A method for calculating the comprehensive health of information systems

Abstract

The present invention provides a method for calculating the comprehensive health of an information system, including: health evaluation based on abnormal impact and hierarchical weights of system architecture, aggregation of health of multiple systems, health calculation based on global deterministic failures, and health correction based on comprehensive status. The calculation method takes into account both deterministic and non-deterministic failures with unknown causes, and assigns weights to each system according to the characteristics of the hierarchical structure of the information system. In addition, it also combines the organization mode of the system, operation strategy, etc. to correct the health calculation, which can overcome the shortcomings of the current information system health evaluation and realize the system health calculation under complex conditions.

Description

Technical Field

[0001] The invention relates to a method for calculating the comprehensive health of an information system. Background Art

[0002] An information system is a man-machine integrated system composed of computer hardware, network and communication equipment, computer software, information resources, information users and rules and regulations for the purpose of processing information flow.

[0003] With the continuous development and progress of technology, especially the widespread promotion and application of various open source technologies, current information systems have shown characteristics such as complex relationships, large size, and difficult maintenance. The risk of information system abnormalities caused by various unknown failures is increasing. How to determine the operation of information systems has become an important problem faced by current information system operation and maintenance personnel. At present, the information system health evaluation method is mainly to integrate various indicators, obtain the weight of the indicators through a certain method, and then summarize the health value. The above method has the following problems:

[0004] 1. Insufficient analysis of the impact of unknown or unmonitorable parameter indicators on the health of information systems. Existing information system health calculation methods are usually based on known monitoring parameter indicators, combined with technical means such as trend prediction, and health analysis is achieved through a similar weighted average method; the impact of unknown or unmonitorable parameter indicators on the health of information systems is not considered.

[0005] 2. Insufficient analysis of the difference in the impact of a single indicator on the health of the information system caused by differences in the composition of the information system modules. The existing information system health calculation method fails to take into account the comprehensive architecture of the information system including computer hardware, network and communication equipment, computer software, information resources, etc., and cannot reflect the difference in the impact of the same indicator on the health of the information system under different information system architectures.

[0006] 3. Insufficient ability to correct the health of information systems under special conditions. Information systems are complex systems. Although the current information system health assessment method takes into account the weight differences of various parameter indicators in different scenarios, it cannot adapt to the health correction requirements of information systems under different operating modes, different application organization methods, and different backup strategies. Summary of the invention

[0007] Purpose of the invention: The technical problem to be solved by the present invention is to provide a method for calculating the comprehensive health of an information system in view of the deficiencies in the prior art, to achieve quantitative evaluation of the operating status of the information system under complex conditions, and to provide a reference for operation and maintenance personnel.

[0008] The solution to achieve the purpose of the present invention is: a method for calculating the comprehensive health of an information system, comprising the following steps:

[0009] Step 1: Perform health assessment based on abnormal impact and hierarchical weights of system architecture;

[0010] Step 2: Aggregate system health;

[0011] Step 3, calculate health based on global deterministic failures;

[0012] Step 4: Make health correction.

[0013] Step 1 includes: calculating the health of a single subsystem according to the following calculation formula:

[0014]

[0015]

[0016] Among them, Sig(x) is the judgment function, and x is the intermediate parameter;

[0017] S i represents the i-th subsystem, ScoreStep1(S i ) represents the health of the i-th subsystem;

[0018] S i,m Represents the i-th subsystem S i The decisive failure of the mth subsystem level, p is the i-th subsystem S i The number of all decisive failures; when S i,m If the decisive fault represented by i,m ) takes the value of 0, otherwise, takes the value of 1;

[0019] Weight parameter P1(S i ) represents the i-th subsystem S i The importance of the entire information system ranges from 0 to 1. n represents the number of subsystems under the information system;

[0020] e i,k Represents the i-th subsystem S i The kth non-deterministic fault of i,k ) represents fault e i,k The importance of subsystem health calculation, P2(e i,k )<1, q represents the i-th subsystem S i There are q types of non-deterministic faults.

[0021] In step 1, the importance of non-deterministic failures is derived and calculated using a fuzzy algorithm. The impact of non-deterministic failures on the system is set to t levels, where t is an integer greater than 1. According to the scale of the information system, t can be between 20 and 100. The domain of the failure weight of non-deterministic failures is A = {h1,h2,…,h t}, h i is the mapping value of the weight of the i-th indicator, i ranges from 1 to t, 0 <h i <1; Set the system to have u indicators, fault e i,k The fault characterization is B(e i,k )=[g1,g2,…,g u ], where g x It represents the measurement value mapping value of the xth indicator, x ranges from 1 to u. After actual monitoring and measurement, the fuzzy relationship is calculated by fuzzy reasoning theory, R is a matrix with u rows and t columns, then the fault e i,k The fuzzy output A′(e i,k )for:

[0022] A′(e i,k )=B(e i,k ).

[0023] Among them. represents the product of matrices;

[0024] Make a decision through the maximum membership method to determine the fault e i,k The fault weight P2(e i,k );

[0025] W(e i,k ) represents the i-th subsystem S i The impact of the kth non-deterministic fault is calculated if the i-th subsystem S i If the kth non-deterministic fault of does not occur, then W(e i,k ) is 0, otherwise it is 1;

[0026] The i-th subsystem S i The abnormality degree value of all non-deterministic faults is expressed as

[0027] Step 2 includes: aggregating the health of each subsystem by summing up, the formula is as follows:

[0028]

[0029] Among them, b j represents the information system, n represents the number of subsystems under the information system, ScoreStep2(b j ) represents the aggregated information system b j health.

[0030] Step 3 includes: using the following formula to calculate the health score ScoreStep3(b j ):

[0031] ScoreStep3(b j )=F(b j )×ScoreStep2(S i )

[0032] Among them, the function F(b j ) takes a value of 0 or 1. j When a decisive failure occurs in the information system represented by j ) value is 0 if yes, otherwise it is 1.

[0033] Step 4 includes: introducing a correction factor to adjust the health of different organizational methods and backup strategies. The specific formula is as follows:

[0034] hp=ScoreStep3(b j )+facotor(b j )

[0035] Among them, facotor(b j ) represents system b j The corresponding correction coefficient, hp represents the corrected health.

[0036] In step 4, the correction factor is calculated using the following steps:

[0037] Step 4-1, obtain the system running configuration:

[0038] Obtain information system operation mode and backup strategy from the prefabricated information system configuration library;

[0039] Step 4-2, obtain the correction coefficient calculation strategy:

[0040] Loading the calculation strategy table of correction coefficients from the prefabricated information system correction coefficient database;

[0041] Step 4-3, calculate the correction factor:

[0042] According to the operation configuration of the information system, the calculation strategy table is traversed to obtain the corresponding correction coefficient.

[0043] The method of the present invention comprehensively considers all monitoring parameters and decisive failure factors, and formulates health correction strategies for information systems under different operating modes, different application organization methods, and different backup strategies, so as to obtain more accurate and reasonable information system health under complex conditions.

[0044] Compared with the prior art, the present invention has the following advantages: it takes into account both deterministic and non-deterministic failures with unknown causes, and assigns weights to each system according to the characteristics of the information system hierarchical structure. In addition, it also combines the system's organizational mode, operating strategy, etc. to correct the health calculation. Not only does it consider a wide range of factors and can be applied to more complex scenarios, but the accuracy of the health is also higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0046] Figure 1 It is a schematic diagram of the method of the present invention.

[0047] Figure 2 It is a schematic diagram of the correction coefficient calculation method. DETAILED DESCRIPTION

[0048] The present invention provides a method for calculating the comprehensive health of an information system, which performs the following steps, and the flowchart is referenced as follows: Figure 1 :

[0049] Step 1: Health assessment based on abnormal impact and system architecture hierarchical weights. Calculate the health of a single subsystem according to the following calculation formula:

[0050]

[0051]

[0052] Among them, S i represents the i-th subsystem, S i,m represents the decisive failure at the mth subsystem level of the ith subsystem. Since the system cannot operate normally when any decisive failure occurs, such as power failure or network disconnection, all decisive failures are aggregated in the form of product. i,m If the decisive fault represented by i,m ) takes the value of 0, otherwise, it takes the value of 1. Considering the different importance of different subsystems in the entire information system, the weight parameter P1(x) is introduced to represent the weight of subsystem S i The importance of E(S i,m ) is multiplied. In addition, for non-deterministic faults, since the exact cause of the fault cannot be determined, the influence of this part of the fault abnormality is removed from the health of the subsystem using subtraction. i,k represents the kth non-deterministic fault of the ith subsystem, W(e i,k) is the occurrence of the kth non-deterministic fault of the i-th subsystem. If it occurs, it is 1, otherwise it is 0. Since the same fault has different impacts on the health status of different subsystems, the fault weight P2(e i,k ) represents the importance of the fault in the calculation of the health of the subsystem P2(e i,k )<1, q represents the i-th subsystem S i There are q types of non-deterministic faults. The importance of non-deterministic faults can be derived and calculated using fuzzy algorithms. Assuming that the impact of non-deterministic faults on the system is divided into t levels (t>1), the domain of the fault weight of non-deterministic faults is A = {h1,h2,…,h t}, h i (i ranges from 1 to t) is the mapping value of the weight (0 <h i <1); Assume that the system has u indicators, fault e i,k The fault can be characterized as B(e i,k )=[g1,g2,…,g u ], where g x Represents the measurement value mapping value of the xth indicator (i ranges from 1 to u). After actual monitoring and measurement, the fuzzy relationship is calculated through fuzzy reasoning theory. R is a matrix with u rows and t columns. Then the fault e i,k The fuzzy output A′(e i,k )for:

[0053] A′(e i,k )=B(e i,k ).

[0054] Through the maximum membership method, make a decision and determine the fault i,k The fault weight P2(e i,k ).

[0055] Then the abnormality degree value of the non-deterministic fault of this subsystem is expressed as Considering that the result of subtraction may be negative, the Sig(x) function is used to normalize the value to a non-negative number.

[0056] Among them, determine the fault e i,k The fault weight P2(e i,k ) is as follows:

[0057] The weight domain of non-deterministic failures of a certain information system is {1, 2, 3, 4, 5}, the corresponding weights are (0.1, 0.25, 0.63, 0.76, 0.82}, and the system index value is (1, 2, 3, 4, 5}. Through actual detection and expert evaluation, the fuzzy quantities of "failure 1" and "failure probability 1" are expressed as

[0058]

[0059]

[0060] It is known that "fault n" can be expressed as:

[0061]

[0062] To infer the fault weight corresponding to "fault n", first determine the fuzzy relationship R:

[0063]

[0064] Then fuzzy reasoning is performed, and the fuzzy quantity corresponding to "fault n" is:

[0065]

[0066] The decision was made based on the maximum membership method, and the corresponding weight was 0.82.

[0067] Step 2: Aggregation of the health of multiple systems. Based on the health of each subsystem calculated in step 1, the health of each subsystem is aggregated by summing them up. The specific formula is as follows:

[0068]

[0069] Among them, b j represents the information system, and n represents the number of subsystems under the information system.

[0070] Step 3: Calculate the health degree based on global deterministic failures. The preliminary information system health degree calculated in step 2 is multiplied by the global deterministic failure factor to handle failures such as global power outages and network outages that affect the system status and have a decisive impact on the normal operation of the information system. The specific formula is as follows:

[0071] ScoreStep3(b j )=F(b j )×ScoreStep2(S i )

[0072] The value of F(x) is 0 or 1. When a decisive failure occurs, the value is 0, otherwise the value is 1.

[0073] Step 4, health correction based on comprehensive status. The health calculated up to step 3 only considers the health status of each subsystem. However, under different organizational methods and backup strategies, even with the same subsystem health status, the overall health value of the information system should be different. Taking the backup strategy as an example, a good backup strategy will enhance the overall disaster recovery capability of the system, so its overall health score should be higher. Therefore, a correction coefficient is introduced to adjust the health under different organizational methods and backup strategies. The specific formula is as follows:

[0074] hp=ScoreStep3(b j )+facotor(b j )

[0075] Among them, facotor(b j Represents system b j The corresponding correction coefficient. The correction coefficient is strongly related to the system's organization, backup strategy, etc., so its calculation method adopts a multiple judgment method. The specific process is as follows Figure 2 As shown, it mainly includes the following three steps:

[0076] (1) Obtaining system running configuration

[0077] Obtain information system operation modes, backup strategies, etc. from the pre-made information system configuration library.

[0078] (2) Obtaining the Correction Coefficient Calculation Strategy

[0079] The calculation strategy table of the correction coefficient is loaded from the prefabricated information system correction coefficient database. The calculation strategy is stored in the database in the format of (system operation configuration, correction coefficient value), which is convenient for the subsequent adjustment and expansion of the calculation strategy.

[0080] (3) Calculation of correction coefficient

[0081] According to the operation configuration of the information system, the calculation strategy table is traversed to obtain the corresponding correction coefficient.

[0082] Based on the integration of the formulas of the above four steps, the complete formula for calculating the health of the information system can be obtained as follows:

[0083]

[0084] Among them, facotor(b j ) is implemented as follows:

[0085] (1) Obtaining system running configuration

[0086] Obtain information system operation modes, backup strategies, etc. from the pre-made information system configuration library.

[0087] An example of an information system configuration library is as follows:

[0088]

[0089] (2) Obtaining the Correction Coefficient Calculation Strategy

[0090] The calculation strategy table of the correction coefficient is loaded from the prefabricated information system correction coefficient database. The calculation strategy is stored in the database in the format of (system operation configuration, correction coefficient value), which is convenient for the subsequent adjustment and expansion of the calculation strategy.

[0091] An example of an information system correction factor database is as follows:

[0092]

[0093] (3) Calculation of correction coefficient

[0094] According to the operation configuration of the information system, the calculation strategy table is traversed to obtain the corresponding correction coefficient.

[0095] The calculation strategy is shown below:

[0096] Assumptions The current operation mode is "multi-system joint operation" and the backup mode is "real-time backup". According to the table above, the correction coefficient is 0.6. Then facotor(b j )=(1-T(b j ))×0.6.

[0097] The present invention provides a method for calculating the comprehensive health of an information system. There are many methods and ways to implement the technical solution. The above is only a preferred implementation of the present invention. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. A method for calculating the comprehensive health of an information system, characterized in that: The steps include: Step 1: Perform health assessment based on abnormal impact and hierarchical weights of system architecture; Step 2: Aggregate system health; Step 3, calculate health based on global deterministic failures; Step 4, make health correction; In step 1, the importance of non-deterministic faults is calculated by fuzzy algorithm. The impact of non-deterministic faults on the system is divided into t levels, where t is an integer greater than 1. The domain of the fault weight of non-deterministic faults is A = {h1, h2, …, h t }, h i is the mapping value of the weight of the i-th indicator, i ranges from 1 to t, 0 <h i <1; Set the system to have u indicators, fault e i,k The fault characterization is B(e i,k )=[g1,g2,…,g u ], where g x It represents the measurement value mapping value of the xth indicator, x ranges from 1 to u. After actual monitoring and measurement, the fuzzy relationship is calculated by fuzzy reasoning theory, R is a matrix with u rows and t columns, then the fault e i,k The fuzzy output A′(e i,k )for: A′(e i,k )=B(e i,k )。R Among them. represents the product of matrices; Make a decision through the maximum membership method to determine the fault e i,k The fault weight P2(e i,k ); W(e i,k ) represents the i-th subsystem S i The impact of the kth non-deterministic fault is calculated if the i-th subsystem S i If the kth non-deterministic fault of does not occur, then W(e i,k ) is 0, otherwise it is 1; The i-th subsystem S i The abnormality degree value of all non-deterministic faults is expressed as Step 2 includes: aggregating the health of each subsystem by summing up, the formula is as follows: Among them, b j represents the information system, n represents the number of subsystems under the information system, ScoreStep2(b j ) represents the aggregated information system b j health; Step 3 includes: using the following formula to calculate the health score ScoreStep3(b j ): ScoreStep3(b j )=F(b j )×ScoreStep2(S i ) Among them, the function F(b j ) takes a value of 0 or 1. j When a decisive failure occurs in the information system represented by j ) value is 0, otherwise it is 1; Step 4 includes: introducing a correction factor to adjust the health of different organizational methods and backup strategies. The specific formula is as follows: hp=ScoreStep3(b j )+facotor(b j ) Among them, facotor(b j ) represents system b j The corresponding correction coefficient, hp represents the corrected health.

2. The method according to claim 1, characterized in that Step 1 includes: calculating the health of a single subsystem according to the following calculation formula: Among them, Sig(x) is the judgment function, and x is the intermediate parameter; S i represents the i-th subsystem, ScoreStep1(S i ) represents the health of the i-th subsystem; S i,m Represents the i-th subsystem S i The decisive failure of the mth subsystem level, p is the i-th subsystem S i The number of all decisive failures; when S i,m If the decisive fault represented by i,m ) takes the value of 0, otherwise, takes the value of 1; Weight parameter P1(S i ) represents the i-th subsystem S i The importance of the entire information system ranges from 0 to 1. n represents the number of subsystems under the information system; e i,k Represents the i-th subsystem S i The kth non-deterministic fault of i,k ) represents fault e i,k The importance of subsystem health calculation, P2(e i,k )<1, q represents the i-th subsystem S i There are q types of non-deterministic faults.

3. The method according to claim 2, characterized in that In step 4, the correction factor is calculated using the following steps: Step 4-1, obtain the system running configuration: Obtain the operation mode and backup strategy of the information system from the prefabricated information system configuration library; Step 4-2, obtain the correction coefficient calculation strategy: Loading the calculation strategy table of correction coefficients from the prefabricated information system correction coefficient database; Step 4-3, calculate the correction factor: According to the operation configuration of the information system, the calculation strategy table is traversed to obtain the corresponding correction coefficient.

Images