Intelligent database fault diagnosis system and method

Abstract

The invention relates to the technical field of intelligent databases, and discloses an intelligent database fault diagnosis system and method.The intelligent database fault diagnosis system comprises a data patrol module, a data carding module, a data reasoning module and a data output module; the database system inspection module is used for collecting database system inspection data and then sending the database system inspection data to the data sorting module, the data sorting module is used for sorting and numbering the collected database system data and calculating a database fault index according to the sorted and numbered data, and the data sorting module sends a database fault index calculation result to the data reasoning module. And the data reasoning module is used for reasoning and judging the database fault condition according to the database fault index calculation result and sending the database fault condition reasoning and judging result to the data output module, and the data output module is used for outputting the database fault condition reasoning and judging result to comprehensively and accurately diagnose the database system.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent databases, and in particular to an intelligent database fault diagnosis system and method. Background Art

[0002] With the rapid development of computer technology, the amount of data generated in various fields has exploded. From simple business data in the early days to the massive amounts of data generated by the Internet and the Internet of Things today, traditional data storage and management methods are no longer able to meet the needs. This necessitates a more efficient and powerful data storage and management approach, which is where databases come in. As the volume of data increases dramatically, so does its complexity. Modern applications need to process data of various types and structures, including structured data (such as numbers and dates), semi-structured data (such as data in formats like XML and JSON), and unstructured data (such as text, images, audio, and video). Databases provide a flexible way to store and manage this complex data. Through the design of appropriate data models, these different types of data can be easily integrated and manipulated.

[0003] The emergence of databases is to meet people's needs for efficient data storage, management and utilization. In the operation of modern enterprises, many key businesses are highly dependent on database systems. In the event of a database failure, the database cannot be opened and the business is paralyzed. Existing database systems are complex, involving multiple components and layers. Failures may occur in hardware, operating systems, networks or database software itself, increasing the difficulty of diagnosis and making it impossible to accurately predict failure conditions, resulting in data unusable and system paralysis. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In response to the shortcomings of the existing technology, the present invention provides an intelligent database fault diagnosis system and method, which has the advantages of combining multiple data parameters to comprehensively analyze and monitor the database system, improve monitoring accuracy, promptly discover possible anomalies in the database system, accurately predict fault conditions, and provide guarantees for database operation and use.

[0006] (2) Technical solution

[0007] To achieve the above-mentioned object, the present invention provides the following technical solutions: an intelligent database fault diagnosis system, comprising a data inspection module, a data combing module, a data reasoning module and a data output module;

[0008] The data inspection module is used to inspect the database system and collect the database system inspection data and send it to the data combing module;

[0009] The data combing module is used to comb and number the collected database system data, and calculate the database fault index based on the combed and numbered data. The data combing module sends the database fault index calculation result to the data reasoning module;

[0010] The data reasoning module is used to perform fault diagnosis on the database based on the calculation result of the database fault index and send the result to the data output module;

[0011] The data output module is used to output the database fault diagnosis and judgment results.

[0012] Preferably, the database system inspection data includes hardware inspection data, network inspection data, operation inspection data, and performance inspection data. The hardware inspection data is collected through the hardware monitoring software built into the data inspection module, the network inspection data is collected through the network monitoring software deployed by the data inspection module, the operation inspection data is collected through the database management system built into the data inspection module, and the performance inspection data is collected through the database performance monitoring tool built into the data inspection module.

[0013] Preferably, the data sorting module sorts and numbers the collected database system data as follows:

[0014] The numbering representation of the hardware inspection data is: [Ys w 、Ys y 、Ys l 、Ys s 、Ys g ], represents the formula, Ys w Represents the temperature data of the hardware, Ys y Represents the voltage data of the hardware, Ys l Represents the current data of the hardware, Ys s Represents the comprehensive hardware utilization data, Ys g Represents hardware failure rate data;

[0015] The network inspection data number representation is: [, Ws y 、Ws c 、Ws j ], represents the formula, Ws l Represents network bandwidth utilization, Ws y Represents the network delay rate, Ws c Represents the transmission error rate, Ws j Represents the connection error rate;

[0016] The numbering representation of the operation inspection data is: [Xs d 、Xs l 、Xsx 、Xs q 、Xs s ], represents the formula, Xs d Represents the number of interruptions, Xs l Represents the number of connection exceptions, Xs x Represents the running response time, Xs q Represents the startup time, Xs s Represents the running speed;

[0017] The numbering representation of the performance inspection data is: [Ns z 、Ns h 、Ns c 、Ns m ], represents the formula, Ns z Represents execution time, Ns h Represents the cache hit rate, Ns c Represents the query response time, Ns m Indicates the number of slow queries.

[0018] Preferably, the calculation formula of the database failure index Sg is:

[0019] Sg=Yy*Wy*Xy*Ny

[0020] In the calculation formula, Sg represents the database failure index, Yy represents the hardware anomaly index, Wy represents the network transmission anomaly index, Xy represents the operation status anomaly index, and Ny represents the performance anomaly index.

[0021] Preferably, the calculation formula of the hardware anomaly index Yy is:

[0022]

[0023] In the calculation formula, Yy represents the hardware abnormality index. Respectively represent the historical average values of the hardware temperature data, hardware voltage data, and hardware current data. They represent the maximum allowable differences of the hardware temperature data, the hardware voltage data, and the hardware current data, respectively. a1, a2, a3, a4, and a5 represent the weights of the hardware temperature data, the hardware voltage data, the hardware current data, the hardware comprehensive utilization rate data, and the hardware failure rate data, respectively, and a1+a2+a3+a4+a5=1.

[0024] Preferably, the network transmission anomaly index Wy is calculated as follows:

[0025]

[0026] In the calculation formula, Wy represents the network transmission anomaly index, i represents the i-th parameter data in the network inspection data, and i o Represents the historical average value of the i-th parameter data in the network inspection data, i v Represents the maximum allowable difference of the i-th parameter data in the network inspection data, i β Represents the weight of the i-th parameter data in the network inspection data, Represents all parameters in the network inspection data from i=Ws l Start to i=Ws j End, the sum of the ratios between the weights and differences of all i-th parameters and the maximum allowed difference.

[0027] Preferably, the calculation formula of the abnormal operating status index Xy is:

[0028]

[0029] In the calculation formula, They represent the maximum allowable values of the number of operation interruptions, number of connection exceptions, startup time, operation speed, and operation response time. Represents the ratio between the number of operation interruptions and the maximum allowed value, represents the nonlinear correction function;

[0030] Represents the ratio between the number of connection exceptions and the maximum allowed value;

[0031] Represents the ratio between the startup time and the maximum allowed value;

[0032] Represents the ratio of actual efficiency to reference efficiency, Xs s / Xs x Indicates the processing capacity per unit response time. Indicates the ideal processing capacity per unit response time;

[0033] γ1, γ2, γ3, and γ4 represent the weight factors of each part respectively, and γ1+γ2+γ3+γ4=1.

[0034] Preferably, the calculation formula of the performance abnormality index Ny is:

[0035]

[0036] In the calculation formula, They represent the maximum allowed values of execution time, cache hit rate, query response time, and number of slow queries, respectively. Represents the ratio between execution time and the maximum allowed value, represents a nonlinear function;

[0037] 1-Ns h Represents the performance loss caused by quantifying the cache hit rate, Used to measure the degree of deviation of cache efficiency, Represents the impact of the actual cache hit rate on the system;

[0038] represents the ratio between the query response time and the maximum allowed value, k represents the nonlinear coefficient, represents the effect of amplified response time;

[0039] Represents the ratio between the number of slow queries and the maximum allowed value;

[0040] θ1, θ2, θ3, and θ4 represent the weights of execution time, cache hit rate, query response time, and number of slow queries, respectively, and θ1+θ2+θ3+θ4=1.

[0041] Preferably, a database fault index standard threshold Bg is set inside the data inference module. When the database fault index Sg is greater than the database fault index standard threshold Bg, it means that there is a fault in the current database, and the data output module outputs a database fault signal.

[0042] The intelligent database fault diagnosis method, based on the intelligent database fault diagnosis system, includes the following steps:

[0043] Step 1: Inspect the database system and collect database system inspection data;

[0044] Step 2: Sort and number the collected database system data and calculate the database failure index;

[0045] Step 3: diagnose the database fault condition according to the calculation result of the database fault index;

[0046] Step 4: Output the diagnosis results of the database fault situation.

[0047] Compared with the prior art, the present invention provides an intelligent database fault diagnosis system and method, which has the following beneficial effects:

[0048] 1. The present invention combines hardware inspection data, network inspection data, operation inspection data, and performance inspection data to form database system inspection data, thereby achieving comprehensive analysis and monitoring of the database system, with wide coverage, improving monitoring accuracy, timely discovering possible anomalies in the database system, accurately predicting fault conditions, and providing protection for database operation and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a step diagram of the method of the present invention;

[0050] Figure 2 Schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0052] See also Figure 1-2 Intelligent database fault diagnosis system, including data inspection module, data combing module, data reasoning module and data output module;

[0053] The data inspection module is used to inspect the database system and collect the inspection data of the database system and send it to the data combing module;

[0054] Database system inspection data includes hardware inspection data, network inspection data, operation inspection data, and performance inspection data. Hardware inspection data is collected through the hardware monitoring software built into the data inspection module. Network inspection data is collected through the network monitoring software deployed in the data inspection module. Operation inspection data is collected through the database management system built into the data inspection module. Performance inspection data is collected through the database performance monitoring tool built into the data inspection module.

[0055] The data combing module is used to comb and number the collected database system data, and calculate the database fault index based on the combed and numbered data. The data combing module sends the database fault index calculation result to the data inference module;

[0056] The data combing module combs and numbers the collected database system data as follows:

[0057] The hardware inspection data number representation is: [Ys w 、Ys y 、Ys l 、Ys s 、Ys g ], represents the formula, Ys w Represents the temperature data of the hardware, Ys y Represents the voltage data of the hardware, Ys l Represents the current data of the hardware, Ys s Represents the comprehensive hardware utilization data, Ys gRepresents hardware failure rate data;

[0058] The numbering representation of network inspection data is: [, Ws y 、Ws c 、Ws j ], represents the formula, Ws l Represents network bandwidth utilization, Ws y Represents the network delay rate, Ws c Represents the transmission error rate, Ws j Represents the connection error rate;

[0059] The numbering formula of the operation inspection data is: [Xs d 、Xs l 、Xs x 、Xs q 、Xs s ], represents the formula, Xs d Represents the number of interruptions, Xs l Represents the number of connection exceptions, Xs x Represents the running response time, Xs q Represents the startup time, Ns s Represents the running speed;

[0060] The numbering representation of performance inspection data is: [Ns z 、Ns h 、Ns c 、Ns m ], represents the formula, Ns z Represents execution time, Ns h Represents the cache hit rate, Ns c Represents the query response time, Ns m Represents the number of slow queries;

[0061] Database system inspection data is composed of hardware inspection data, network inspection data, operation inspection data, and performance inspection data. This allows for comprehensive analysis and monitoring of the database system, with extensive coverage, improved monitoring accuracy, and timely detection of possible anomalies in the database system, providing assurance for database operation and use.

[0062] The calculation formula of database failure index Sg is:

[0063] Sg=Yy*Wy*Xy*Ny

[0064] In the calculation formula, Sg represents the database failure index, Yy represents the hardware anomaly index, Wy represents the network transmission anomaly index, Xy represents the operation status anomaly index, and Ny represents the performance anomaly index;

[0065] The calculation formula of hardware abnormality index Yy is:

[0066]

[0067] In the calculation formula, Yy represents the hardware abnormality index. Respectively represent the historical average values of the hardware temperature data, hardware voltage data, and hardware current data. Respectively represent the maximum allowable differences of the hardware temperature data, the hardware voltage data, and the hardware current data; a1, a2, a3, a4, and a5 represent the weights of the hardware temperature data, the hardware voltage data, the hardware current data, the hardware comprehensive utilization rate data, and the hardware failure rate data, and a1+a2+a3+a4+a5=1;

[0068] The calculation formula of network transmission anomaly index Wy is:

[0069]

[0070] In the calculation formula, Wy represents the network transmission anomaly index, i represents the i-th parameter data in the network inspection data, and i o Represents the historical average value of the i-th parameter data in the network inspection data, i v Represents the maximum allowable difference of the i-th parameter data in the network inspection data, i β Represents the weight of the i-th parameter data in the network inspection data, Represents all parameters in the network inspection data from i=Ws l Start to i=Ws j End, the sum of the ratios between the sum of all i-th parameter weights and differences and the maximum allowed difference;

[0071] The calculation formula of the operating status abnormality index Xy is:

[0072]

[0073] In the calculation formula, They represent the maximum allowable values of the number of operation interruptions, number of connection exceptions, startup time, operation speed, and operation response time. Represents the ratio between the number of operation interruptions and the maximum allowed value, Represents a nonlinear correction function, which is used to amplify the impact of high-frequency interruptions. The more interruptions there are, the greater the contribution to the anomaly index.

[0074] It represents the ratio between the number of connection exceptions and the maximum allowed value, which directly reflects the stability of the network or service connection;

[0075] Represents the ratio of the startup time to the maximum allowed value. The longer the startup time, the lower the system initialization efficiency, which may indicate resource contention or configuration problems.

[0076] Represents the ratio of actual efficiency to reference efficiency, Xs s / Xs x Indicates the processing capacity per unit response time. Indicates the ideal processing capacity per unit response time, It represents the reverse processing of the efficiency ratio. The lower the efficiency, the higher the contribution value.

[0077] γ1, γ2, γ3, and γ4 represent the weight factors of each part respectively, and γ1+γ2+γ3+γ4=1.

[0078] By combining speed and response time into efficiency terms, we directly quantify system resource utilization, avoiding optimizing a single metric while ignoring overall performance. We also apply nonlinear weighting to high-frequency interruptions and efficiency losses, which is more in line with actual usage sensitivity. By quantifying core indicators, we provide an operational status assessment method that balances simplicity and flexibility, enabling accurate diagnosis of database faults.

[0079] The calculation formula of performance abnormality index Ny is:

[0080]

[0081] In the calculation formula, They represent the maximum allowed values of execution time, cache hit rate, query response time, and number of slow queries, respectively. Represents the ratio between execution time and the maximum allowed value, Represents a nonlinear function used to amplify the impact of timeouts. The longer the execution time, the worse the performance.

[0082] 1-Ns h It represents the performance loss caused by quantifying the cache hit rate. The higher the hit rate, the more frequently the system needs to access slow resources due to cache failure, that is, the worse the performance. Used to measure the degree of deviation of cache efficiency, Represents the impact of the actual cache hit rate on the system.

[0083] represents the ratio between the query response time and the maximum allowed value, k represents the nonlinear coefficient, Represents the impact of amplified response time. The longer the response time, the worse the performance.

[0084] Represents the ratio between the number of slow queries and the maximum allowed value. The more slow queries there are, the more unstable the system performance is, and the contribution value increases linearly.

[0085] θ1, θ2, θ3, and θ4 represent the weights of execution time, cache hit rate, query response time, and number of slow queries, respectively, and θ1+θ2+θ3+θ4=1.

[0086] By quantifying core performance indicators and combining them with nonlinear correction and expansion parameters, a flexible method for evaluating system performance anomalies is provided;

[0087] The data reasoning module is used to perform database fault diagnosis based on the database fault index calculation results and send the results to the data output module;

[0088] A database fault index standard threshold Bg is set inside the data inference module. When the database fault index Sg is greater than the database fault index standard threshold Bg, it means that there is a fault in the current database, and the data output module outputs a database fault signal.

[0089] The above-mentioned intelligent database-based fault diagnosis system includes the following steps:

[0090] Step 1: Inspect the database system and collect database system inspection data;

[0091] Step 2: Sort and number the collected database system data and calculate the database failure index;

[0092] Step 3: diagnose the database fault condition according to the calculation result of the database fault index;

[0093] Step 4: Output the diagnosis results of the database fault situation.

[0094] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. Intelligent database fault diagnosis system, characterized by: Including data inspection module, data combing module, data reasoning module and data output module; The data inspection module is used to inspect the database system and collect the database system inspection data and send it to the data combing module; The data combing module is used to comb and number the collected database system data, and calculate the database fault index based on the combed and numbered data. The data combing module sends the database fault index calculation result to the data reasoning module; The data reasoning module is used to perform fault diagnosis on the database based on the calculation result of the database fault index and send the result to the data output module; The data output module is used to output the database fault diagnosis and judgment results.

2. The intelligent database fault diagnosis system according to claim 1, characterized in that: The database system inspection data includes hardware inspection data, network inspection data, operation inspection data, and performance inspection data. The hardware inspection data is collected through the hardware monitoring software built into the data inspection module, the network inspection data is collected through the network monitoring software deployed by the data inspection module, the operation inspection data is collected through the database management system built into the data inspection module, and the performance inspection data is collected through the database performance monitoring tool built into the data inspection module.

3. The intelligent database fault diagnosis system according to claim 2, characterized in that: The data sorting module sorts and numbers the collected database system data as follows: The numbering representation of the hardware inspection data is: [Ys w 、Ys y 、Ys l 、Ys s 、Ys g ], represents the formula, Ys w Represents the temperature data of the hardware, Ys y Represents the voltage data of the hardware, Ys l Represents the current data of the hardware, Ys s Represents the comprehensive hardware utilization data, Ys g Represents hardware failure rate data; The network inspection data number representation is: [, Ws y 、Ws c 、Ws j ], represents the formula, Ws l Represents network bandwidth utilization, Ws y Represents the network delay rate, Ws c Represents the transmission error rate, Ws j Represents the connection error rate; The numbering representation of the operation inspection data is: [Xs d 、Xs l 、Xs x 、Xs q 、Xs s ], represents the formula, Xs d Represents the number of interruptions, Xs l Represents the number of connection exceptions, Xs x Represents the running response time, Xs q Represents the startup time, Xs s Represents the running speed; The numbering representation of the performance inspection data is: [Ns z 、Ns h 、Ns c 、Ns m ], represents the formula, Ns z Represents execution time, Ns h Represents the cache hit rate, Ns c Represents the query response time, Ns m Indicates the number of slow queries.

4. The intelligent database fault diagnosis system according to claim 3, characterized in that: The calculation formula of the database failure index Sg is: Sg=Yy*Wy*Xy*Ny In the calculation formula, Sg represents the database failure index, Yy represents the hardware anomaly index, Wy represents the network transmission anomaly index, Xy represents the operation status anomaly index, and Ny represents the performance anomaly index.

5. The intelligent database fault diagnosis system according to claim 4, characterized in that: The calculation formula of the hardware abnormality index Yy is: In the calculation formula, Yy represents the hardware abnormality index. Respectively represent the historical average values of the hardware temperature data, hardware voltage data, and hardware current data. They represent the maximum allowable differences of the hardware temperature data, the hardware voltage data, and the hardware current data, respectively. a1, a2, a3, a4, and a5 represent the weights of the hardware temperature data, the hardware voltage data, the hardware current data, the hardware comprehensive utilization rate data, and the hardware failure rate data, respectively, and a1+a2+a3+a4+a5=1.

6. The intelligent database fault diagnosis system according to claim 5, characterized in that: The calculation formula of the network transmission anomaly index Wy is: In the calculation formula, Wy represents the network transmission anomaly index, i represents the i-th parameter data in the network inspection data, and i o Represents the historical average value of the i-th parameter data in the network inspection data, i v Represents the maximum allowable difference of the i-th parameter data in the network inspection data, i β Represents the weight of the i-th parameter data in the network inspection data, Represents all parameters in the network inspection data from i=Ws l Start to i=Ws j End, the sum of the ratios between the weights and differences of all i-th parameters and the maximum allowed difference.

7. The intelligent database fault diagnosis system according to claim 6, characterized in that: The calculation formula of the abnormal operating state index Xy is: In the calculation formula, They represent the maximum allowable values of the number of operation interruptions, number of connection exceptions, startup time, operation speed, and operation response time. Represents the ratio between the number of operation interruptions and the maximum allowed value, represents the nonlinear correction function; Represents the ratio between the number of connection exceptions and the maximum allowed value; Represents the ratio between the startup time and the maximum allowed value; Represents the ratio of actual efficiency to reference efficiency, Xs s / Xs x Indicates the processing capacity per unit response time. Indicates the ideal processing capacity per unit response time; γ1, γ2, γ3, and γ4 represent the weight factors of each part respectively, and γ1+γ2+γ3+γ4=1.

8. The intelligent database fault diagnosis system according to claim 7, characterized in that: The calculation formula of the performance abnormality index Ny is: In the calculation formula, They represent the maximum allowed values of execution time, cache hit rate, query response time, and number of slow queries, respectively. Represents the ratio between execution time and the maximum allowed value, represents a nonlinear function; 1-Ms h Represents the performance loss caused by quantifying the cache hit rate, Used to measure the degree of deviation of cache efficiency, Represents the impact of the actual cache hit rate on the system; represents the ratio between the query response time and the maximum allowed value, k represents the nonlinear coefficient, represents the effect of amplified response time; Represents the ratio between the number of slow queries and the maximum allowed value; θ1, θ2, θ3, and θ4 represent the weights of execution time, cache hit rate, query response time, and number of slow queries, respectively, and θ1+θ2+θ3+θ4=1.

9. The intelligent database fault diagnosis system according to claim 8, characterized in that: The data inference module is internally provided with a database fault index standard threshold value Bg. When the database fault index Sg is greater than the database fault index standard threshold value Bg, it indicates that a fault exists in the current database, and the data output module outputs a database fault signal.

10. An intelligent database fault diagnosis method based on the intelligent database fault diagnosis system according to claims 1 to 9, characterized in that: The following steps are involved: Step 1: Inspect the database system and collect database system inspection data; Step 2: Sort and number the collected database system data and calculate the database failure index; Step 3: diagnose the database fault condition according to the calculation result of the database fault index; Step 4: Output the diagnosis results of the database fault situation.

Images