An AI intelligent computing center computing and storage integrated cloud platform and security protection method

Abstract

The present invention relates to the field of data information transmission technology, and in particular to an AI intelligent computing center computing and storage integrated cloud platform and security protection method, including edge computing modules for providing edge intelligent computing power; AI intelligent computing units for performing AI intelligent computing; edge storage units for storing data generated by the AI ​​intelligent computing units as intelligent computing data; various time-effectiveness backup modules for real-time backup of intelligent computing data generated by each edge computing module, and planning and management, and also for adjusting the planning and management process of real-time backup, and also for optimizing the data backup state, and also for encrypting and uploading real-time backup data as uploaded data, and adjusting it; verification storage module for storing uploaded data; storage verification unit for performing storage judgment on uploaded data; decryption storage unit for storing uploaded data judged by storage. The present invention improves the intelligent computing efficiency and security of AI intelligent computing.

Description

Technical Field

[0001] The present invention relates to the field of data information transmission technology, and in particular to an AI intelligent computing center computing and storage integrated cloud platform and a security protection method. Background Art

[0002] The integration of computing and storage can significantly improve the computing efficiency of the AI ​​intelligent computing center. By integrating and optimizing computing resources and storage resources, it can reduce the delay in data access and increase the speed of data processing, thereby meeting the real-time and efficiency requirements of AI algorithms. Security protection of data transmission is the key to ensuring the data security of the AI ​​intelligent computing center. Implementing effective security protection measures, such as data encryption, identity authentication, and access control, can ensure the integrity and confidentiality of data during transmission and prevent data from being acquired or tampered with by unauthorized third parties.

[0003] China Patent Publication No.: CN116137622A discloses a highly compatible trusted cloud service platform, which is divided into computing area, security area and storage area according to functional areas from top to bottom. The computing area provides fast and secure computing and processing services for users' massive data. The computing area is connected to a network switch below, and the network is configured and the corresponding port is allocated according to the server IP address; the security area is composed of security audit, firewall and other integrated security mechanisms from top to bottom, and has an independent security protection system; the storage area adopts SAN storage area network to provide large-capacity data storage, which is geographically dispersed and supports compatibility with multiple databases and middleware. However, this solution is only divided into computing area, security area and storage area according to functional areas, and cannot achieve the computing and storage integration and security protection requirements of the AI ​​intelligent computing center. Summary of the invention

[0004] To this end, the present invention provides an AI intelligent computing center computing and storage integrated cloud platform and security protection method to overcome the problems of low intelligent computing efficiency and security caused by the lack of computing and storage planning and security protection for AI intelligent computing in the prior art.

[0005] To achieve the above objectives, on the one hand, the present invention provides an AI intelligent computing center computing and storage integrated cloud platform, including:

[0006] Each edge computing module is used to provide edge intelligent computing power, and is connected to each time-sensitive backup module respectively;

[0007] AI intelligent computing unit, which is set in each edge computing module to perform AI intelligent computing;

[0008] An edge storage unit, which is provided in each edge computing module and is used to store the data generated by the AI ​​intelligent computing unit as intelligent computing data, and the edge storage unit is connected to the AI ​​intelligent computing unit;

[0009] Each time-limited backup module is used to back up the intelligent computing data generated by each edge computing module in real time, and to plan and manage the real-time backup data. It is also used to adjust the planning and management process of real-time backup according to external abnormal access conditions, and to optimize the data backup status according to the edge computing module status. It is also used to set the upload key to encrypt and upload the real-time backup data as the uploaded data, and adjust the uploaded data;

[0010] Verify that the storage module is used to store the uploaded data and that the storage module is connected to each time-limited backup module;

[0011] A storage verification unit, which is arranged in the verification storage module and is used to perform storage judgment on the uploaded data according to the upload key;

[0012] The decryption storage unit is arranged in the verification storage module and is used to store the uploaded data that passes the storage judgment. The decryption storage unit is connected to the storage verification unit.

[0013] Furthermore, the AI ​​intelligent computing unit performs AI intelligent computing, and stores the intelligent computing data generated by the AI ​​intelligent computing in the edge storage unit through a virtual private network.

[0014] Further, the time-limited backup module establishes an SSH connection with the edge storage unit, the time-limited backup module provides an SSH key, compares the SSH key with a preset SSH key, and authenticates the time-limited backup module according to the comparison result, wherein:

[0015] When the SSH key is inconsistent with the preset SSH key, it is determined that the time-based backup module identity authentication fails;

[0016] When the SSH key is consistent with the preset SSH key, the time-limited backup module identity authentication is determined to be successful. The time-limited backup module and the edge storage unit open an SFTP session, and the intelligent computing data generated by the edge computing module is acquired in real time through the SFTP session and stored and backed up as real-time backup data.

[0017] Furthermore, the time-effective backup module compares the upload time interval T with the preset upload time interval T0, and plans and manages the real-time backup data according to the comparison result, wherein:

[0018] When T<T0, the time-effective backup module continues to store the real-time backup data;

[0019] When T≥T0, the time-effective backup module sets the real-time backup data stored in the upload time interval as the upload data, encrypts the upload data and uploads it to the verification storage module with the upload key, and re-times the upload time interval.

[0020] Furthermore, the time-effective backup module determines the data backup status according to the external abnormal access situation, and adjusts the planning and management process of the real-time backup according to the determination result, wherein:

[0021] When there is an abnormal external access situation, the time-effective backup module determines that the data backup state is abnormal, and sets the preset upload time interval T0 in the real-time backup planning management process to 0;

[0022] When there is no external abnormal access, the time-effective backup module determines that the data backup state is normal and does not adjust the planning and management process of the real-time backup.

[0023] Furthermore, the time-effective backup module optimizes the data backup state according to the edge computing module state, wherein:

[0024] When the edge computing module is in an abnormal state, the time-effective backup module optimizes the data backup state and optimizes the data backup state when there is no external abnormal access to an abnormal state;

[0025] When the edge computing module is in normal status, the time backup module does not optimize the data backup status.

[0026] Furthermore, when setting the upload key, the time-limited backup module generates the upload key using the genpkey command in OpenSSL, and shares the upload key with the storage verification unit through a key exchange protocol;

[0027] When encrypting the uploaded data, the time-limited backup module performs data padding on the uploaded data through ZeroPadding, generates an initialization vector according to the uploaded data after the data padding, divides the uploaded data after the data padding into blocks of a preset fixed size, selects the cipher block chaining mode in the AES symmetric encryption algorithm to encrypt each block separately to obtain a ciphertext, converts the ciphertext into an ASCII string using Base64 encoding, and uploads the converted ASCII string, the initialization vector and the upload key.

[0028] Furthermore, the time-effective backup module adjusts the uploaded data according to the data adjustment signal, and deletes the uploaded data corresponding to the data adjustment signal after a safe time interval.

[0029] Furthermore, the storage verification unit compares the upload key with the shared key, makes a storage judgment on the uploaded data according to the comparison result, and sends a data adjustment signal to the timeliness backup module according to the judgment result, wherein:

[0030] When the upload key is inconsistent with the shared key, the storage verification unit determines that the uploaded data has not passed the storage judgment and does not send a data adjustment signal to the time-limited backup module;

[0031] When the upload key is consistent with the shared key, the storage verification unit determines that the uploaded data passes the storage judgment and sends a data adjustment signal to the time-effectiveness backup module;

[0032] The decryption storage unit uses the same AES symmetric encryption algorithm as the time-effectiveness backup module to decrypt the uploaded data, and stores the decrypted uploaded data.

[0033] On the other hand, the present invention also provides a security protection method for an AI intelligent computing center computing and storage integrated cloud platform, comprising:

[0034] Step S1, performing AI intelligent calculation by an AI intelligent calculation unit;

[0035] Step S2, storing the data generated by the AI ​​intelligent computing unit as intelligent computing data through the edge storage unit;

[0036] Step S3, backing up the intelligent computing data generated by each edge computing module in real time through the time-effectiveness backup module;

[0037] Step S4, planning and managing the real-time backup data through the time-effective backup module;

[0038] Step S5, adjusting the real-time backup planning and management process according to the external abnormal access situation through the time-effective backup module;

[0039] Step S6, optimizing the data backup status according to the edge computing module status through the time backup module;

[0040] Step S7, setting an upload key through the time-based backup module to encrypt and upload the real-time backup data as the upload data, and adjusting the upload data;

[0041] Step S8, performing storage determination on the uploaded data according to the upload key by the storage verification unit;

[0042] Step S9: storing the uploaded data that passes the storage judgment in the decryption storage unit.

[0043] Compared with the prior art, the beneficial effect of the present invention lies in that the platform provides edge intelligent computing power through each edge computing module, so as to push data processing and intelligent analysis tasks closer to the data source, the platform performs AI intelligent computing through the AI ​​intelligent computing unit and stores the data generated by the AI ​​intelligent computing unit as intelligent computing data through the edge storage unit, so as to reduce the delay of data transmission, improve the response speed, and meet the application scenarios with high real-time requirements, the platform performs real-time backup of the intelligent computing data generated by each edge computing module through each time-effect backup module, and plans and manages the real-time backup data, so as to encrypt and upload the uploaded data at every upload time interval, thereby improving the security of data transmission, the platform adjusts the planning and management process of real-time backup through each time-effect backup module according to external abnormal access conditions, so as to It is convenient to upload data in time, reduce the impact of external abnormal access on data upload, and ensure the integrity of data. The platform optimizes the data backup status according to the edge computing module status through each time-limited backup module to upload data in time, reduce the impact of abnormal edge computing module status on data upload, and ensure the integrity of data. The platform sets the upload key through each time-limited backup module to encrypt and upload the real-time backup data as the uploaded data so as to transmit the uploaded data safely. The platform stores the uploaded data through the verification storage module. The platform performs storage judgment on the uploaded data according to the upload key through the storage verification unit and stores the uploaded data that passes the storage judgment through the decryption storage unit. While realizing the integration of computing and storage of the AI ​​intelligent computing center, it ensures the safe transmission of data, breaks the information island, and improves the efficiency and security of AI intelligent computing. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic diagram of the structure of the AI ​​intelligent computing center computing and storage integrated cloud platform of this embodiment;

[0045] Figure 2 The figure is a flow chart of the security protection method of the integrated computing and storage cloud platform of the AI ​​intelligent computing center in this embodiment. DETAILED DESCRIPTION

[0046] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0047] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0048] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] See also Figure 1 As shown, it is a schematic diagram of the structure of the AI ​​intelligent computing center computing and storage integrated cloud platform of this embodiment, and the cloud platform includes:

[0050] Each edge computing module is used to provide edge intelligent computing power, and is connected to each time-sensitive backup module respectively;

[0051] AI intelligent computing unit, which is set in each edge computing module to perform AI intelligent computing;

[0052] An edge storage unit, which is provided in each edge computing module and is used to store the data generated by the AI ​​intelligent computing unit as intelligent computing data, and the edge storage unit is connected to the AI ​​intelligent computing unit;

[0053] Each time-limited backup module is used to back up the intelligent computing data generated by each edge computing module in real time, and to plan and manage the real-time backup data. It is also used to adjust the planning and management process of real-time backup according to external abnormal access conditions, and to optimize the data backup status according to the edge computing module status. It is also used to set the upload key to encrypt and upload the real-time backup data as the uploaded data, and adjust the uploaded data;

[0054] Verify that the storage module is used to store the uploaded data and that the storage module is connected to each time-limited backup module;

[0055] A storage verification unit, which is arranged in the verification storage module and is used to perform storage judgment on the uploaded data according to the upload key;

[0056] The decryption storage unit is arranged in the verification storage module and is used to store the uploaded data that passes the storage judgment. The decryption storage unit is connected to the storage verification unit.

[0057] Specifically, the cloud platform is applied to intelligent computing systems with high security requirements. By setting the AI ​​intelligent computing center to an integrated computing and storage mode and performing security protection on the intelligent computing data, the security of data transmission is improved. The platform provides edge intelligent computing power through each edge computing module, so as to push data processing and intelligent analysis tasks closer to the data source. The platform performs AI intelligent computing through the AI ​​intelligent computing unit and stores the data generated by the AI ​​intelligent computing unit as intelligent computing data through the edge storage unit, so as to reduce the delay in data transmission, improve the response speed, and meet the application scenarios with high real-time requirements. The platform performs real-time back-up of the intelligent computing data generated by each edge computing module through each time backup module, and plans and manages the real-time backup data, so as to encrypt and upload the uploaded data at every upload time interval, thereby improving the security of data transmission. The platform uses each time backup module based on The planning and management process of real-time backup is adjusted according to external abnormal access conditions to facilitate timely data upload, reduce the impact of external abnormal access conditions on data upload, and ensure data integrity. The platform optimizes the data backup status according to the edge computing module status through each time-limited backup module to upload data in time, reduce the impact of abnormal edge computing module status on data upload, and ensure data integrity. The platform sets an upload key through each time-limited backup module to encrypt and upload the real-time backup data as uploaded data so as to safely transmit the uploaded data. The platform stores the uploaded data through a verification storage module. The platform performs storage judgment on the uploaded data according to the upload key through a storage verification unit and stores the uploaded data that passes the storage judgment through a decryption storage unit. While realizing the integration of computing and storage in the AI ​​intelligent computing center, it ensures data security transmission, breaks information islands, and improves the efficiency and security of AI intelligent computing.

[0058] Specifically, the number of edge computing modules and the number of time-sensitive backup modules are set to be the same, and both are K+n, where K and n are positive integers greater than 0.

[0059] It can be understood that this embodiment does not limit the setting method of each edge computing module. Technical personnel in this field can set it according to actual needs. For example, each edge computing module can be set according to computing power requirements. The edge intelligent computing power refers to the intelligent computing capability performed on the edge nodes of the network.

[0060] Specifically, the edge computing module pushes data processing and intelligent analysis tasks closer to the data source in order to reduce data transmission delays, improve response speed, and meet application scenarios with high real-time requirements.

[0061] Specifically, the AI ​​intelligent computing unit performs AI intelligent computing, and stores the intelligent computing data generated by the AI ​​intelligent computing in the edge storage unit through a virtual private network.

[0062] It can be understood that this embodiment does not limit the setting method of the virtual private network, and technical personnel in this field can freely set it according to actual needs. For example, a secure data tunnel can be created on the public network through a virtual private network, and the intelligent computing data is transmitted through the secure data tunnel. The intelligent computing data in the tunnel is encrypted data. After the edge storage unit is authenticated, the intelligent computing data in the tunnel is decrypted and the intelligent computing data is obtained for storage.

[0063] Specifically, the AI ​​intelligent computing unit creates a secure data tunnel through virtual private network tunnel technology to transmit intelligent computing data, so as to ensure the integrity and authenticity of the data and prevent the data from being tampered with or destroyed during transmission.

[0064] Specifically, the time-limited backup module establishes an SSH connection with the edge storage unit, the time-limited backup module provides an SSH key, compares the SSH key with a preset SSH key, and authenticates the time-limited backup module according to the comparison result, wherein:

[0065] When the SSH key is inconsistent with the preset SSH key, it is determined that the time-based backup module identity authentication fails;

[0066] When the SSH key is consistent with the preset SSH key, the time-limited backup module identity authentication is determined to be successful. The time-limited backup module and the edge storage unit open an SFTP session, and the intelligent computing data generated by the edge computing module is acquired in real time through the SFTP session and stored and backed up as real-time backup data.

[0067] Specifically, the SSH refers to the file transfer protocol, the SSH connection refers to the data transmission connection established according to the file transfer protocol, the SSH key refers to the key used for identity authentication during the data transmission process, the preset SSH key refers to the preset comparison key used for identity authentication during the data transmission process, the SFTP refers to the secure transmission protocol for encrypted data transmission, and the SFTP session refers to the encrypted transmission method used to browse, upload and download data after successful identity authentication.

[0068] Specifically, the time-effective backup module stores and backs up the real-time backup data through an SFTP session to ensure the confidentiality and integrity of the data during transmission.

[0069] Specifically, the time-effective backup module compares the upload time interval T with the preset upload time interval T0, and plans and manages the real-time backup data according to the comparison result, wherein:

[0070] When T<T0, the time-effective backup module continues to store the real-time backup data;

[0071] When T≥T0, the time-effective backup module sets the real-time backup data stored in the upload time interval as the upload data, encrypts the upload data and uploads it to the verification storage module with the upload key, and re-times the upload time interval.

[0072] Specifically, the upload time interval refers to the time interval from the current time to the last time the data was set to be uploaded. The preset upload time interval refers to the preset value of the time interval that reflects the upload data setting requirements, such as T0=3mins. The upload key refers to the key used to authenticate the identity of the data uploader, such as the timeliness backup module. The re-timing refers to resetting the upload time interval to zero and then re-timing.

[0073] Specifically, the time-based backup module plans and manages the real-time backup data. When the upload time interval is greater than or equal to the preset upload time interval, the uploaded data is encrypted and uploaded to the verification storage module along with the upload key, and the upload time interval is re-timed so that the uploaded data can be encrypted and uploaded every upload time interval.

[0074] Specifically, the time-effective backup module determines the data backup status according to the external abnormal access situation, and adjusts the planning and management process of real-time backup according to the determination result, wherein:

[0075] When there is an abnormal external access situation, the time-effective backup module determines that the data backup state is abnormal, and sets the preset upload time interval T0 in the real-time backup planning management process to 0;

[0076] When there is no external abnormal access, the time-effective backup module determines that the data backup state is normal and does not adjust the planning and management process of the real-time backup.

[0077] Specifically, the external abnormal access situation refers to the abnormal access situation to each edge computing module and each time backup module from the outside. This embodiment does not limit the method for obtaining the external abnormal access situation. Technical personnel in this field can freely set it according to actual needs. It only needs to meet the identification requirements of external abnormal access. For example, real-time analysis of network traffic patterns and trends can be set, including the transmission rate, size and frequency of data packets. If an abnormal traffic pattern is found, such as a sudden surge in traffic or an abnormal data packet size, it indicates that there is an external abnormal access situation. The source IP address of the data packet can also be set to analyze to find IP addresses that do not belong to normal business traffic. If a large number of accesses from unknown or suspicious IP addresses are found, it indicates that there is an external abnormal access situation.

[0078] Specifically, when there is an external abnormal access situation, the time-effective backup module sets the preset upload time interval T0=0 in the real-time backup planning management process to facilitate timely data upload, reduce the impact of external abnormal access on data upload, and ensure data integrity.

[0079] Specifically, the time-effective backup module optimizes the data backup state according to the edge computing module state, wherein:

[0080] When the edge computing module is in an abnormal state, the time-effective backup module optimizes the data backup state and optimizes the data backup state when there is no external abnormal access to an abnormal state;

[0081] When the edge computing module is in normal status, the time backup module does not optimize the data backup status.

[0082] Specifically, the edge computing module status refers to the fault status of the AI ​​intelligent computing unit and the edge storage unit in the edge computing module. When the AI ​​intelligent computing unit and the edge storage unit are faulty, the edge computing module status is determined to be abnormal. When the AI ​​intelligent computing unit and the edge storage unit are not faulty, the edge computing module status is determined to be normal. This embodiment does not limit the method for obtaining the fault status of the AI ​​intelligent computing unit and the edge storage unit. Technical personnel in this field can freely set it according to actual needs. For example, the AI ​​intelligent computing unit can be set to perform performance monitoring, such as processing speed, calculation accuracy, etc. When the performance degrades, it is determined that the AI ​​intelligent computing unit is faulty. Error log analysis, input and output comparison, and system resource inspection can also be performed on the AI ​​intelligent computing unit. Data consistency check can be performed on the edge storage unit to verify whether the data in the storage unit is consistent and complete. Read and write tests, error code and log analysis, and hardware status inspection can also be performed on the edge storage unit.

[0083] Specifically, when the edge computing module is in an abnormal state, the time-sensitive backup module optimizes the data backup state when there is no external abnormal access to an abnormal state, so as to upload data in a timely manner, reduce the impact of the abnormal edge computing module state on data upload, and ensure data integrity.

[0084] Specifically, when setting the upload key, the time-effective backup module uses the genpkey command in OpenSSL to generate the upload key, and shares the upload key with the storage verification unit through a key exchange protocol.

[0085] Specifically, the OpenSSL refers to a secure socket layer cryptographic library that includes various commonly used cryptographic algorithms, key generation, certificate encapsulation management functions, and the SSL protocol. The genpkey command refers to a tool in OpenSSL that generates private key values ​​for various keys. This embodiment does not specifically limit the type of key exchange protocol. Technical personnel in this field can freely set it according to actual needs. It is only necessary to meet the sharing requirements of the uploaded key. For example, the key exchange protocol can be set to the Otway-Rees protocol.

[0086] Specifically, when encrypting the uploaded data, the time-limited backup module fills the uploaded data with data through ZeroPadding, generates an initialization vector according to the uploaded data after the data is filled, and divides the uploaded data after the data is filled into blocks of a preset fixed size, selects the cipher block chaining mode in the AES symmetric encryption algorithm to encrypt each block separately to obtain a ciphertext, uses Base64 encoding to convert the ciphertext into an ASCII string, and uploads the converted ASCII string, initialization vector and upload key.

[0087] Specifically, the ZeroPadding refers to a padding operation performed on plaintext data before encryption begins to ensure that the data length meets the padding requirements of the encryption algorithm. The initialization vector refers to a random number or pseudo-random number used together with the key to ensure that even if the plaintext is the same, the present embodiment does not limit the preset fixed size, and those skilled in the art can freely set it according to actual needs, as long as the requirements for the division of uploaded data are met, such as the preset fixed size can be set to 128 bits, 192 bits or 256 bits. The AES symmetric encryption algorithm refers to, the cipher block chaining mode refers to an encryption method for plaintext data that introduces a chaining mechanism, the Base64 encoding refers to a ciphertext processing method for converting binary ciphertext into an ASCII character string, and the ASCII character string refers to a character string composed of ASCII code characters.

[0088] Specifically, data padding is performed by adding specific bytes at the end of the data so that the data length reaches the block size expected by the encryption algorithm. An initialization vector is generated and securely transmitted together with the ciphertext for use in decryption, thereby ensuring that the security of the uploaded data encryption is enhanced even if the plaintext data is the same. The ciphertext is converted into an ASCII string so that the uploaded data can be securely transmitted.

[0089] Specifically, the time-effective backup module adjusts the uploaded data according to the data adjustment signal, and deletes the uploaded data corresponding to the data adjustment signal after a safe time interval has passed.

[0090] Specifically, the safety time interval refers to a preset value of the time interval from when the time-effective backup module receives the data adjustment signal to when the uploaded data is deleted. For example, the safety time interval can be set to 3 minutes.

[0091] Specifically, the time-limited backup module deletes the uploaded data corresponding to the data adjustment signal after a safe time interval, so as to release backup storage space, ensure the storage margin of the time-limited backup module, and improve the backup storage efficiency of the time-limited backup module.

[0092] Specifically, the storage verification unit compares the upload key with the shared key, makes a storage judgment on the uploaded data according to the comparison result, and sends a data adjustment signal to the timeliness backup module according to the judgment result, wherein:

[0093] When the upload key is inconsistent with the shared key, the storage verification unit determines that the uploaded data has not passed the storage judgment and does not send a data adjustment signal to the time-limited backup module;

[0094] When the upload key is consistent with the shared key, the storage verification unit determines that the uploaded data passes the storage judgment and sends a data adjustment signal to the timeliness backup module.

[0095] Specifically, the shared key refers to the upload key shared by the timeliness backup module and the storage verification unit through a key exchange protocol, and the data adjustment signal refers to the signal sent by the storage verification unit to the timeliness backup module to delete the uploaded data that passes the storage judgment.

[0096] Specifically, when the upload key is consistent with the shared key, the storage verification unit determines that the uploaded data passes the storage judgment, so as to facilitate the secure transmission of the uploaded data, thereby improving the security of data transmission.

[0097] Specifically, the decryption storage unit uses the same AES symmetric encryption algorithm as the time-effectiveness backup module to decrypt the uploaded data, and stores the decrypted uploaded data.

[0098] Specifically, the decryption storage unit decrypts and stores the data generated by the edge computing power so that in the event of equipment failure or data loss, it can be recovered from the stored data to ensure business continuity and facilitate subsequent analysis and mining to obtain valuable information, thereby supporting decision-making or business optimization, enabling data sharing and collaboration between different users, departments or organizations, breaking down information silos, promoting cross-departmental communication and cooperation, improving work efficiency, and improving the efficiency of AI intelligent computing while realizing the integration of computing and storage in the AI ​​intelligent computing center.

[0099] See also Figure 2As shown, it is a flow chart of the security protection method of the AI ​​intelligent computing center computing and storage integrated cloud platform in this embodiment, and the method includes:

[0100] Step S1, performing AI intelligent calculation by an AI intelligent calculation unit;

[0101] Step S2, storing the data generated by the AI ​​intelligent computing unit as intelligent computing data through the edge storage unit;

[0102] Step S3, backing up the intelligent computing data generated by each edge computing module in real time through the time-effectiveness backup module;

[0103] Step S4, planning and managing the real-time backup data through the time-effective backup module;

[0104] Step S5, adjusting the real-time backup planning and management process according to the external abnormal access situation through the time-effective backup module;

[0105] Step S6, optimizing the data backup status according to the edge computing module status through the time backup module;

[0106] Step S7, setting an upload key through the time-based backup module to encrypt and upload the real-time backup data as the upload data, and adjusting the upload data;

[0107] Step S8, performing storage determination on the uploaded data according to the upload key by the storage verification unit;

[0108] Step S9: storing the uploaded data that passes the storage judgment in the decryption storage unit.

[0109] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An AI intelligent computing center computing and storage integrated cloud platform, characterized in that: include: Each edge computing module is used to provide edge intelligent computing power, and is connected to each time-sensitive backup module respectively; AI intelligent computing unit, which is set in each edge computing module to perform AI intelligent computing; An edge storage unit, which is provided in each edge computing module and is used to store the data generated by the AI ​​intelligent computing unit as intelligent computing data, and the edge storage unit is connected to the AI ​​intelligent computing unit; Each time-limited backup module is used to back up the intelligent computing data generated by each edge computing module in real time, and to plan and manage the real-time backup data. It is also used to adjust the planning and management process of real-time backup according to external abnormal access conditions, and to optimize the data backup status according to the edge computing module status. It is also used to set the upload key to encrypt and upload the real-time backup data as the uploaded data, and adjust the uploaded data; Verify that the storage module is used to store the uploaded data and that the storage module is connected to each time-limited backup module; A storage verification unit, which is arranged in the verification storage module and is used to perform storage judgment on the uploaded data according to the upload key; A decryption storage unit, which is disposed in the verification storage module, for storing the uploaded data determined by the storage, the decryption storage unit is connected to the storage verification unit; The AI ​​intelligent computing unit performs AI intelligent computing and stores the intelligent computing data generated by the AI ​​intelligent computing in the edge storage unit through a virtual private network; The time-limited backup module establishes an SSH connection with the edge storage unit, the time-limited backup module provides an SSH key, compares the SSH key with a preset SSH key, and authenticates the time-limited backup module according to the comparison result, wherein: When the SSH key is inconsistent with the preset SSH key, it is determined that the time-based backup module identity authentication fails; When the SSH key is consistent with the preset SSH key, the time-limited backup module identity authentication is determined to be successful, and the time-limited backup module opens an SFTP session with the edge storage unit, and obtains the intelligent computing data generated by the edge computing module in real time through the SFTP session and stores and backs it up as real-time backup data; The time-effective backup module compares the upload time interval T with the preset upload time interval T0, and plans and manages the real-time backup data according to the comparison result, wherein: When T<T0, the time-effective backup module continues to store the real-time backup data; When T≥T0, the time-limited backup module sets the real-time backup data stored in the upload time interval as the upload data, encrypts the upload data and uploads it to the verification storage module with the upload key, and re-times the upload time interval; The time-effective backup module determines the data backup status according to the external abnormal access situation, and adjusts the real-time backup planning and management process according to the judgment result, wherein: When there is an external abnormal access situation, the time-effective backup module determines that the data backup state is abnormal, and sets the preset upload time interval T0 in the real-time backup planning management process to 0; When there is no external abnormal access, the time-based backup module determines that the data backup status is normal and does not adjust the real-time backup planning and management process; The time-effective backup module optimizes the data backup status according to the edge computing module status, wherein: When the edge computing module is in an abnormal state, the time-effective backup module optimizes the data backup state and optimizes the data backup state when there is no external abnormal access to an abnormal state; When the edge computing module is in normal state, the time backup module does not optimize the data backup state; When setting the upload key, the time-limited backup module uses the genpkey command in OpenSSL to generate the upload key, and shares the upload key with the storage verification unit through the key exchange protocol; When encrypting the uploaded data, the time-limited backup module fills the uploaded data with data through ZeroPadding, generates an initialization vector according to the uploaded data after the data filling, divides the uploaded data after the data filling into blocks of a preset fixed size, selects the cipher block chaining mode in the AES symmetric encryption algorithm to encrypt each block separately to obtain a ciphertext, converts the ciphertext into an ASCII string using Base64 encoding, and uploads the converted ASCII string, the initialization vector and the upload key; The time backup module adjusts the uploaded data according to the data adjustment signal, and deletes the uploaded data corresponding to the data adjustment signal after a safe time interval; The storage verification unit compares the upload key with the shared key, makes a storage judgment on the uploaded data according to the comparison result, and sends a data adjustment signal to the timeliness backup module according to the judgment result, wherein: When the upload key is inconsistent with the shared key, the storage verification unit determines that the uploaded data has not passed the storage judgment and does not send a data adjustment signal to the time-limited backup module; When the upload key is consistent with the shared key, the storage verification unit determines that the uploaded data passes the storage judgment and sends a data adjustment signal to the time-effectiveness backup module; The decryption storage unit uses the same AES symmetric encryption algorithm as the time-effectiveness backup module to decrypt the uploaded data, and stores the decrypted uploaded data.

2. A security protection method applied to the AI ​​intelligent computing center computing and storage integrated cloud platform as claimed in claim 1, characterized in that: include: Step S1, performing AI intelligent calculation by an AI intelligent calculation unit; Step S2, storing the data generated by the AI ​​intelligent computing unit as intelligent computing data through the edge storage unit; Step S3, backing up the intelligent computing data generated by each edge computing module in real time through the time-effectiveness backup module; Step S4, planning and managing the real-time backup data through the time-effective backup module; Step S5, adjusting the real-time backup planning and management process according to the external abnormal access situation through the time-effective backup module; Step S6, optimizing the data backup status according to the edge computing module status through the time backup module; Step S7, setting an upload key through the time-based backup module to encrypt and upload the real-time backup data as the upload data, and adjusting the upload data; Step S8, performing storage determination on the uploaded data according to the upload key by the storage verification unit; Step S9: storing the uploaded data that passes the storage judgment in the decryption storage unit.

Images