Method and system for transmitting disaster recovery backup capability of emergency recovery unit

Abstract

The invention relates to the field of data storage and backup, and discloses a method and a system for transmitting disaster recovery backup capability of an emergency recovery unit. The method comprises the steps of establishing a quantum encryption transmission channel, monitoring a network index in real time and predicting transmission quality to start preventive optimization, calculating a link quality score to carry out link switching, verifying integrity and consistency of received data and executing retransmission, carrying out multi-mode data backup, and switching to an emergency recovery unit when a main data center fails. The system comprises a quantum encryption transmission module, a transmission monitoring and switching module, a data verification and retransmission module, a multi-mode backup module and an emergency switching and synchronization module. Through the technical scheme, the problems of data transmission security and high efficiency can be effectively solved, the disaster recovery backup capability of the emergency recovery unit is improved, the service continuity is ensured, and the method and the device are suitable for scenes with high requirements on data security and service continuity.

Description

Technical Field

[0001] The present invention relates to the field of data storage and backup, and particularly to a method and system for disaster recovery backup capabilities of a transmission emergency recovery unit. Background Art

[0002] In today's digital age, the dependence of various industries on data is increasing day by day, and the security and continuity of data have become key factors to ensure the normal operation of business. Information systems face many potential threats, such as natural disasters: earthquakes, floods, fires, etc., hardware failures: server damage, storage device failures, software errors: system vulnerabilities, program crashes, human errors: misoperations, malicious sabotage, and network attacks: hacker intrusions, virus infections. The above factors may all lead to data loss and system paralysis.

[0003] To cope with the above risks, disaster recovery backup technology has emerged. Traditional disaster recovery backup methods can, to a certain extent, achieve data protection and system recovery, but with the rapid development of information technology and the continuous expansion of business scale, their limitations have gradually become prominent.

[0004] For example, some traditional disaster recovery backup solutions only focus on data backup while ignoring the security and efficiency during data transmission. In the data transmission link, problems such as network latency, insufficient bandwidth, data loss or tampering may occur, which will seriously affect the emergency recovery unit's acquisition of the latest and accurate data, resulting in the inability to perform data recovery and business takeover in a timely and effective manner during a disaster.

[0005] At the same time, some disaster recovery backup systems lack flexibility and intelligence in the deployment and management of emergency recovery units, and it is difficult to quickly adapt to the disaster recovery requirements in different business scenarios and complex environments. When the main system fails, the emergency recovery unit may not be able to quickly and seamlessly switch and take on the heavy responsibility of business operation, resulting in a long business interruption and a very poor experience for users. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for disaster recovery backup capabilities of a transmission emergency recovery unit to solve the above technical problems.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A method for disaster recovery backup capabilities of a transmission emergency recovery unit includes:

[0009] S1. Establish a quantum encryption transmission channel between the main data center and the emergency recovery unit, encrypt data using the quantum key distribution protocol, and dynamically adjust the transmission rate based on the real-time monitoring results of network bandwidth;

[0010] S2. Real-time collect the index data of transmission delay, packet loss rate, and bandwidth utilization rate through the monitoring nodes deployed on the transmission path, use the transmission quality model to predict the transmission quality within a future set time period to obtain a predicted value, and initiate preventive optimization measures when the predicted value exceeds the preset threshold;

[0011] S3. Synchronously calculate the real-time link quality score S corc , when the link quality score S corc < the score threshold S th for a continuous set duration Δt, or when the link quality score difference ΔS corc > the reference link quality score difference ΔS0, perform a forced network link switch, and the real-time link quality score S corc is calculated through a multi-factor dynamic model;

[0012] S4. After receiving the data, the emergency recovery unit first verifies the data integrity through the SHA-3 hash algorithm, and then compares the data header characteristic value with the metadata to achieve consistency verification. When the verification fails, it triggers an automatic retransmission mechanism;

[0013] S5. Perform multi-mode backup on the data that passes the verification;

[0014] S6. When it is detected that the main data center fails, switch to the emergency recovery unit within the set time limit.

[0015] As a further technical solution, the transmission quality model is a pre-trained long short-term memory neural network LSTM model; the preventive optimization measures include: adjusting the data transmission priority, enabling a compression algorithm to reduce the transmission volume, and preheating the standby link in advance.

[0016] As a further technical solution, the expression of the multi-factor dynamic model is:

[0017]

[0018] where μ1, μ2, μ3, μ4 are the weight coefficients corresponding to transmission delay, packet loss rate, bandwidth utilization rate, and continuous stable connection time respectively, determined based on historical data analysis, Delay c is the value after normalizing the transmission delay, PLR c is the value after normalizing the packet loss rate, BU c is the value after normalizing the bandwidth utilization rate, C c is the continuous stable connection duration, and C0 is the reference value of the continuous stable connection duration.

[0019] As a further technical solution, the network link switch in step S3 includes three-level redundant links:

[0020] The primary link is a fiber optic dedicated line;

[0021] The secondary link adopts SDN-controlled dynamic multipath transmission;

[0022] The tertiary link is a 5GNR-U carrier aggregation channel;

[0023] When handover is executed, select the standby link with the highest link quality score; if there are standby links with the same link quality score, preferentially select the link with the largest difference in type from the primary link; if the standby link with the highest link quality score is unavailable, sequentially attempt other standby links in descending order of link quality score.

[0024] As a further technical solution, the automatic retransmission mechanism in step S4 includes:

[0025] Set two-level retransmission queues: real-time service data enters the priority queue, and the retransmission interval ≤ 200 ms;

[0026] Non-real-time service data enters the regular queue, and the retransmission frequency is dynamically adjusted based on the link quality score S corc value. When S corc > the preset threshold S0, use the exponential backoff algorithm; when S corc ≤ the preset threshold S0, enable parallel multi-channel retransmission based on the three-level redundant link, allocate the retransmitted data to the primary link, secondary link, and / or tertiary link for simultaneous transmission according to the link scores of each link, and mark the retransmitted data packets with timestamps and sequence numbers to prevent duplicate reception.

[0027] As a further technical solution, the multi-mode backup in S5 includes: a combination of full backup, incremental backup, and differential backup. Among them, the full backup is performed once a week, the incremental backup is performed once a day, and the differential backup is performed once every three days;

[0028] At the same time, store the backup data in multiple different types of storage media, including disk arrays, tape libraries, and cloud storage.

[0029] As a further technical solution, the process of comparing the data header feature value with the metadata to achieve consistency verification in step S4 is as follows:

[0030] S41. Extract the data header feature value from the received data. The data header feature value includes the data type identifier, data generation time, data source identifier, data version number, and data length;

[0031] S42. Retrieve the metadata related to the received data from the local metadata repository. The metadata includes the expected type of data, generation time range, source identifier, version requirement, and data length standard;

[0032] S43. Compare the extracted data header feature values with the corresponding metadata item by item. When all the data header feature values match the metadata or are within a pre-set error range, it is confirmed that the data is consistent;

[0033] S44. If at least one mismatch is found between the data header feature values and the metadata or it exceeds the pre-set error range during the comparison process, it is confirmed that the verification fails.

[0034] As a further technical solution, the transmission quality model predicts the transmission delay, packet loss rate, and bandwidth utilization rate within a future set time period;

[0035] S21. When the predicted transmission delay exceeds the preset threshold of the transmission delay, start dynamic bandwidth adjustment, including:

[0036] Prioritize the allocation of real-time service traffic to the reserved bandwidth channel;

[0037] Increase the compression ratio of non-critical data to more than 60%;

[0038] Automatically degrade the resolution of video streaming media;

[0039] S22. When the predicted packet loss rate exceeds the preset threshold of the packet loss rate, perform multi-path traffic diversion: direct 20% - 40% of the current traffic to the backup link;

[0040] S23. When the predicted bandwidth utilization rate continues to decline, trigger a pre-check of the link quality, including:

[0041] Send probe data packets to the backup link to measure the round-trip delay;

[0042] Preheat the backup link transmission channel to 20% of the load capacity.

[0043] As a further technical solution, the process of the predicted continuous decline of the bandwidth utilization rate is as follows:

[0044] S231. Dynamic baseline calculation:

[0045] Through the formula: Calculate the adaptive baseline value B base,t , is the average value of the bandwidth utilization rate at the current moment t within the past 24 hours, and σ 7d,t is the standard deviation of the bandwidth utilization rate at the same moment t within the past 7 days;

[0046] S32. Downward trend detection:

[0047] Fit the slope k of the bandwidth data for the most recent N sampling points. When k < the preset slope threshold k0, it is initially determined that it is continuously declining;

[0048] S33. Composite verification condition:

[0049] If the current bandwidth utilization rate B t < B base,t and k < k0, then the final predicted bandwidth utilization rate continues to decline.

[0050] A system for disaster recovery and backup of the transmission emergency recovery unit, characterized by comprising:

[0051] A quantum encryption transmission module, used to establish a quantum encryption transmission channel, perform data encryption, and dynamically adjust the transmission rate based on the network bandwidth;

[0052] A transmission monitoring and switching module, including monitoring nodes, a transmission quality model, a link quality scoring model, and a three - level redundant link, used to collect transmission data, predict transmission quality, and perform link switching;

[0053] A data verification and re - transmission module, used for data integrity and consistency verification, and triggering an automatic re - transmission mechanism;

[0054] A multi - mode backup module, used to perform full - volume backup, incremental backup, differential backup, and store the backup data on different media;

[0055] An emergency switching and synchronization module, used to complete service switching in case of a failure of the primary data center and achieve data synchronization.

[0056] The beneficial effects of the present invention:

[0057] (1) By adopting a quantum encryption transmission channel and a quantum key distribution protocol, the risk of eavesdropping and leakage during data transmission is theoretically eliminated, ensuring data security; at the same time, based on real - time monitoring of the network bandwidth, the transmission rate is dynamically adjusted, and preventive optimization measures are started in combination with the prediction of transmission quality, such as adjusting the data transmission priority, enabling a compression algorithm, pre - heating the standby link, etc., effectively reducing the transmission delay and packet loss rate, and ensuring stable data transmission; in the financial data transmission scenario, it can prevent the leakage of transaction information and ensure the fast and accurate transmission of transaction instructions;

[0058] (2) A multi - factor dynamic model calculates the link quality score, and a scientific forced network link switching is achieved by combining the score threshold and the link score difference; the three - level redundant link design, combined with a reasonable standby link selection strategy, can quickly switch when the primary link fails or has poor quality, ensuring uninterrupted data transmission; the emergency recovery unit can quickly take over the service in case of a failure of the primary data center, and the multi - mode backup strategy combined with multiple storage media ensures the comprehensiveness and reliability of data backup, greatly shortening the service interruption time, improving the overall fault tolerance and recovery ability of the system, reducing the economic losses caused by failures, and ensuring data security and service continuity;

[0059] (3) The SHA-3 hashing algorithm verifies data integrity, and the consistency check between the data header feature value and the metadata ensures the accuracy of key information such as the data source and type. When the verification fails, the automatic retransmission mechanism corrects errors in a timely manner; through multiple guarantee mechanisms, the data is ensured to be transmitted to the emergency recovery unit completely and accurately, providing a reliable basis for subsequent data processing and service recovery. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The present invention will be further described below in conjunction with the accompanying drawings.

[0061] Figure 1 It is a method step diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0062] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

[0063] Please refer to Figure 1 As shown, the present invention is a method for transmitting the disaster tolerance and backup capabilities of an emergency recovery unit, including:

[0064] S1. Establish a quantum encryption transmission channel between the main data center and the emergency recovery unit, encrypt the data using the quantum key distribution protocol, and dynamically adjust the transmission rate based on the real-time monitoring results of the network bandwidth; the dynamic adjustment of the transmission rate based on the real-time monitoring results of the network bandwidth can be achieved based on a deep learning algorithm or a proportional-integral-derivative (PID) control algorithm. Among them, the principle of the deep learning algorithm is to use the powerful feature extraction and prediction capabilities of the deep learning model to learn and train a large amount of historical network bandwidth data, and establish a mapping relationship between the network bandwidth and the transmission rate. For example, the long short-term memory network (LSTM) can process time series data and capture the changing rules of the network bandwidth over time; the convolutional neural network (CNN) can extract and analyze the spatial features in the network traffic data; by continuously optimizing the model parameters, it can accurately predict the optimal transmission rate under different network states. The above two algorithms are both existing technologies and will not be introduced in detail;

[0065] S2. Real-time collect the index data of transmission delay, packet loss rate, and bandwidth utilization rate through the monitoring nodes deployed on the transmission path, use the transmission quality model to predict the transmission quality within a future set time period to obtain a predicted value, and start preventive optimization measures when the predicted value exceeds the preset threshold;

[0066] S3. Synchronously calculate the real-time link quality score S corc, when the link quality score S corc < the score threshold S th for a continuous set duration Δt, or the link quality score difference ΔS between the primary link and the backup link corc > the reference link quality score difference ΔS0, perform a forced network link switch. The real-time link quality score S corc is calculated through a multi-factor dynamic model; the score threshold S th is determined by statistical analysis of historical data according to the service type and network environment;

[0067] S4. After receiving the data, the emergency recovery unit first verifies the data integrity through the SHA-3 hash algorithm, and then compares the data header feature value with the metadata to achieve consistency verification. When the verification fails, trigger the automatic retransmission mechanism;

[0068] S5. Perform multi-mode backup on the data that passes the verification;

[0069] S6. When a main data center failure is detected, switch to the emergency recovery unit within the set time limit.

[0070] The transmission quality model is a pre-trained long short-term memory neural network LSTM model; the preventive optimization measures include: adjusting the data transmission priority, enabling a compression algorithm to reduce the transmission volume, and preheating the backup link in advance.

[0071] In this embodiment, first, by establishing a quantum encryption transmission channel and adopting the quantum key distribution protocol, data encryption is achieved using the principles of quantum mechanics. Any eavesdropping behavior during the key distribution process will change the quantum state, which can be detected and discovered, theoretically ensuring the absolute security of data transmission, eliminating the risk of data leakage, and providing a solid security foundation for disaster recovery backup; based on real-time monitoring of network bandwidth, the transmission rate is dynamically adjusted to avoid the problem of low transmission efficiency caused by network congestion or bandwidth idleness; secondly, combining transmission quality prediction with preventive optimization measures, intervention is carried out in advance before the network quality deteriorates, such as adjusting the data transmission priority to ensure critical services, enabling compression algorithms to reduce the amount of data transmitted, and preheating standby links in advance, effectively reducing transmission latency and packet loss rate, and maintaining stable and efficient data transmission; thirdly, the SHA-3 hash algorithm is used to verify the data integrity. Any data tampering will cause the hash value to change, ensuring that the received data is the same as the sent data; the consistency check of the data header characteristic value and metadata further confirms that key information such as the data source and type is accurate. When the check fails, the automatic retransmission mechanism corrects the error in time to ensure that the data is transmitted to the emergency recovery unit completely and accurately; subsequently, a multi-factor dynamic model calculates the link quality score, and based on the score threshold and link score difference judgment, scientific and reasonable forced switching of network links is achieved; the three-level redundant link and parallel multi-channel transmission strategy can quickly switch to a high-quality standby link when the main link fails or has poor quality, making full use of the bandwidth resources of multiple links, improving the system's fault tolerance and transmission performance; finally, through a multi-mode backup strategy (full backup, incremental backup, differential backup) combined with various storage media, the comprehensiveness and reliability of data backup are ensured; when the main data center fails, it quickly switches to the emergency recovery unit, and the business system operation is restored within the set time limit, greatly shortening the business interruption time, ensuring that the enterprise's core business is not interrupted, and reducing the economic losses caused by the failure.

[0072] The expression of the multi-factor dynamic model is as follows:

[0073]

[0074] where μ1, μ2, μ3, and μ4 are the weight coefficients corresponding to transmission latency, packet loss rate, bandwidth utilization rate, and continuous stable connection time respectively, determined based on historical data analysis. Delay c is the value after normalization processing of the transmission latency, PLR c is the value after normalization processing of the packet loss rate, BU c is the value after normalization processing of the bandwidth utilization rate, C c is the duration of continuous stable connection, and C0 is the reference value of the duration of continuous stable connection.

[0075] In this example, transmission delay affects the timeliness of data transmission, packet loss rate is related to the integrity of data, bandwidth utilization reflects the resource utilization degree of the link, and continuous stable connection time reflects the stability of the link. Through the multi-factor dynamic model it is able to comprehensively consider four factors: transmission delay, packet loss rate, bandwidth utilization, and continuous stable connection time, and accurately evaluate the link quality. Taking data transmission during major e-commerce promotions as an example, a large amount of transaction data places extremely high requirements on the network link. The model determines the weights of each factor based on historical data and calculates the link quality score in real time. If the packet loss rate of a certain link increases under high concurrency, the model can quickly reduce its score. According to the score judgment, when the link quality score is lower than the threshold or the score difference from the backup link reaches a certain level, the transmission strategy is adjusted in a timely manner, and the data is switched to a more stable link, effectively avoiding the loss or delay of transaction data and ensuring the smooth progress of transactions.

[0076] The network link switching in step S3 includes three levels of redundant links:

[0077] The first-level link is a fiber-optic dedicated line;

[0078] The second-level link adopts SDN-controlled dynamic multi-path transmission;

[0079] The third-level link is a 5GNR-U carrier aggregation channel;

[0080] When the switching is executed, the backup link with the highest link quality score is selected; if there are backup links with the same link quality score, the link with the largest difference in type from the main link is preferred; if the backup link with the highest link quality score is unavailable, other backup links are tried in order from high to low according to the link quality score.

[0081] The three-level redundant links set in this embodiment, namely the fiber-optic dedicated line, SDN-controlled dynamic multi-path transmission, and 5GNR-U carrier aggregation channel, build multiple defenses for data transmission. When the network suffers a sudden failure, such as the fiber being accidentally cut off, the abnormality of the main link can be immediately detected. At this time, through the link quality score, it is quickly switched to the backup link. If the fiber-optic dedicated line is unavailable, other links with the highest score, such as the SDN-controlled dynamic multi-path transmission link, will be selected. The above redundant design greatly improves the fault tolerance of the network and ensures the continuity of data transmission.

[0082] Through the above technical solution, intelligent link switching is achieved; in a complex network environment, the performance of different links changes over time; for example, within a certain period, the 5GNR-U carrier aggregation channel may experience performance degradation and a decrease in link quality score due to surrounding signal interference; according to the score change, data is automatically switched to a fiber-optic dedicated line or an SDN-controlled dynamic multi-path transmission link with a higher score; at the same time, if there are standby links with the same score, a link with a large difference in type from the main link is preferentially selected to avoid affecting transmission due to potential problems that may be common to links of the same type; through the above intelligent switching mechanism, the advantages of each link are fully utilized, effectively improving data transmission efficiency, reducing transmission latency, and enhancing the user experience.

[0083] The automatic retransmission mechanism in step S4 includes:

[0084] Set up two-level retransmission queues: real-time service data enters the priority queue, and the retransmission interval ≤ 200 ms;

[0085] Non-real-time service data enters the regular queue, and the retransmission frequency is dynamically adjusted based on the link quality score S corc value. When S corc > the preset threshold S0, an exponential backoff algorithm is adopted. For example, the initial backoff time is set to 100 ms, the backoff factor is 2, and after each retransmission failure, the backoff time increases exponentially, with a maximum backoff time not exceeding 5 s; when S corc ≤ the preset threshold S0, parallel multi-channel retransmission based on a three-level redundant link is enabled. The retransmitted data is allocated to the primary link, secondary link, and / or tertiary link for simultaneous transmission according to the scores of each link, and the retransmitted data packets are marked with timestamps and sequence numbers to prevent duplicate reception.

[0086] The preset threshold is a pre-configured critical value of the link quality score, which is specifically determined according to the service type and network environment; the process of starting multi-channel transmission is as follows:

[0087] Preferentially select the standby link with the highest link quality score as the main transmission channel, and according to the remaining link quality scores and system bandwidth requirements, enable one or more other standby links at the same time. After dividing the data according to the traffic allocation strategy, it is transmitted through the parallel three-level redundant links; the traffic allocation strategy is determined according to the ratio of the link quality score of each link to the total link quality score, that is, the link quality score of each level / (link quality score of the primary link + link quality score of the secondary link + link quality score of the tertiary link); for example, if the proportion of the link quality score of the primary link is 40%, the proportion of the link quality score of the secondary link is 30%, and the proportion of the link quality score of the tertiary link is 30%, then 40% of the data is allocated to the primary link, 30% of the data is allocated to the secondary link, and 30% of the data is allocated to the tertiary link.

[0088] The multi-mode backup in S5 includes: a combination of full backup, incremental backup, and differential backup. Among them, the full backup is performed once a week, the incremental backup is performed once a day, and the differential backup is performed once every three days;

[0089] Meanwhile, the backup data is stored in multiple different types of storage media, including disk arrays, tape libraries, and cloud storage.

[0090] In this embodiment, by combining full backup, incremental backup, and differential backup, the comprehensiveness and recoverability of the data are guaranteed; the full backup once a week provides a complete data copy, providing a basis for long-term data recovery; the daily incremental backup and the differential backup every three days can quickly recover recent data changes; storing the backup data in multiple media enhances data security and avoids data loss caused by the failure of a single storage medium. In case of data loss or damage, the business data can be quickly recovered according to the backup strategy and storage medium, reducing data loss.

[0091] The process of comparing the data header feature values with the metadata to achieve consistency verification in step S4 is as follows:

[0092] S41. Extract the data header feature values from the received data. The data header feature values include data type identifier, data generation time, data source identifier, data version number, and data length;

[0093] S42. Retrieve the metadata related to the received data from the local metadata repository. The metadata includes the expected type of data, generation time range, source identifier, version requirement, and data length standard;

[0094] S43. Compare the extracted data header feature values with the corresponding metadata item by item. When all the data header feature values match the metadata or are within a preset error range, it is confirmed that the data is consistent; for the data type identifier, directly judge whether the two are exactly the same; for the data generation time, check whether the generation time of the received data is within the reasonable time range specified in the metadata; for the data source identifier, confirm whether it is consistent with the expected source in the metadata; for the data version number, verify whether it meets the version required by the system; for the data length, judge whether it is within the length tolerance range set in the metadata.

[0095] S44. If at least one mismatch is found between the data header feature value and the metadata or the value exceeds the pre-set error range during the comparison process, it is confirmed that the verification fails. Immediately trigger the exception handling mechanism, record detailed error information, including the specific feature items of the inconsistency, the differences between the actual value and the expected value, etc., and according to the preset rules, send a data retransmission request to the master data center. At the same time, isolate and store the abnormal data to prevent it from entering the subsequent data processing process. After the verification passes, the data will be allowed to enter the subsequent data storage and processing links.

[0096] Data consistency verification is a key link to ensure data quality. During data transmission, data errors may occur due to factors such as network interference. In this embodiment, a reliable verification method is provided, and the automatic retransmission mechanism is an effective remedy for data errors, ensuring data integrity, improving the accuracy and reliability of data transmission, and ensuring the availability of disaster recovery backup data; by comparing the data header feature value with the metadata, the consistency of the data can be accurately verified, and errors or tampering in the data during transmission can be detected in a timely manner; when the verification fails, the automatic retransmission mechanism is triggered to ensure that the data reaching the emergency recovery unit is accurate and complete, effectively preventing incorrect data from entering the subsequent processing process, ensuring the reliability of the data, and providing an accurate basis for further analysis and use of the data.

[0097] The transmission quality model predicts the transmission delay, packet loss rate, and bandwidth utilization rate within a future set time period;

[0098] S21. When the predicted transmission delay exceeds the preset threshold of the transmission delay, start dynamic bandwidth adjustment, including:

[0099] Allocate real-time service traffic to the reserved bandwidth channel preferentially;

[0100] Increase the compression ratio of non-critical data to more than 60%;

[0101] Automatically degrade the resolution of video streaming media;

[0102] S22. When the predicted packet loss rate exceeds the preset threshold of the packet loss rate, perform multi-path traffic diversion: direct 20%-40% of the current traffic to the backup link;

[0103] S23. When the predicted bandwidth utilization rate continues to decline, trigger a pre-check of the link quality, including:

[0104] Send probe data packets to the backup link to measure the round-trip delay;

[0105] Preheat the backup link transmission channel to 20% of the load capacity.

[0106] Transmission quality prediction and preventive optimization measures can detect potential network problems in advance and intervene in time; take corresponding measures before the network condition deteriorates to avoid data transmission interruption or delay and ensure the normal operation of the business; the preventive strategy provided in this embodiment improves the adaptability and stability of the system, reduces the impact of network problems on the business, and improves the overall system performance; according to the prediction results of the transmission quality model, take different optimization measures in a targeted manner, and when the transmission delay is predicted to exceed the threshold, the real-time business is guaranteed through dynamic bandwidth adjustment; when the packet loss rate is predicted to exceed the threshold, multi-path traffic diversion reduces the risk of packet loss; when the bandwidth utilization rate is predicted to continue to decrease, the link quality pre-inspection prepares the backup link in advance. The above measures improve the stability of network transmission, reduce data transmission anomalies, and improve the business experience.

[0107] The process of predicting the continuous decrease in bandwidth utilization is as follows:

[0108] S231, Dynamic baseline calculation:

[0109] By formula: Calculate the adaptive baseline value B base,t , is the average bandwidth utilization rate at the current time t in the past 24 hours, σ 7d,t is the standard deviation of bandwidth utilization at the same time t in the past 7 days;

[0110] S32, downtrend detection:

[0111] The bandwidth data of the most recent N sampling points are fitted with a slope k. When k < the preset slope threshold k0, it is preliminarily determined to be continuously decreasing.

[0112] S33. Composite verification conditions:

[0113] If the current bandwidth utilization is B t <B base,t And k<k0, then the final predicted bandwidth utilization rate continues to decrease.

[0114] Accurately predicting changes in bandwidth utilization is an important means to ensure network performance; dynamic baseline calculation combined with historical data is more in line with actual network usage patterns; downward trend detection and compound verification conditions improve prediction accuracy; the above prediction mechanism provides support for the system to adjust the transmission strategy in advance, ensure reasonable allocation of network resources, improve network resource utilization, and ensure stable operation of services; in this embodiment, through dynamic baseline calculation, downward trend detection and compound verification conditions, it is possible to accurately predict the continuous decline in bandwidth utilization; timely discover bandwidth problems, provide accurate basis for triggering optimization measures such as link quality pre-check, avoid slow or interrupted data transmission due to insufficient bandwidth, and ensure the stability and reliability of network transmission.

[0115] A system for transmitting the disaster recovery and backup capabilities of an emergency recovery unit, characterized in that it includes:

[0116] A quantum encryption transmission module, used to establish a quantum encryption transmission channel, perform data encryption and dynamic adjustment of the transmission rate based on network bandwidth;

[0117] A transmission monitoring and switching module, including monitoring nodes, a transmission quality model, a link quality scoring model, and a three-level redundant link, used to collect transmission data, predict transmission quality, and perform link switching;

[0118] A data verification and retransmission module, used for data integrity and consistency verification, and triggering an automatic retransmission mechanism;

[0119] A multi-mode backup module, used to perform full backup, incremental backup, differential backup, and store the backup data on different media;

[0120] An emergency switching and synchronization module, used to complete service switching in case of a failure in the primary data center and achieve data synchronization.

[0121] It should be noted that: the calculation formulas and each parameter participating in the operation in the present invention have been dimensionless processed, and the process of dimensionless processing is well known in the industry and will not be described here.

[0122] The above has described a specific embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. Any equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A method for transmitting the disaster recovery and backup capabilities of an emergency recovery unit, characterized in that Including: S1. Establish a quantum encryption transmission channel between the main data center and the emergency recovery unit, perform data encryption using the quantum key distribution protocol, and dynamically adjust the transmission rate based on the real-time monitoring results of the network bandwidth; S2. Real-time collect the index data of transmission delay, packet loss rate, and bandwidth utilization rate through the monitoring nodes deployed on the transmission path, use the transmission quality model to predict the transmission quality within a future set period to obtain a predicted value, and start preventive optimization measures when the predicted value exceeds the preset threshold; S3. Synchronously calculate the real-time link quality score S corc , when the link quality score S corc < the score threshold S th for a continuously set duration Δt, or the link quality score difference ΔS between the primary link and the backup link corc > the reference link quality score difference ΔS0, perform a forced network link switch. The real-time link quality score S corc is calculated through a multi-factor dynamic model; S4. After receiving the data, the emergency recovery unit first verifies the data integrity through the SHA-3 hash algorithm, and then compares the data header characteristic values with the metadata to achieve consistency verification. When the verification fails, an automatic retransmission mechanism is triggered; S5. Perform multi-mode backup on the data that passes the verification; S6. When a main data center failure is detected, switch to the emergency recovery unit within the set time limit.

2. The method for the disaster tolerance and backup capability of the transmission emergency recovery unit according to claim 1, wherein The transmission quality model is a pre-trained long short-term memory neural network (LSTM) model; The preventive optimization measures include: adjusting the data transmission priority, enabling a compression algorithm to reduce the transmission volume, and preheating the standby link in advance.

3. The method for disaster recovery backup capability of the transmission emergency recovery unit according to claim 2, characterized in that, The expression of the multi-factor dynamic model is: Among them, μ1, μ2, μ3, and μ4 are the weight coefficients corresponding to transmission delay, packet loss rate, bandwidth utilization rate, and duration of continuous stable connection, respectively, which are determined based on historical data analysis. Delay c is the value after normalization of the transmission delay, and PLR c is the value after normalization of the packet loss rate, and BU c is the value after normalization of the bandwidth utilization rate, and C c is the duration of continuous stable connection, and C0 is the reference value of the duration of continuous stable connection.

4. The method for disaster recovery backup capability of the transmission emergency recovery unit according to claim 3, characterized in that, The network link switching in step S3 includes three-level redundant links: The first-level link is a fiber-optic dedicated line; The second-level link uses software-defined networking (SDN)-controlled dynamic multi-path transmission; The third-level link is a 5G NR-U carrier aggregation channel; When switching is executed, select the standby link with the highest link quality score; if there are standby links with the same link quality score, preferentially select the link with the largest difference in type from the main link; if the standby link with the highest link quality score is unavailable, try other standby links in order from high to low link quality score.

5. The method for disaster recovery backup capability of the transmission emergency recovery unit according to claim 4, characterized in that, The automatic retransmission mechanism in step S4 includes: Set two-level retransmission queues: Real-time service data enters the priority queue, and the retransmission interval ≤ 200 ms; Non-real-time service data enters the regular queue and dynamically adjusts the retransmission frequency based on the link quality score S corc value. When S corc > the preset threshold S0, the exponential backoff algorithm is adopted; when S corc ≤ the preset threshold S0, parallel multi-channel retransmission based on a three-level redundant link is enabled. The retransmitted data is allocated to the primary link, secondary link, and / or tertiary link for simultaneous transmission according to the link scores of each link, and the retransmitted data packets are marked with timestamps and sequence numbers to prevent duplicate reception.

6. The method for disaster recovery backup capability of the transmission emergency recovery unit according to claim 1, characterized in that, The multi-mode backup in S5 includes: combining full backup, incremental backup, and differential backup. Among them, full backup is performed once a week, incremental backup is performed once a day, and differential backup is performed once every three days; At the same time, store the backup data in multiple different types of storage media, including disk arrays, tape libraries, and cloud storage.

7. The method for disaster recovery backup capability of the transmission emergency recovery unit according to claim 1, wherein The process of comparing the data header characteristic values with the metadata to achieve consistency verification in step S4 is: S41. Extract the data header characteristic values from the received data. The data header characteristic values include data type identifier, data generation time, data source identifier, data version number, and data length; S42. Retrieve the metadata related to the received data from the local metadata repository. The metadata includes the expected type of data, generation time range, source identifier, version requirements, and data length standard; S43. Compare the extracted data header characteristic values with the corresponding metadata item by item. When all data header characteristic values match the metadata or are within a preset error range, it is confirmed that the data is consistent; S44. If at least one mismatch or exceeding the preset error range is found between the data header characteristic values and the metadata during the comparison process, it is confirmed that the verification fails.

8. The method for disaster recovery backup capability of a transmission emergency recovery unit according to claim 1, characterized in that, The transmission quality model predicts the transmission delay, packet loss rate, and bandwidth utilization within a future set time period; S21. When it is predicted that the transmission delay exceeds the preset threshold of the transmission delay, dynamic bandwidth adjustment is initiated, including: Real-time service traffic is preferentially allocated to the reserved bandwidth channel; The compression ratio of non-critical data is increased to more than 60%; The resolution of video streaming media is automatically degraded; S22. When the predicted packet loss rate exceeds the preset threshold of the packet loss rate, multi-path traffic shunting is performed: 20%-40% of the current traffic is guided to the backup link; S23. When it is predicted that the bandwidth utilization continues to decline, link quality pre-inspection is triggered, including: Sending probe data packets to the backup link to measure the round-trip delay; Preheating the backup link transmission channel to 20% of the load capacity.

9. The method for disaster recovery backup capability of a transmission emergency recovery unit according to claim 8, characterized in that, The process of predicting the continuous decline of bandwidth utilization is as follows: S231. Dynamic baseline calculation: Through the formula: the adaptive baseline value B is calculated base,t , where is the average bandwidth utilization rate of the current time t within the past 24 hours, and σ 7d,t is the standard deviation of the bandwidth utilization rate at the same time t within the past 7 days; S32. Decline trend detection: The fitting slope k of the bandwidth data of the last N sampling points is calculated. When k < the preset slope threshold k0, it is preliminarily determined that the decline is continuous; S33. Composite verification conditions: If the current bandwidth utilization rate is B t <B base,t and k < k0, then the final predicted bandwidth utilization rate continues to decline.

10. A system for implementing the disaster recovery and backup capability of the transmission emergency recovery unit described in claim 3, characterized in that, Including: The quantum encryption transmission module is used to establish a quantum encryption transmission channel, perform data encryption, and dynamically adjust the transmission rate based on the network bandwidth; The transmission monitoring and switching module includes monitoring nodes, a transmission quality model, a link quality scoring model, and a three-level redundant link, and is used to collect transmission data, predict transmission quality, and perform link switching; The data verification and retransmission module is used for data integrity and consistency verification, and to trigger the automatic retransmission mechanism; The multi-mode backup module is used to perform full backup, incremental backup, differential backup, and store the backup data on different media; The emergency switching and synchronization module is used to complete service switching and achieve data synchronization when the main data center fails.

Images