifier, obtain the corresponding decryption key from the key management system; Data decryption: Use the obtained decryption key and encryption algorithm to decrypt the encrypted data and restore the original data.

2. The object storage data encryption method based on a large model according to claim 1 is characterized in that: In step S2, a deep learning large model is constructed based on the Transformer or RNN architecture.

3. The object storage data encryption method based on a large model according to claim 1 is characterized in that: In step S3, the encryption strategy includes a symmetric encryption AES algorithm, an asymmetric encryption RSA algorithm, and a combined encryption algorithm; The combined encryption algorithm combines the symmetric encryption AES algorithm and the asymmetric encryption RSA algorithm to improve security.

4. The object storage data encryption method based on a large model according to claim 3 is characterized in that: In step S3, the key management strategy adopts a distributed key management solution to prevent single point failure and key leakage.

5. A large-model-based object storage data encryption device, characterized in that: include: Data preprocessing module: responsible for preprocessing the data to be stored, including data segmentation and data compression operations, to improve the efficiency of subsequent encryption and decryption operations; Large model training module: responsible for building and training large deep learning models, training models based on historical data and known threats, and continuously optimizing model performance according to actual needs and changes in the security environment; Encryption strategy selection module: responsible for selecting the optimal encryption strategy based on the prediction results of the deep learning large model, realizing the intelligentization and dynamic adjustment of encryption strategy to ensure the security and flexibility of data; Data encryption module: responsible for performing data encryption operations and applying the selected encryption strategy to encrypt data to provide higher security; Data storage module: responsible for storing encrypted data in the object storage system, generating corresponding metadata, and encrypting and backing up the metadata to ensure its security; Data decryption module: responsible for reading metadata and decrypting encrypted data to restore the original data to ensure that users can efficiently and securely access stored data when needed.

6. The object storage data encryption system based on a large model according to claim 5, characterized in that: The large model training module builds a deep learning large model based on the Transformer or RNN architecture.

7. The object storage data encryption system based on a large model according to claim 5, characterized in that: The data encryption module supports symmetric encryption AES algorithm, asymmetric encryption RSA algorithm, and combined encryption algorithm; The combined encryption algorithm combines the symmetric encryption AES algorithm and the asymmetric encryption RSA algorithm to improve security.

8. The object storage data encryption system based on a large model according to claim 5 or 7, characterized in that: The data encryption module supports a distributed key management scheme to prevent single point failure and key leakage.

9. An object storage data encryption device based on a large model, characterized in that: The method comprises a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method according to any one of claims 1 to 4 when executing the computer program.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images