A computing power data operation and maintenance processing method and system based on AI intelligence

Abstract

The present application relates to a method and system for processing computing power data operation and maintenance based on AI intelligence, the method comprising: obtaining comprehensive performance information of all backup devices within a preset distance range, and analyzing and determining the capacity values ​​of all backup devices within a preset distance range according to a preset capacity value quantitative evaluation method of the comprehensive performance information, the comprehensive performance information including storage performance information, processing performance information, transmission performance information, stability performance information and security performance information; receiving and backing up relatively high-priority computing power data from high to low according to the backup capacity value of the backup device, until the remaining backup computing power data of the backup device is cleared or the computing power data to be backed up has been completely backed up. The present application has the following effects: it provides an efficient, safe and reliable solution for computing power data operation and maintenance, helps to ensure the integrity and recoverability of data, and reduces the risk of data loss.

Description

Technical Field

[0001] The present invention relates to the field of data storage and backup technology, and in particular to a computing power data operation and maintenance processing method and system based on AI intelligence. Background Art

[0002] With the rapid development of information technology, the importance of computing power data in various fields has become increasingly prominent. Whether it is the key business operations of enterprises, data analysis of scientific research, or digital services in daily life, they are inseparable from the support of a large amount of computing power data. Effective computing power data operation and maintenance processing is essential to ensure the security, availability and integrity of data. Through intelligent backup trigger mechanisms, accurate priority determination, comprehensive backup device capability assessment, and efficient data backup strategies, the risk of data loss can be minimized, business continuity and stability can be guaranteed, reliable data support can be provided for decision-making, and overall operational efficiency and competitiveness can be improved.

[0003] In the existing technology, the operation and maintenance of computing power data usually adopts some relatively traditional methods. Generally, a fixed backup time interval is set, such as daily or weekly backup. The distinction between data categories is relatively simple, and may only be roughly divided into important data and ordinary data, and the determination of backup priority is not precise enough.

[0004] On the one hand, the simple classification of important data and ordinary data cannot accurately reflect the differences in value and importance of different types of computing data. For example, critical business data and high-value data may require higher backup priority and stricter security protection, but this may not be fully reflected in existing technologies. On the other hand, the selection of backup devices is one-sided, and only storage capacity is used as the main indicator to select backup devices, which easily ignores other important performance factors. For example, low processing performance may lead to slow backup speed, poor transmission performance may affect the efficiency of data transmission, insufficient stability performance may increase the risk of data loss, and weak security performance may expose data to the risk of being attacked or leaked. Summary of the invention

[0005] In order to provide an efficient, secure and reliable solution for computing power data operation and maintenance, ensure the integrity and recoverability of data, and reduce the risk of data loss, this application provides a computing power data operation and maintenance processing method and system based on AI intelligence.

[0006] In the first aspect, the present application provides a computing power data operation and maintenance processing method based on AI intelligence, which adopts the following technical solutions:

[0007] A computing power data operation and maintenance processing method based on AI intelligence, comprising:

[0008] Obtain backup trigger information for computing power data;

[0009] Obtain computing power data categories and the amount of remaining backup computing power data of all backup devices within a preset distance. The computing power data categories include key business data, system configuration data, high-value data, log data, temporary data, user preference data, and statistical data.

[0010] According to the mapping relationship between computing power data categories and backup priority levels, analyze and determine the backup priority levels of different computing power data categories;

[0011] Obtaining comprehensive performance information of all backup devices within a preset distance range, and analyzing and determining the capacity values ​​of all backup devices within a preset distance range according to a preset capacity value quantitative evaluation method of the comprehensive performance information, wherein the comprehensive performance information includes storage performance information, processing performance information, transmission performance information, stability performance information, and security performance information;

[0012] According to the backup capacity value of the backup device, high-priority computing power data is received and backed up from high to low until the remaining backup computing power data of the backup device is cleared or the computing power data to be backed up has been completely backed up.

[0013] By adopting the above technical solution, first of all, the priority levels of different computing power data categories are clarified to ensure that important data such as key business are backed up first, thereby improving data security. Secondly, considering the comprehensive performance of the backup device within the preset distance and the remaining computing power, the backup tasks are reasonably allocated to optimize resource utilization. Furthermore, high-priority data is received from high to low according to the backup capacity value to improve backup efficiency. At the same time, the backup strategy can be flexibly adjusted to meet different business needs. On the whole, this method provides an efficient, safe and reliable solution for computing power data operation and maintenance, which helps to ensure the integrity and recoverability of data and reduce the risk of data loss.

[0014] Optionally, the backup trigger information for obtaining computing power data includes:

[0015] Analyze whether the backup trigger conditions of computing power data are met. The backup trigger conditions include time trigger conditions, data volume trigger conditions, and event trigger conditions.

[0016] If yes, continue with the next steps;

[0017] If not, stop the subsequent steps.

[0018] By adopting the above technical solution, by analyzing the trigger conditions such as time, data volume and events, it can accurately determine whether to start the backup. When the conditions are met, the subsequent steps are carried out to ensure that the backup is carried out in time; if they are not met, it is stopped to avoid unnecessary waste of resources and improve the efficiency and pertinence of the backup.

[0019] Optionally, obtaining the preset distance range includes:

[0020] Obtain the backup trigger conditions associated with the backup trigger information of this computing power data;

[0021] According to the correspondence between the backup trigger conditions, computing power data categories and the preset distance range, the preset distance range is analyzed and determined.

[0022] By adopting the above technical solution, on the one hand, the range is determined according to the backup trigger conditions, so that it can be flexibly adjusted in different situations. For example, when the time is urgent, you can choose a close-range device for fast backup; when the data volume is triggered, the distance can be reasonably planned according to the scale. On the other hand, combined with the computing power data category, it ensures that important data is backed up at a more appropriate distance to improve security. For example, critical business data can be backed up on a close and highly reliable device. This not only improves the backup efficiency, but also meets the needs of different scenarios, realizes the reasonable allocation of resources, provides effective protection for the safe backup of computing power data, and improves the overall data operation and maintenance level.

[0023] Optionally, according to the mapping relationship between the computing power data category and the backup priority level, the backup priority levels of different computing power data categories are analyzed and determined, including:

[0024] Analyze whether the computing power data obtained is at a critical business node period;

[0025] If yes, then according to the mapping relationship between the computing power data category at the business critical node period, the backup trigger condition and the backup priority level, analyze and determine the backup priority level of the computing power data category at the business critical node period;

[0026] If not, the backup priority levels of different computing power data categories are analyzed and determined based on the mapping relationship between computing power data categories, backup trigger conditions and backup priority levels.

[0027] By adopting the above technical solution, during non-business critical nodes, the backup priority is determined according to the conventional mapping relationship to ensure that different types of computing power data are reasonably arranged. During business critical nodes, the mapping relationship is adjusted in a targeted manner to increase the backup priority of key data, which can better ensure business continuity and stability. This method flexibly adapts to different business states, avoids waste of resources, and provides stronger protection for important data at critical moments, improves the effectiveness and pertinence of the overall data backup strategy, helps improve the reliability of the system and the ability to respond to emergencies, and provides strong support for the business operations of the enterprise.

[0028] Optionally, according to the mapping relationship between the computing power data category and the backup priority level, the backup priority levels of different computing power data categories are analyzed and determined, including:

[0029] Analyze whether the computing power data obtained is at a critical business node period and / or a stage of important business changes;

[0030] If yes, then according to the mapping relationship between the computing power data categories that are simultaneously in the business critical node period and the business important change stage, the priority level of the computing power data categories that are simultaneously in the business critical node period and the business important change stage, the priority level of the computing power data categories that are simultaneously in the business critical node period, the mapping relationship between the backup trigger conditions and the backup priority level, the priority level of the computing power data categories that are individually in the business critical node period, the analysis and determination of the analysis; according to the mapping relationship between the computing power data categories that are individually in the business important change stage, the backup trigger conditions and the backup priority level, the priority level of the computing power data categories that are individually in the business important change stage are analyzed and determined;

[0031] If not, the backup priority levels of different computing power data categories are analyzed and determined based on the mapping relationship between computing power data categories, backup trigger conditions and backup priority levels.

[0032] By adopting the above technical solutions, the priority levels of computing power data in different situations are determined during business key nodes and important business change stages, which can accurately ensure the backup of key data. Higher priority is given to computing power data that is also in special periods to ensure its security. Priority is determined according to conventional mapping during non-special periods, and resources are allocated reasonably. Consider the backup trigger conditions, further optimize the backup arrangement, and improve backup efficiency and reliability. This method adapts to different business states, enhances data security and recoverability, provides strong support for the stable operation of the business, helps to achieve reasonable backup and management of computing power data, and improves the overall data operation and maintenance level.

[0033] Optionally, analyzing and determining the capability values ​​of all backup devices within a preset distance range includes:

[0034] Collect historical performance data and related environmental data of all backup devices, where historical performance data includes the number of read and write times, error rate, and temperature of the storage medium, and environmental data includes temperature, humidity, and voltage;

[0035] The collected historical performance data of all backup devices and related environmental data are used as input objects, and are input into a storage medium life prediction model and a backup device failure prediction model that have been established and trained in advance, and a quantified value of the backup device failure probability and a quantified value of the remaining life of the storage medium are output;

[0036] Analyze and determine the decline coefficient of the backup device capability value based on the interval range in which the backup device failure probability quantified value falls, the interval range in which the storage medium remaining life quantified value falls, and the decline coefficient of the capability value;

[0037] The effective capacity value of the backup device is calculated according to the decline coefficient of the backup device capacity value and the backup device capacity value.

[0038] By adopting the above technical solutions, by collecting historical performance and environmental data and using the prediction model to derive quantitative values, the status of backup equipment can be accurately evaluated. Determining the decline coefficient and then calculating the effective capacity value can dynamically reflect the actual performance of the equipment. In practical applications, backup tasks can be reasonably allocated to avoid faulty equipment from taking on too many tasks. At the same time, it helps to plan equipment maintenance and replacement in advance and improve the reliability and stability of the backup system. It can better adapt to different equipment states, optimize resource utilization, provide strong guarantees for the safe backup of computing power data, improve the overall data operation and maintenance level, and ensure efficient and reliable data backup under different circumstances.

[0039] Optionally, the method further includes receiving and backing up relatively high-priority computing power data from high to low according to the backup capacity value of the backup device until the remaining backup computing power data of the backup device is cleared or the computing power data to be backed up has been completely backed up in parallel, as follows:

[0040] Obtain the priority of different types of computing power data and the effective capacity value of backup devices;

[0041] Analyze and determine the data compression schemes for different categories of computing power data based on the priorities of different categories of computing power data, the intervals into which the effective capacity values ​​of the backup devices fall, and the corresponding relationship between the data compression schemes;

[0042] Execute data compression schemes for different computing power data.

[0043] By adopting the above technical solutions, the priorities of different types of computing power data and the effective capacity values ​​of backup devices can be obtained, and the compression scheme can be determined in a targeted manner. Determining the scheme based on the corresponding relationship can optimize storage without affecting the backup progress. Executing specific compression schemes for different data improves backup efficiency and saves storage space. Adapting to different data requirements and device capabilities ensures that high-priority data is quickly backed up under reasonable compression. It enhances the flexibility and adaptability of the backup system, improves the overall data operation and maintenance level, provides strong support for the safe storage and efficient management of computing power data, and helps to achieve reliable backup and rapid recovery of data in various situations.

[0044] In the second aspect, the present application provides a computing power data operation and maintenance processing system based on AI intelligence, which adopts the following technical solutions:

[0045] A computing power data operation and maintenance processing system based on AI intelligence, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement the computing power data operation and maintenance processing method based on AI intelligence as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of a computing power data operation and maintenance processing method based on AI intelligence in an embodiment of the present application.

[0047] Figure 2 It is a flowchart of another embodiment of the present application for obtaining backup trigger information of computing power data.

[0048] Figure 3 It is a schematic diagram of the process of obtaining a preset distance range according to another embodiment of the present application.

[0049] Figure 4 Another embodiment of the present invention is a flowchart of analyzing and determining the backup priority levels of different computing power data categories according to the mapping relationship between computing power data categories and backup priority levels. Figure 1 .

[0050] Figure 5 Another embodiment of the present invention is a flowchart of analyzing and determining the backup priority levels of different computing power data categories according to the mapping relationship between computing power data categories and backup priority levels. Figure 2 . DETAILED DESCRIPTION

[0051] The present application is further described in detail below in conjunction with the accompanying drawings.

[0052] Reference Figure 1 , is a computing power data operation and maintenance processing method based on AI intelligence disclosed in this application, including:

[0053] Step S100, obtaining backup trigger information of computing power data.

[0054] Backup trigger information can be generated in a variety of ways. For example:

[0055] Time trigger condition: Set the backup to be automatically triggered at 12:00 every night. At this time, the enterprise's server backs up the various computing power data generated that day to ensure the security and recoverability of the data.

[0056] Data volume trigger condition: When the storage volume of computing power data reaches a certain threshold, the backup is triggered. For example, when the amount of data in the database grows to 1TB, the system automatically starts the backup program and backs up the newly added data to the backup device within the preset distance range.

[0057] Event trigger conditions: Backup is triggered when certain important events occur. For example, when a major system upgrade occurs or when a major change occurs to key business data, backup is triggered immediately to prevent data loss or damage.

[0058] Step S200, obtaining the computing power data category and the remaining backup computing power data amount of all backup devices within a preset distance range.

[0059] Computing power data categories include key business data, system configuration data, high-value data, log data, temporary data, user preference data, and statistical data.

[0060] Critical business data: For example, transaction data in financial transaction systems. This data is vital to the core business of the enterprise. Once lost, it may cause serious economic losses and business interruption.

[0061] System configuration data: including server configuration parameters, network settings, etc. These data are the basis for ensuring the normal operation of the system.

[0062] High-value data: such as a company’s R&D data, intellectual property documents, etc., which have high commercial value.

[0063] Log data: records the system operation log information for troubleshooting and performance analysis.

[0064] Temporary data: For example, temporary cache files generated when users browse the web are relatively unimportant and can be given a lower priority when backing up.

[0065] User preference data: such as the user’s browsing history and preference settings on shopping websites, used for personalized recommendations.

[0066] The remaining backup computing power data of all backup devices within the preset distance range: Assuming that an enterprise has set up multiple backup devices in a data center, the network monitoring system can obtain information such as the remaining storage space and processing capacity of each backup device in real time. For example, the remaining storage space of backup device A is 500GB, and the processing capacity is 100MB of data per second; the remaining storage space of backup device B is 300GB, and the processing capacity is 80MB of data per second.

[0067] Step S300, analyzing and determining the backup priority levels of different computing power data categories based on the mapping relationship between computing power data categories and backup priority levels.

[0068] Specifically, a mapping table can be established to clearly define the backup priorities of different computing power data categories. For example, the backup priority of critical business data can be set to high priority, the backup priority of system configuration data can be set to high priority, the backup priority of high-value data can be set to medium-high priority, the backup priority of log data can be set to medium priority, and the backup priority of temporary data can be set to low priority.

[0069] For example, in an enterprise's production system, when a backup is needed, the system first determines that critical business data and system configuration data have the highest backup priority because these data are essential to the normal operation of the enterprise. Temporary data has the lowest backup priority and can be backed up when resources are sufficient.

[0070] Step S400, obtain the comprehensive performance information of all backup devices within a preset distance range, and analyze and determine the capacity values ​​of all backup devices within the preset distance range according to the preset capacity value quantification evaluation method of the comprehensive performance information, wherein the comprehensive performance information includes storage performance information, processing performance information, transmission performance information, stability performance information and security performance information.

[0071] Storage performance information: including storage capacity, read / write speed, storage media type, etc. For example, backup device A uses a solid-state hard disk with a storage capacity of 2TB and a read / write speed of 500MB per second; backup device B uses a mechanical hard disk with a storage capacity of 4TB and a read / write speed of 200MB per second.

[0072] Processing performance information: mainly refers to the processor performance of the backup device, such as the CPU model, number of cores, main frequency, etc. For example, the processor of backup device C is Intel Core i7, four cores and eight threads, and the main frequency is 3.0GHz; the processor of backup device D is AMD Ryzen 5, six cores and twelve threads, and the main frequency is 2.8GHz.

[0073] Transmission performance information: including network bandwidth, transmission protocol, etc. For example, backup device E is connected via Gigabit Ethernet with a network bandwidth of 1000Mbps; backup device F is connected via a wireless network with a network bandwidth of 500Mbps.

[0074] Stability performance information: Consider the stability indicators of the backup device, such as mean time between failures (MTBF), failure rate, etc. For example, the MTBF of backup device G is 50,000 hours, and the failure rate is 1% per year; the MTBF of backup device H is 40,000 hours, and the failure rate is 2% per year.

[0075] Security performance information: including data encryption, access control, physical security of backup devices, etc. For example, backup device I uses the AES-256 encryption algorithm to encrypt data, sets a strict access control policy, and is placed in a secure computer room; backup device J has a weaker encryption algorithm, looser access control, and lower physical security.

[0076] The following is an example of how the scores for the comprehensive performance information for each section are calculated:

[0077] The storage performance score is calculated as follows: 1. Storage capacity score: It can be scored based on the actual storage capacity and the preset different capacity intervals. For example, the score for the 0 - 1TB interval is 20 points, the score for the 1 - 2TB interval is 40 points, the score for the 2 -3TB interval is 60 points, and so on. Assuming that the storage capacity of the backup device is X TB, its storage capacity score = 20 * (X / / 1 + 1) (X / / 1 means rounding X). 2. Read and write speed score: It is also divided into different speed intervals for scoring. For example, the score for the 0 - 200MB / s interval is 20 points, the score for the 200 - 400MB / s interval is 40 points, the score for the 400 -600MB / s interval is 60 points, and so on. If the read and write speed of the backup device is Y MB / s, its read and write speed score can be determined according to the interval where Y is located. 3. Storage media type score: You can preset the solid-state drive score to be 40 points, the mechanical hard disk score to be 30 points, and so on. Assuming that the storage medium type used by the backup device is Z, its storage medium type score is directly the corresponding preset score.

[0078] Total storage performance score = storage capacity score * 0.4 + read / write speed score * 0.4 + storage media type score * 0.2.

[0079] The processing performance score is calculated as follows: CPU model score: You can subjectively score the performance of different processor models. For example, Intel Core i9 scores 50 points, Intel Core i7 scores 40 points, AMD Ryzen 7 scores 45 points, AMD Ryzen 5 scores 35 points, etc. If the CPU model of the backup device is W, its CPU model score is directly the corresponding preset score. Core number score: Assuming that each core scores 5 points, if the backup device has N cores, the core number score = 5 * N. Main frequency score: Scoring is divided into different main frequency ranges. For example, the score of 0-2.0GHz range is 20 points, the score of 2.0-2.5GHz range is 30 points, and the score of 2.5-3.0GHz range is 40 points. If the main frequency of the backup device is Z GHz, its main frequency score is determined according to the range where Z is located. Total processing performance score = CPU model score * 0.4 + core number score * 0.3 + main frequency score * 0.3.

[0080] The transmission performance score is calculated as follows: 1. Network bandwidth score: Scored according to different bandwidth intervals. For example, the score for the 0-500Mbps interval is 30 points, the score for the 500-1000Mbps interval is 60 points, and the score for 1000Mbps and above is 80 points, etc. If the network bandwidth of the backup device is X Mbps, its network bandwidth score is determined according to the interval where X is located. 2. Transmission protocol score: Assume that the Gigabit Ethernet transmission protocol score is 40 points, the wireless network transmission protocol score is 30 points, etc. If the transmission protocol of the backup device is Y, its transmission protocol score is directly the corresponding preset score. Total transmission performance score = network bandwidth score * 0.6 + transmission protocol score * 0.4.

[0081] The stability performance score is calculated as follows: 1. MTBF score: The score is divided into different mean time between failures intervals. For example, the score for the interval of 0 - 30,000 hours is 20 points, the score for the interval of 30,000 - 40,000 hours is 40 points, the score for the interval of 40,000 - 50,000 hours is 60 points, etc. If the MTBF of the backup device is Z hours, then its MTBF score is determined according to the interval in which Z is located. 2. Failure rate score: Low failure rates are scored high, and high failure rates are scored low. For example, the score for the interval of 0 - 1% failure rate is 40 points, and the score for the interval of 1% - 2% is 30 points, etc. If the failure rate of the backup device is X%, then its failure rate score is determined according to the interval in which X is located. 3. Total stability performance score = MTBF score * 0.6 + failure rate score * 0.4.

[0082] The security performance score is calculated as follows: 1. Data encryption score: Strong encryption algorithms get high scores, and weak encryption algorithms get low scores. For example, the AES-256 encryption algorithm scores 40 points, and the weaker encryption algorithm scores 20 points, etc. If the encryption algorithm used by the backup device is Y, its data encryption score is directly the corresponding preset score. 2. Access control score: Strict access control policies get high scores, and loose access control policies get low scores. For example, strict access control scores are 30 points, and loose access control scores are 20 points, etc. If the access control policy of the backup device is Z, its access control score is directly the corresponding preset score. 3. Physical security score: Placed in a secure environment such as a secure computer room, it gets a high score, and low physical security gets a low score. For example, a secure computer room scores 30 points, and a common environment scores 20 points, etc. If the physical security of the backup device is W, its physical security score is directly the corresponding preset score. 4. Total security performance score = data encryption score * 0.4 + access control score * 0.3 + physical security score * 0.3.

[0083] Capability value quantitative evaluation method: The weighted average method can be used to assign different weights to each performance indicator, and then calculate the comprehensive capability value of the backup device. For example, the storage performance weight is 0.3, the processing performance weight is 0.2, the transmission performance weight is 0.2, the stability performance weight is 0.2, and the security performance weight is 0.1. Assuming that the storage performance score of backup device K is 80 points, the processing performance score is 70 points, the transmission performance score is 60 points, the stability performance score is 75 points, and the security performance score is 85 points, then its comprehensive capability value is: 80×0.3 + 70×0.2 + 60×0.2 + 75×0.2 + 85×0.1 = 74.5 points.

[0084] Step S500, receiving and backing up relatively high priority computing power data from high to low according to the backup capability value of the backup device, until the remaining backup computing power data of the backup device is cleared or the computing power data to be backed up has been completely backed up.

[0085] Assume that the enterprise has three backup devices, the capacity value of backup device L is 90 points, the capacity value of backup device M is 80 points, and the capacity value of backup device N is 70 points. Currently, there is a batch of critical business data that needs to be backed up. The system first selects backup device L with the highest backup capacity value for backup. If the remaining storage space of backup device L is insufficient to accommodate all critical business data, the system will continue to select backup device M for backup, and so on, until all high-priority computing power data is backed up or the remaining backup computing power data of the backup device is cleared.

[0086] Reference Figure 2 , the backup trigger information for obtaining computing power data includes:

[0087] Step S110, analyzing whether the backup trigger conditions of the computing power data are met, the backup trigger conditions include time trigger conditions, data volume trigger conditions, and event trigger conditions. If yes, execute step S120; if no, execute step S130.

[0088] Time trigger conditions can be set at different time points according to the business needs of the enterprise and the importance of data. This method can ensure that data is backed up regularly at specific time points to prevent data loss.

[0089] The data volume trigger condition can be set according to the capacity of the storage device and the data growth rate. This can prevent the storage device from running at full capacity due to excessive data volume, and can also ensure timely backup when the data volume is large, thereby improving data security.

[0090] Event trigger conditions can be set according to the specific business conditions of the enterprise. This method can back up data in time when key events occur, ensuring data integrity and recoverability.

[0091] Step S120, continue with subsequent steps.

[0092] If the backup trigger condition is met, the system will continue to execute the subsequent steps, namely, obtaining the computing power data category and the remaining backup computing power data volume of all backup devices within the preset distance range. For example, when the time trigger condition is met, the system will automatically enter step S200, obtain the categories of various computing power data, and check the remaining backup computing power data volume of the backup devices within the preset distance range to prepare for the subsequent backup operation.

[0093] Step S130, stop the subsequent steps.

[0094] If the backup trigger condition is met, the system will continue to execute the subsequent steps, namely, obtaining the computing power data category and the remaining backup computing power data volume of all backup devices within the preset distance range. For example, when the time trigger condition is met, the system will automatically enter step S200, obtain the categories of various computing power data, and check the remaining backup computing power data volume of the backup devices within the preset distance range to prepare for the subsequent backup operation.

[0095] Reference Figure 3 , the acquisition of the preset distance range includes:

[0096] Step S410, obtaining the backup trigger condition associated with the backup trigger information of the current computing power data.

[0097] The backup trigger condition may be time trigger, data volume trigger or event trigger. For example: if it is time trigger, the system may set a specific time point every day for backup, such as 12 o'clock every night. At this time, the backup trigger condition obtained is the condition that the time reaches 12 o'clock at night. For data volume trigger, suppose it is set to trigger the backup when the amount of data in the database reaches a certain value, such as the amount of data reaches 1TB. When the amount of data in the database approaches or reaches 1TB, the obtained data volume trigger condition is activated. In the case of event trigger, the backup is triggered when the system undergoes a major upgrade or a major change occurs in key business data. When the system is preparing for a major upgrade or a major change is detected in key business data, the corresponding event trigger condition will be obtained.

[0098] Step S420, analyzing and determining the preset distance range according to the correspondence between the backup trigger condition, the computing power data category and the preset distance range.

[0099] Assume that there are different combinations of backup trigger conditions and computing power data categories, corresponding to different preset distance ranges.

[0100] For example, when the backup trigger condition is time-triggered and the computing power data category is critical business data, the preset distance range may be set to backup devices within 100 kilometers from the main server. Because critical business data requires faster and more reliable backup devices in time-triggered backup, choosing a backup device that is closer can reduce data transmission time and potential risks.

[0101] If it is triggered by data volume, and the computing power data category is log data, the preset distance range may be set to backup devices within 500 kilometers from the main server. Because log data is relatively less important and the data volume may be large, you can choose a backup device that is slightly farther away but still has a certain degree of reliability and storage capacity.

[0102] For event triggers, such as a major system upgrade, and the computing power data category is system configuration data, the preset distance range may be set to a backup device within 200 kilometers from the main server. In critical events such as system upgrades, system configuration data needs to be backed up to a reliable device as soon as possible to prevent data loss due to problems during the upgrade process, so a closer backup device is selected.

[0103] Reference Figure 4 According to the mapping relationship between computing power data categories and backup priority levels, the backup priority levels of different computing power data categories are analyzed and determined as follows:

[0104] Step S310, analyzing whether the acquired computing power data category is in a business critical node period. If yes, executing step S320; if yes, executing step S330.

[0105] Business critical node periods can be defined based on specific business situations. For example, in an e-commerce company, the “Double Eleven” shopping festival is a business critical node period.

[0106] Assume that the computing power data being analyzed is order data. The system will check whether the current time is in a business-critical node period such as "Double Eleven". If so, step S320 will be executed; if not, step S330 will be executed.

[0107] Step S320, based on the mapping relationship between the computing power data category at the business critical node period, the backup trigger condition and the backup priority level, analyze and determine the backup priority level of the computing power data category at the business critical node period.

[0108] During business-critical periods, different computing power data categories may have different backup priorities.

[0109] Taking e-commerce companies as an example, during the "Double Eleven" period, the backup priority of key business data such as order data and inventory data will be increased. Assume that the backup trigger condition is time-triggered, and backup is performed every hour. During the critical business node period, the backup priority of order data may be increased from the usual medium priority to the high priority.

[0110] For example, based on pre-set mapping relationships, the system determines that during the "Double Eleven" period, order data has a high backup priority under time-triggered conditions, in order to ensure the security and recoverability of critical data during business peak periods.

[0111] Step S330, analyzing and determining the backup priority levels of different computing power data categories based on the mapping relationship between computing power data categories, backup trigger conditions and backup priority levels.

[0112] During non-business-critical periods, the backup priority is determined according to the conventional mapping relationship.

[0113] For example, in normal times, user browsing history data may be set as a low priority backup. If the backup trigger condition is data volume trigger, the backup will be performed when the browsing history data volume reaches a certain value. Based on the conventional mapping relationship, the system will determine that during non-critical node periods, browsing history data has a low backup priority under the data volume trigger condition.

[0114] Reference Figure 5 According to the mapping relationship between computing power data categories and backup priority levels, the backup priority levels of different computing power data categories are analyzed and determined as follows:

[0115] Step S3a0, analyzing whether the acquired computing power data category is in a critical business node period and / or a major business change stage. If yes, execute step S3b0; if no, execute step S3c0.

[0116] Key business node periods can be major promotional activities of the enterprise, such as "618" and "Double 11" of e-commerce platforms. Important business change stages can be periods when the enterprise conducts major system upgrades and adjusts its business model.

[0117] For example, for an e-commerce platform, the computing power data category currently being analyzed is order data. The system will determine whether it is currently in the "Double 11" period (a critical business node period), and will also determine whether the platform is undergoing a new payment system upgrade (an important business change phase).

[0118] Step S3b0, analyze and determine the priority of the computing power data category that is simultaneously in the business critical node period and the business important change stage according to the mapping relationship between the computing power data category, backup trigger conditions and backup priority levels that are simultaneously in the business critical node period and the business important change stage; analyze and determine the priority of the computing power data category that is alone in the business critical node period according to the mapping relationship between the computing power data category, backup trigger conditions and backup priority levels that are alone in the business critical node period; analyze and determine the priority of the computing power data category that is alone in the business important change stage according to the mapping relationship between the computing power data category, backup trigger conditions and backup priority levels that are alone in the business important change stage.

[0119] Assume that order data is in the "Double 11" and payment system upgrade stages at the same time. According to the pre-set mapping relationship, the backup priority of order data in this case may be set to the highest level. Because in this special period, order data is extremely important, and once lost, it may cause huge losses to the company.

[0120] If it is only during the "Double 11" period (a critical business node period), the backup priority of the order data may be set to a high priority, but it is slightly lower than the priority during two special stages at the same time.

[0121] If you are only in the payment system upgrade phase (or in the phase of major business changes alone), the backup priority of order data will be increased accordingly, but it may be lower than the situation where you are in both phases at the same time or the situation where you are only in the "Double 11" period.

[0122] Step S3c0, analyzing and determining the backup priority levels of different computing power data categories based on the mapping relationship between computing power data categories, backup trigger conditions and backup priority levels.

[0123] During non-business critical node periods and non-business important change phases, the backup priority is determined according to the conventional mapping relationship.

[0124] For example, the user's browsing history data may be set as a low priority backup in normal times. If the backup trigger condition is data volume trigger, the backup will be performed when the browsing history data volume reaches a certain value. The system will determine that the browsing history data has a low backup priority under the data volume trigger condition during non-special periods based on the conventional mapping relationship.

[0125] The capacity values ​​of all backup devices within the preset distance range are analyzed and determined, including:

[0126] Step S410: Collect historical performance data and related environmental data of all backup devices.

[0127] Among them, historical performance data includes the number of read and write times, error rate, and temperature of the storage medium, and environmental data includes temperature, humidity, and voltage.

[0128] Assume that there are multiple backup devices, such as backup devices A, B, C, etc.

[0129] For backup device A: The number of read and write operations of the storage medium can be obtained through the device's system log or special monitoring software. For example, in the past week, the storage medium was read 1,000 times and written 500 times. The error rate can be determined by checking the number of errors that occurred during the reading and writing process of the storage medium. If a total of 10,000 read and write operations were performed in the past month and 10 errors occurred, the error rate is 0.1%. The temperature can be obtained through the device's built-in temperature sensor. Assume that the current temperature of the storage medium of backup device A is 40 degrees Celsius.

[0130] In terms of environmental data, temperature can be obtained through the temperature sensor in the computer room, assuming that the temperature of the computer room is 25 degrees Celsius; humidity can be measured by a hygrometer, for example, the humidity is 50%; voltage can be obtained through power monitoring equipment, for example, the voltage is 220V.

[0131] Step S420, taking the collected historical performance data of all backup devices and related environmental data as input objects, inputting them into the pre-established and trained storage medium life prediction model and backup device failure prediction model, and outputting the quantified value of the possibility of backup device failure and the quantified value of the remaining life of the storage medium.

[0132] The general process of establishing and training the storage medium life prediction model and the backup device failure prediction model is as follows:

[0133] 1. Data collection: For the storage media life prediction model, collect historical performance data of the storage media, including the number of read and write times, error rate, temperature and other indicators in different time periods. This data can be obtained from the system log of the backup device, storage management software and other channels. Record the actual service life of the storage media, for example, the time span from the start of use to the failure or end of service life of a hard disk. This information can be obtained through equipment replacement records, maintenance logs, etc. For the backup device failure prediction model: Collect historical performance data of the backup device, such as processor utilization, memory utilization, network bandwidth utilization, and the above-mentioned related indicators of the storage medium. Collect environmental data, including temperature, humidity, voltage, etc., which may affect the stability and reliability of the backup device. Mark the actual failure of the backup device, including the time when the failure occurred, the type of failure, etc.

[0134] 2. Data preprocessing: 1. Data cleaning, removing outliers, erroneous data, and incomplete data records. For example, if the number of reads and writes of a data point is obviously abnormally high or low, it may be a data collection error and needs to be corrected or deleted. 2. Data normalization: Normalize the data of different indicators to make them have the same scale. For example, map the values ​​of indicators such as temperature, humidity, and the number of reads and writes to the [0, 1] interval so that the model can handle it better. 3. Feature engineering: Extract useful features based on the characteristics of the data and the needs of the problem. For example, the growth rate of the number of reads and writes over a period of time, the trend of the error rate, etc. can be calculated as new features.

[0135] 3. Model selection: For the storage medium life prediction model, you can choose regression models such as linear regression, random forest regression, support vector regression, etc. These models can predict the remaining life of the storage medium based on the input storage medium performance indicators. For the backup device failure prediction model, you can choose classification models such as decision trees, random forests, support vector machines, neural networks, etc. These models can predict whether the device will fail based on the input device performance and environmental data.

[0136] 4. Model training: Divide the preprocessed data into a training set and a test set. Usually, a certain ratio can be used, such as 70% of the data as a training set and 30% of the data as a test set. Use the training set to train the selected model. By adjusting the parameters of the model, the prediction error of the model on the training set is minimized. For example, for a random forest model, parameters such as the number of trees and the depth of the trees can be adjusted. During the training process, techniques such as cross-validation can be used to evaluate the performance and stability of the model. Cross-validation divides the data into multiple subsets, and uses one of the subsets as a test set and the remaining subsets as a training set in turn. The model is trained and evaluated multiple times to obtain a more reliable performance evaluation.

[0137] 5. Model evaluation and optimization: Use the test set to evaluate the trained model. Evaluation indicators can include mean square error (MSE), mean absolute error (MAE), accuracy, recall, F1 value, etc., which are selected according to the type of model and the needs of the problem. If the performance of the model does not meet the requirements, the following optimization measures can be taken: Adjust the parameters of the model: Try different parameter combinations to find the optimal parameter settings. Increase the amount of data: Collect more historical data to improve the generalization ability of the model. Improve feature engineering: Extract more effective features, or use feature selection techniques to remove irrelevant or redundant features. Try other models: If the current model does not work well, you can try different types of models to see if you can get better performance.

[0138] The storage media life prediction model and backup device failure prediction model can be trained through machine learning algorithms. For example, a large amount of historical data, including different read and write times, error rates, temperature, humidity, voltage and other data, as well as the corresponding storage media actual life and device failure conditions, can be used to train the model.

[0139] Assuming that the data of backup device A is input into the model, the storage medium life prediction model outputs a quantitative value of the remaining storage medium life of 80% (indicating the proportion of the total life of the storage medium that is expected to be used normally), and the backup device failure prediction model outputs a quantitative value of the possibility of backup device failure of 10% (indicating the probability of device failure in the future).

[0140] Step S430, analyzing and determining the decline coefficient of the backup device capability value according to the interval range in which the backup device failure possibility quantization value falls, the interval range in which the storage medium remaining life quantization value falls, and the decline coefficient of the capability value.

[0141] For example, you can set several interval ranges:

[0142] Quantitative value of backup equipment failure probability: 0% - 5% is a low-risk interval, corresponding to a decline coefficient of 0.1; 5% -15% is a medium-risk interval, corresponding to a decline coefficient of 0.3; and above 15% is a high-risk interval, corresponding to a decline coefficient of 0.5.

[0143] Quantitative value of the remaining life of the storage medium: above 80% is the high life interval, corresponding to a decline coefficient of 0.1; 50% -80% is the medium life interval, corresponding to a decline coefficient of 0.3; below 50% is the low life interval, corresponding to a decline coefficient of 0.5.

[0144] Assume that the backup device failure probability of backup device A is 10%, which is in the medium risk range, and the remaining life of the storage medium is 80%, which is in the high life range. Taking all factors into consideration, it can be determined that the capacity value decline coefficient of backup device A is 0.2 (the final decline coefficient can be determined by weighted average or other methods according to the specific situation).

[0145] Step S440: Calculate the effective capacity value of the backup device according to the decline coefficient of the backup device capacity value and the backup device capacity value.

[0146] Assume that the initial capacity value of backup device A is 100 (the initial capacity value can be obtained by evaluating the comprehensive performance information mentioned above). Based on the slip coefficient of 0.2, the effective capacity value is calculated to be 100 * (1 - 0.2) = 80. That is, the effective capacity value of backup device A is 80 after considering factors such as the possibility of failure and the life of the storage medium.

[0147] A computing power data operation and maintenance processing method based on AI intelligence also includes the steps of receiving and backing up relatively high-priority computing power data from high to low according to the backup capacity value of the backup device until the remaining backup computing power data of the backup device is cleared or the computing power data to be backed up has been completely backed up in parallel, specifically as follows:

[0148] Get the priority of different categories of computing power data and the effective capacity value of the backup device.

[0149] The priorities of different categories of computing power data can be determined according to the methods mentioned above, such as high priority for critical business data and low priority for temporary data. Suppose we have the following categories of computing power data and their priorities: Critical business data: high priority. System configuration data: high priority. High-value data: medium-high priority. Log data: medium priority. Temporary data: low priority. User preference data: medium priority. Statistical data: medium-low priority.

[0150] The effective capacity value of the backup device can be obtained by analyzing and determining the capacity values ​​of all backup devices within the preset distance range as described above. For example, after calculation, the effective capacity value of a backup device is 80 (assuming the full score is 100).

[0151] According to the priority of different categories of computing power data, the range in which the effective capacity value of the backup device falls and the corresponding relationship between the data compression scheme, the data compression scheme for different categories of computing power data is analyzed and determined.

[0152] First, determine the range of the backup device's effective capacity value. For example, you can divide it into several ranges:

[0153] High ability range: effective ability value is above 80. Medium ability range: effective ability value is between 50 - 80. Low ability range: effective ability value is below 50.

[0154] Then establish the corresponding relationship between the priority of different categories of computing power data and the range of effective capacity values ​​of backup devices and data compression schemes. For example: For high-priority critical business data and system configuration data: If the backup device is in the high capacity range, a lossless compression algorithm, such as the DEFLATE algorithm, can be used to ensure data integrity. If the backup device is in the medium capacity range, a medium-intensity lossless compression algorithm, such as the LZMA algorithm, can be used to minimize the impact on the data while ensuring a certain compression ratio. If the backup device is in the low capacity range, a lightweight lossless compression algorithm, such as the LZ4 algorithm, can be used to increase the compression speed. For high-value data with medium and high priorities: In the high capacity range, a combination of lossy compression and lossless compression can be used, such as the JPEG2000 compression algorithm for lossy compression of image data, while using lossless compression for important text information.

[0155] In the medium capability range, a medium-strength lossy compression algorithm, such as the JPEG compression algorithm, can be used. In the low capability range, a lower-quality lossy compression algorithm can be used, or some data can be selectively compressed. For medium-priority log data and user preference data:

[0156] In the high-capacity range, a compression algorithm with a higher compression ratio, such as the GZIP algorithm, can be used. In the medium-capacity range, a common compression algorithm, such as the ZIP algorithm, can be used. In the low-capacity range, it can be decided whether to perform compression or use a very simple compression method according to the situation.

[0157] For low-priority temporary data and statistical data: In the high-capacity range, you can perform deeper compression, such as using the 7-Zip algorithm. In the medium-capacity range, you can use general compression methods. In the low-capacity range, you can do no compression or just perform simple archiving.

[0158] Assume that the effective capacity value of the current backup device is 70, which is in the medium capacity range. For key business data, select the LZMA algorithm for compression; for high-value data, use a medium-strength lossy compression algorithm (such as the JPEG compression algorithm); for log data, use the ZIP algorithm for compression; for temporary data, decide whether to compress it according to the situation.

[0159] Execute data compression schemes for data of different computing power.

[0160] Specifically, according to the determined compression scheme, different categories of computing power data are compressed. For example, for key business data, the LZMA algorithm compression software is used to compress it. For high-value data, the corresponding lossy compression tool is used for compression.

[0161] During the compression process, the security and integrity of the data must be ensured. For important data, data verification can be performed before and after compression to ensure that the compression process does not cause data corruption.

[0162] After compression is complete, the compressed data is stored and backed up according to the backup plan to ensure that the data can be quickly restored when needed.

[0163] Based on the same inventive concept, an embodiment of the present invention provides a computing power data operation and maintenance processing system based on AI intelligence, including a memory and a processor, wherein the memory stores a data that can be executed on the processor to implement the following Figures 1 to 5 Procedure for either method.

[0164] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A computing power data operation and maintenance processing method based on AI intelligence, characterized in that: include: Obtain backup trigger information for computing power data; Obtain computing power data categories and the amount of remaining backup computing power data of all backup devices within a preset distance. The computing power data categories include key business data, system configuration data, high-value data, log data, temporary data, user preference data, and statistical data. According to the mapping relationship between computing power data categories and backup priority levels, analyze and determine the backup priority levels of different computing power data categories; Obtaining comprehensive performance information of all backup devices within a preset distance range, and analyzing and determining the capacity values ​​of all backup devices within a preset distance range according to a preset capacity value quantitative evaluation method of the comprehensive performance information, wherein the comprehensive performance information includes storage performance information, processing performance information, transmission performance information, stability performance information, and security performance information; Receive and back up relatively high-priority computing power data from high to low according to the backup capacity value of the backup device, until the remaining backup computing power data of the backup device is cleared or all the computing power data to be backed up has been backed up; The capacity values ​​of all backup devices within the preset distance range are analyzed and determined, including: Collect historical performance data and related environmental data of all backup devices, where historical performance data includes the number of read and write times, error rate, and temperature of the storage medium, and environmental data includes temperature, humidity, and voltage; The collected historical performance data of all backup devices and related environmental data are used as input objects, and are input into a storage medium life prediction model and a backup device failure prediction model that have been established and trained in advance, and a quantified value of the backup device failure probability and a quantified value of the remaining life of the storage medium are output; Analyze and determine the decline coefficient of the backup device capability value based on the interval range in which the backup device failure probability quantified value falls, the interval range in which the storage medium remaining life quantified value falls, and the decline coefficient of the capability value; The effective capacity value of the backup device is calculated according to the decline coefficient of the backup device capacity value and the backup device capacity value; According to the mapping relationship between computing power data categories and backup priority levels, the backup priority levels of different computing power data categories are analyzed and determined as follows: Analyze the key business nodes and important business change stages of the acquired computing power data categories; If the business is in a critical node period and a stage of important business changes at the same time, the priority level of the computing power data category that is in a critical node period and a stage of important business changes shall be analyzed and determined based on the mapping relationship between the computing power data category, backup trigger conditions and backup priority level that are in a critical node period and a stage of important business changes at the same time; If it is in the critical business node period alone, the priority level of the computing power data category in the critical business node period alone is analyzed and determined according to the mapping relationship between the computing power data category in the critical business node period alone, the backup trigger condition and the backup priority level; If the business is in a stage of significant business changes, the priority of the computing power data category that is in a stage of significant business changes shall be analyzed and determined based on the mapping relationship between the computing power data category that is in a stage of significant business changes, the backup trigger conditions, and the backup priority level; If the business is not in a critical node period or a major business change stage, the backup priority levels of different computing power data categories are analyzed and determined based on the mapping relationship between computing power data categories, backup trigger conditions, and backup priority levels. The method further includes receiving and backing up relatively high-priority computing power data from high to low according to the backup capability value of the backup device until the remaining backup computing power data of the backup device is cleared or the computing power data to be backed up has been completely backed up, which is as follows: Obtain the priority of different types of computing power data and the effective capacity value of backup devices; Analyze and determine the data compression schemes for different categories of computing power data based on the priorities of different categories of computing power data, the intervals into which the effective capacity values ​​of the backup devices fall, and the corresponding relationship between the data compression schemes; Execute data compression schemes for data of different computing power.

2. According to the AI-based computing data operation and maintenance processing method of claim 1, it is characterized in that: The backup trigger information for obtaining computing power data includes: Analyze whether the backup trigger conditions of computing power data are met. The backup trigger conditions include time trigger conditions, data volume trigger conditions, and event trigger conditions. If yes, continue with the next steps; If not, stop the subsequent steps.

3. According to the AI-based computing data operation and maintenance processing method of claim 2, it is characterized in that: The acquisition of the preset distance range includes: Obtain the backup trigger conditions associated with the backup trigger information of this computing power data; According to the correspondence between the backup trigger conditions, computing power data categories and the preset distance range, the preset distance range is analyzed and determined.

4. According to the AI ​​intelligence-based computing power data operation and maintenance processing method of claim 2, it is characterized in that: According to the mapping relationship between computing power data categories and backup priority levels, the backup priority levels of different computing power data categories are analyzed and determined as follows: Analyze whether the acquired computing power data category is at a critical business node period; If yes, then according to the mapping relationship between the computing power data category at the business critical node period, the backup trigger condition and the backup priority level, analyze and determine the backup priority level of the computing power data category at the business critical node period; If not, the backup priority levels of different computing power data categories are analyzed and determined based on the mapping relationship between computing power data categories, backup trigger conditions and backup priority levels.

5. A computing power data operation and maintenance processing system based on AI intelligence, characterized in that: It includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a computing power data operation and maintenance processing method based on AI intelligence as described in any one of claims 1 to 4.

Images