System problem positioning method and device, readable storage medium and program product

Abstract

The invention discloses a system problem positioning method and device, a readable storage medium and a program product, and relates to the technical field of computer application, and the method comprises the steps: pre-constructing a problem knowledge base, after obtaining a current system feature, enabling one or more features similar to the current system feature to possibly exist in a plurality of dimension features of the problem knowledge base, searching similar dimension features of the current system features from a question knowledge base; according to the method, particle swarm sub-group initialization is carried out according to searched similar dimension features, iterative learning is carried out by using each sub-group obtained by sub-group initialization and the similar dimension features corresponding to each sub-group, and system problem positioning is realized through the convergence state of each sub-group, so that the problem that a solution is searched from a network manually according to a system problem phenomenon is solved; the technical problems of low credibility, low system problem positioning efficiency and low system security in the prior art are solved, and the technical effects of high credibility, high system problem positioning efficiency and high system security are achieved.

Description

Technical Field

[0001] This application relates to the field of computer application technologies, and particularly to a system problem localization method, device, readable storage medium, and program product. Background Art

[0002] The operating system is the basic software of a computer, and all user software runs on top of the operating system. When problems occur in the system, especially in complex business systems, problem troubleshooting is very time-consuming.

[0003] There may be many reasons for operating system problems, including hardware failures, software conflicts, configuration errors, resource shortages, etc. The currently commonly used system problem localization method is for manual workers to search for solutions on the Internet based on system problem phenomena, with low credibility, low system problem localization efficiency, and low system security. Summary of the Invention

[0004] This application provides a system problem localization method to at least solve the problems in the related technologies.

[0005] This application provides a system problem localization method, including:

[0006] Obtain the current system characteristics, and search for similar dimensional characteristics of the current system characteristics in a pre-constructed problem knowledge base;

[0007] Initialize subgroups of the particle swarm according to the found similar dimensional characteristics to obtain each subgroup;

[0008] Use each subgroup to perform iterative learning according to each similar dimensional characteristic to obtain the convergence state of each subgroup;

[0009] Locate the system problem according to the convergence state of each subgroup.

[0010] This application also provides a system problem localization device, including: a memory for storing a computer program; a processor for implementing the steps of any of the above system problem localization methods when executing the computer program.

[0011] This application also provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any of the above system problem localization methods.

[0012] This application also provides a computer program product, including a computer program, and the computer program, when executed by a processor, implements the steps of any of the above system problem localization methods.

[0013] With this application, since a problem knowledge base is pre-constructed, and the problem knowledge base stores features in multiple dimensions when the system has problems. After obtaining the current system features, one or more features similar to the current system features may exist among the multiple-dimensional features of the problem knowledge base. Search for the similar-dimensional features of the current system features in the problem knowledge base. Then, initialize the subgroups of the particle swarm according to the found similar-dimensional features. Furthermore, perform iterative learning using each subgroup obtained from the subgroup initialization and the corresponding similar-dimensional features of each subgroup, and statistically analyze the convergence states of the iterative learning of each subgroup. The convergence state of each subgroup reflects the degree of consistency between the features of the corresponding dimension of each subgroup and the current system features. The system problem is located through the convergence states of each subgroup. Therefore, it is possible to solve the technical problems of low credibility, low system problem location efficiency, and low system security when manually searching for solutions on the Internet based on system problem phenomena, and achieve the technical effects of high credibility, high system problem location efficiency, and high system security. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1 It is a flowchart of an implementation of a system problem location method provided by an embodiment of the present application;

[0016] Figure 2 It is a flowchart of an implementation of another system problem location method provided by an embodiment of the present application;

[0017] Figure 3 It is a structural block diagram of a system problem location device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present application.

[0019] It should be noted that in the description of this application, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. The terms "first", "second", etc. in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0020] To enable those skilled in the art of this technology to better understand the solution of this application, the following further detailed description of this application will be given in conjunction with the accompanying drawings and specific embodiments.

[0021] An embodiment of this application provides a method for locating system problems. The method will be described in detail in combination with the execution process of the method for locating system problems.

[0022] See Figure 1 , Figure 1 which is the flowchart of the implementation of a method for locating system problems provided by an embodiment of this application. The method may include the following steps:

[0023] S101: Obtain the current system characteristics and search for similar dimensional characteristics of the current system characteristics in a pre-constructed problem knowledge base.

[0024] When system problem location is required, obtain the current system characteristics. For example, the problem phenomenon, resource occupancy, function abnormality, system status, etc. of the current system can be obtained.

[0025] Pre-construct a problem knowledge base, which stores characteristics of multiple dimensions when the system has problems. After obtaining the current system characteristics, search for similar dimensional characteristics of the current system characteristics in the pre-constructed problem knowledge base.

[0026] As shown in Table 1, Table 1 is the structure table of a problem knowledge base in an embodiment of this application.

[0027] Table 1

[0028]

[0029] It can be seen from Table 1 that the problem knowledge base can be divided into four major categories: business layer, application layer, kernel layer, and hardware layer. Each major category is composed of subcategories such as Category 1, Category 2, Category 3, Category 4,..., Category N, and each subcategory corresponds to a dimension.

[0030] In a specific embodiment of this application, searching for similar dimensional characteristics of the current system characteristics in the pre-constructed problem knowledge base may include the following steps:

[0031] Search for similar dimensional features of the current system characteristics from the business layer, application layer, kernel layer, and hardware layer included in the problem knowledge base.

[0032] As shown in Table 1, the problem knowledge base contains problem characteristics in four major categories: the business layer, application layer, kernel layer, and hardware layer. After obtaining the current system characteristics, search for similar dimensional features of the current system characteristics from the business layer, application layer, kernel layer, and hardware layer included in the problem knowledge base. By setting the problem characteristics of multiple layers in the problem knowledge base and searching for similar dimensional features of the current system characteristics from multiple layers, fine-grained search of problem characteristics is achieved, thereby improving the accuracy of problem location.

[0033] S102: Initialize subgroups of the particle swarm according to the found similar dimensional features to obtain each subgroup.

[0034] After finding the similar dimensional features of the current system characteristics, initialize subgroups of the particle swarm according to the found similar dimensional features to obtain each subgroup. By setting multiple subgroups, each subgroup is optimized separately with the problem characteristics of the corresponding dimension as the learning target.

[0035] The Particle Swarm Optimization (PSO) algorithm is an optimization algorithm based on swarm intelligence and is widely used in various optimization problems due to its simplicity, ease of implementation, and strong global search ability.

[0036] In the particle swarm optimization algorithm, each problem is regarded as a particle in the search space. Each particle has a position and a velocity in the search space. The position of the particle represents the current problem, and the velocity determines the moving direction and distance of the particle in the search space. Each particle will record the best solution found, called the personal best (pBest). At the same time, the entire particle swarm will record the set of the best solutions found by all particles, called the global best (gBest).

[0037] S103: Use each subgroup to perform iterative learning according to each similar dimensional feature to obtain the convergence state of each subgroup.

[0038] After initializing each subgroup, use each subgroup to perform iterative learning according to each similar dimensional feature, and separately count the convergence of each subgroup to obtain the convergence state of each subgroup.

[0039] S104: Locate system problems according to the convergence state of each subgroup.

[0040] After obtaining the convergence states of each subgroup, system problem localization is performed according to the convergence states of each subgroup. For example, the subgroup with the fastest convergence can be found based on the convergence states of each subgroup, and further learning is carried out using the subgroup with the fastest convergence. There is a corresponding relationship between the positions of the particles during the learning process and the system problem. Therefore, system problem localization is performed according to the convergence positions of each particle learned.

[0041] Through this application, since a problem knowledge base is pre-constructed, and the problem knowledge base stores features in multiple dimensions when the system has problems. After obtaining the current system features, there may be one or more features similar to the current system features among the multiple-dimensional features of the problem knowledge base. The similar-dimensional features of the current system features are searched for in the problem knowledge base. Then, the subgroup initialization of the particle swarm is performed according to the found similar-dimensional features. Furthermore, iterative learning is carried out using each subgroup obtained by the subgroup initialization and the corresponding similar-dimensional features of each subgroup, and the convergence states of the iterative learning of each subgroup are statistically analyzed. The convergence states of each subgroup reflect the consistency level between the features of the corresponding dimensions of each subgroup and the current system features. System problem localization is achieved through the convergence states of each subgroup. Therefore, the technical problems of low credibility, low system problem localization efficiency, and low system security caused by manually searching for solutions on the Internet based on the system problem phenomenon can be solved, and the technical effects of high credibility, high system problem localization efficiency, and high system security are achieved.

[0042] See Figure 2 , Figure 2 which is the flowchart of another system problem localization method provided by an embodiment of this application. This method may include the following steps:

[0043] S201: Obtain one or more of the problem phenomenon features, resource occupancy features, function abnormality features, and system status features of the current system.

[0044] When system problem localization is required, one or more of the problem phenomenon features, resource occupancy features, function abnormality features, and system status features of the current system are obtained. The specific features obtained depend on whether the current features can characterize that the system has problems, and the features that can characterize that the system currently has problems are obtained.

[0045] S202: Determine the obtained feature items as the current system features.

[0046] After obtaining one or more of the problem symptom characteristics, resource occupancy characteristics, function abnormality characteristics, and system status characteristics of the current system, the obtained characteristic items are determined as the current system characteristics. By obtaining one or more of the problem symptom characteristics, resource occupancy characteristics, function abnormality characteristics, and system status characteristics, characteristics that can characterize problems in the system can be obtained from multiple aspects, further realizing a fine-grained division of system problem positioning.

[0047] S203: Decompose the current system characteristics to obtain each sub-characteristic.

[0048] After obtaining the current system characteristics, decompose the current system characteristics to obtain each sub-characteristic. As shown in Table 1, the current system characteristics can be decomposed into characteristics of multiple dimensions, that is, corresponding to multiple types of characteristics in Table 1.

[0049] S204: Search for similar-dimension characteristics of each sub-characteristic from the problem knowledge base respectively.

[0050] After decomposing the current system characteristics into each sub-characteristic, search for similar-dimension characteristics of each sub-characteristic from the problem knowledge base respectively. As shown in Table 1, search for similar-dimension characteristics of each sub-characteristic from each type. By decomposing the current system characteristics, a large problem is decomposed into multiple target sub-problems, and the problem positioning efficiency is further improved by optimizing and positioning respectively through the multi-dimensional particle swarm algorithm.

[0051] S205: Initialize the subgroups of the particle swarm according to the found similar-dimension characteristics to obtain each subgroup.

[0052] S206: Initialize the velocities of each subgroup to obtain the initial velocities corresponding to each particle in each subgroup respectively.

[0053] After initializing to obtain each subgroup, initialize the velocities of each particle in each subgroup to obtain the initial velocities corresponding to each particle in each subgroup respectively. By initializing the velocities of each particle in each subgroup, searching for system problems using multiple different subgroups is realized, thereby realizing a fine-grained positioning of system problems.

[0054] S207: Search for the initial entropy values corresponding to each similar-dimension characteristic from the characteristic entropy value table.

[0055] Pre-create a characteristic entropy value table, which pre-stores the corresponding relationship between each dimension characteristic and each entropy value. After finding the similar-dimension characteristics of each sub-characteristic from the problem knowledge base, search for the initial entropy values corresponding to each similar-dimension characteristic from the characteristic entropy value table.

[0056] As shown in Table 2, Table 2 is a comparison table of the relationship between various dimensional features and various entropy values in the embodiments of the present application.

[0057] Table 2

[0058]

[0059] The initial entropy values corresponding to the respective similar dimensional features are obtained by looking up in the feature entropy value table.

[0060] S208: Use each subgroup to perform iterative learning according to each initial entropy value to obtain the convergence state of each subgroup.

[0061] After obtaining the initial entropy values corresponding to the respective similar dimensional features, use each subgroup to perform iterative learning according to each initial entropy value to obtain the convergence state of each subgroup. By looking up the initial entropy values corresponding to the respective similar dimensional features, accurate and unique identification of the respective similar dimensional features is achieved, facilitating the accurate positioning of the position of the particle during the iterative process, thereby improving the accuracy of system problem positioning.

[0062] In a specific implementation manner of the present application, step S208 may include the following steps:

[0063] Step 1: Determine each initial velocity as each current velocity;

[0064] Step 2: Determine the initial positions of the particles in each subgroup according to each initial entropy value, and determine each initial position as each current position;

[0065] Step 3: Update the velocities of the particles according to the current positions and current velocities of the particles to obtain new current velocities;

[0066] Step 4: Update the current position according to the current position and current velocity to obtain a new current position;

[0067] Step 5: Determine the convergence state of each subgroup according to each current position.

[0068] For ease of description, the above five steps can be combined for explanation.

[0069] In the process of using each subgroup to perform iterative learning according to each initial entropy value, determine each initial velocity as each current velocity, determine the initial positions of the particles in each subgroup according to each initial entropy value, and determine each initial position as each current position, and update the velocities of the particles according to the current positions and current velocities of the particles to obtain new current velocities. For example, the particle velocity can be updated by the following formula:

[0070] ;

[0071] Wherein, It represents the weight of the part where the next action of the particle comes from its own experience, which is the acceleration weight that pushes the particle towards the individual optimal position (pbest); It represents the weight of the part where the next action of the particle comes from the experience of other particles, which is the acceleration weight that pushes the particle towards the global optimal position (gbest); It represents the inertia weight, which is used to control the degree of inheritance of the particle's own speed. A larger inertia weight allows the particle to fly farther in the solution space, which is helpful for global search, while a smaller inertia weight enables the particle to conduct a more refined search in the local area; It represents the new current speed of the i-th particle; It represents the current speed of the i-th particle; It represents the random number function; It represents the current position of the particle in the previous iteration; It represents the current position of the particle; It represents the optimal position of the i-th particle so far; It represents the optimal position of the entire subgroup.

[0072] After obtaining the current speed through update, the current position is updated based on the current position and the current speed to obtain the new current position. For example, the particle position can be updated through the following formula:

[0073] ;

[0074] Among them, It represents the current position of the i-th particle, It represents the new current position of the i-th particle; It represents the new current speed of the i-th particle.

[0075] The convergence state of each subgroup is determined according to each current position.

[0076] By continuously iterating the speed and position of each particle in each subgroup, the convergence state of each particle position is obtained, and then the convergence state of each subgroup is statistically obtained, which improves the accuracy of the statistical convergence state of each subgroup, and further improves the accuracy of system problem location.

[0077] In a specific implementation manner of the present application, updating the speed of each particle according to the current position and the current speed of each particle may include the following steps:

[0078] Step 1: Obtain the number of subgroups of each subgroup obtained by initialization;

[0079] Step 2: Generate the random factor to be added according to the number of subgroups;

[0080] Step 3: Update the velocity of each particle according to the current position, current velocity, and random factor of each particle.

[0081] For convenience of description, the above three steps can be combined for illustration.

[0082] When updating the velocity of each particle, obtain the number of subgroups of each subgroup obtained by initialization, generate a random factor to be added according to the number of subgroups, and update the velocity of each particle according to the current position, current velocity, and random factor of each particle. For example, the particle velocity can be updated by the following formula:

[0083] ;

[0084] where represents a random number between 0 and k, and k represents the number of subgroups.

[0085] By adding a random factor during the particle velocity update process to simulate the change in velocity caused by resource or network jitter in the system, the robustness is improved, the accuracy of the particle velocity update result is increased, and thus the accuracy of system problem location is improved.

[0086] In a specific embodiment of the present application, updating the current position according to the current position and current velocity may include the following steps:

[0087] Step 1: Number each subgroup to obtain the subgroup numbers of each subgroup;

[0088] Step 2: Determine whether there is a subgroup in each subgroup that has been in the fastest convergence for continuously greater than or equal to a preset number of times within a preset time period. If so, execute Step 3; if not, execute Step 5;

[0089] Step 3: Obtain the subgroup number of the subgroup with the fastest convergence, and determine the subgroup number of the subgroup with the fastest convergence as the target number;

[0090] Step 4: Update the current position according to the current position, current velocity, number of subgroups, and target number;

[0091] Step 5: Update the current position according to the current position and current velocity.

[0092] For convenience of description, the above five steps can be combined for illustration.

[0093] When updating the positions of the particles in the particle swarm, number each subgroup to obtain the subgroup numbers of each subgroup, and determine whether there is a subgroup in each subgroup that has been in the fastest convergence for continuously greater than or equal to a preset number of times within a preset time period. If so, obtain the subgroup number of the subgroup with the fastest convergence. For example, the subgroup number of the subgroup with the fastest convergence can be obtained by the following formula:

[0094] ;

[0095] Among them, represents the number of the subgroup that is continuously greater than or equal to m times and is in the fastest convergence within a preset time period selected from k subgroups, and m can be set to be greater than 3. represents the subgroup number of the subgroup with the fastest convergence.

[0096] When it is determined that there is a subgroup in each subgroup that is continuously greater than or equal to the preset number of times and is in the fastest convergence within a preset time period, the subgroup number of the subgroup with the fastest convergence is determined as the target number, and the current position is updated according to the current position, the current speed, the number of subgroups, and the target number. For example, the current position can be updated through the following formula:

[0097] ;

[0098] Among them, t is the time duration from the completion of the previous iteration to the completion of the current iteration.

[0099] When it is determined that there is no subgroup in each subgroup that is continuously greater than or equal to the preset number of times and is in the fastest convergence within a preset time period, the current position is updated according to the current position and the current speed. For example, the current position can be updated through the following formula:

[0100] ;

[0101] After determining the subgroup with the fastest convergence, by adding the parameter of the current convergence value to the current position of the particle, it helps to quickly find the optimal position, thereby improving the system problem location efficiency.

[0102] In a specific implementation manner of the present application, updating the current position according to the current position, the current speed, the number of subgroups, and the target number may include the following steps:

[0103] Step 1: Obtain the iteration time of the current iteration;

[0104] Step 2: Update the current position according to the current position, the current speed, the iteration time, the number of subgroups, and the target number.

[0105] For convenience of description, the above two steps can be combined for description.

[0106] When updating the position of the particles in the subgroup, obtain the iteration time of the current iteration, and update the current position according to the current position, the current speed, the iteration time, the number of subgroups, and the target number. By combining the iteration time for particle position update, the accuracy of particle position update is improved, and thus the accuracy of problem location is improved.

[0107] S209: According to the convergence states of each subgroup, determine whether there is a subgroup that is in optimal convergence when reaching each preset number of iterations. If so, execute step S210; if not, execute step S211.

[0108] After obtaining the convergence states of each subgroup, according to the convergence states of each subgroup, determine whether there is a subgroup that is in optimal convergence when reaching each preset number of iterations. For example, determine whether there is a subgroup that is in optimal convergence when learning 1000 times, 1500 times, and 2000 times. If so, it means there is a subgroup that always has a convergence advantage, and execute step S210; if not, it means there is no subgroup that always has a convergence advantage, and execute step S211.

[0109] S210: Determine the subgroup that is in optimal convergence when reaching each preset number of iterations as the target subgroup.

[0110] When it is determined that there is a subgroup that is in optimal convergence when reaching each preset number of iterations among each subgroup, it means there is a subgroup that always has a convergence advantage. Determine the subgroup that is in optimal convergence when reaching each preset number of iterations as the target subgroup.

[0111] S211: Determine the subgroup with the fastest convergence when reaching the first number of iterations as the target subgroup.

[0112] When it is determined that there is no subgroup that is in optimal convergence when reaching each preset number of iterations among each subgroup, determine the subgroup with the fastest convergence when reaching the first number of iterations as the target subgroup.

[0113] S212: Locate the system problem according to the target subgroup.

[0114] After determining the target subgroup, there is interaction between subgroups. The target subgroup with fast convergence will broadcast and notify other subgroups, and other subgroups will stop learning. Then, locate the system problem according to the target subgroup. Thus, select the subgroup most relevant to the current system problem location from multiple subgroups for system problem location, avoiding further learning of multiple subgroups that have little relevance to the system problem location, improving resource utilization, greatly improving the accuracy of system problem location, and improving the efficiency of system problem location. By adopting the improved particle swarm intelligent learning algorithm, it can be compared with the phenomena of the current system, and learn in combination with the dynamic parameter information of the system, and finally locate to a certain problem classification provided by the problem knowledge base, helping users to more accurately locate the cause of the system problem.

[0115] In a specific embodiment of the present application, step S212 may include the following steps:

[0116] Step 1: Use the target subgroup to perform iterative learning based on the similarity dimension features corresponding to the target subgroup;

[0117] Step 2: When the second iteration count is reached, locate the system problem according to the convergence status of each particle in the target subgroup.

[0118] For ease of description, the above two steps can be combined for explanation.

[0119] After determining the target subgroup, use the target subgroup to perform iterative learning based on the similarity dimension features corresponding to the target subgroup. When the second iteration count is reached, locate the system problem according to the convergence status of each particle in the target subgroup. During the iterative learning process of the target subgroup, it will ultimately approach a subclass of a certain problem infinitely, thereby generally finding the specific cause of the problem, improving the efficiency of system problem location, and improving the accuracy of system problem location.

[0120] It should be noted that the first and second in the first iteration count and the second iteration count do not have a size or sequence relationship. They are only for the purpose of jointly learning each subgroup and only learning the target subgroup.

[0121] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0122] The embodiments of the present application also provide a system problem location device, which may include:

[0123] A similarity dimension feature search module 31, configured to obtain the current system features and search for the similarity dimension features of the current system features from a pre-constructed problem knowledge base;

[0124] A subgroup obtaining module 32, configured to perform subgroup initialization of the particle swarm according to the found similarity dimension features to obtain each subgroup;

[0125] A convergence status obtaining module 33, configured to use each subgroup to perform iterative learning according to each similarity dimension feature to obtain the convergence status of each subgroup;

[0126] A system problem location module 34, configured to locate the system problem according to the convergence status of each subgroup.

[0127] Through this application, since a problem knowledge base is pre-constructed, and the problem knowledge base stores features in multiple dimensions when the system has problems. After obtaining the current system features, one or more features similar to the current system features may exist among the features in multiple dimensions of the problem knowledge base. Search for the similar dimensional features of the current system features from the problem knowledge base. Then, initialize the subgroups of the particle swarm according to the found similar dimensional features. Furthermore, perform iterative learning using each subgroup obtained by subgroup initialization and the corresponding similar dimensional features of each subgroup, and count the convergence states of the iterative learning of each subgroup. The convergence state of each subgroup reflects the degree of consistency between the features of the corresponding dimension of each subgroup and the current system features. The system problem is located through the convergence states of each subgroup. Therefore, it is possible to solve the technical problems of low credibility, low efficiency in system problem location, and low system security when manually searching for solutions on the Internet based on system problem phenomena, and achieve the technical effects of high credibility, high efficiency in system problem location, and high system security.

[0128] In a specific embodiment of the present application, the similar dimensional feature search module 31 includes:

[0129] A feature acquisition sub-module, configured to acquire one or more features among the problem phenomenon features, resource occupancy features, function abnormality features, and system current status features of the current system;

[0130] A current system feature determination sub-module, configured to determine the acquired feature items as the current system features.

[0131] In a specific embodiment of the present application, the similar dimensional feature search module 31 includes:

[0132] A sub-feature acquisition sub-module, configured to decompose the current system features into each sub-feature;

[0133] A similar dimensional feature search sub-module, configured to search for the similar dimensional features of each sub-feature from the problem knowledge base respectively.

[0134] In a specific embodiment of the present application, the similar dimensional feature search module 31 is specifically a module for searching for the similar dimensional features of the current system features from the business layer, application layer, kernel layer, and hardware layer included in the problem knowledge base.

[0135] In a specific embodiment of the present application, the device may further include:

[0136] An initial velocity acquisition module, configured to initialize the velocity of each subgroup after obtaining each subgroup, and obtain the initial velocity corresponding to each particle in each subgroup.

[0137] In a specific embodiment of the present application, the convergence state acquisition module 33 includes:

[0138] An initial entropy value lookup sub-module for looking up the initial entropy values corresponding to the respective similar dimensional features from a feature entropy value table;

[0139] An iterative learning sub-module for performing iterative learning by each subgroup according to each initial entropy value.

[0140] In a specific embodiment of the present application, the iterative learning sub-module includes:

[0141] A current speed determination unit for determining each initial speed as each current speed;

[0142] A current position determination unit for determining the initial positions of the particles in each subgroup according to each initial entropy value and determining each initial position as each current position;

[0143] A speed update unit for updating the speed of each particle according to the current position and current speed of each particle to obtain a new current speed;

[0144] A position update unit for updating the current position according to the current position and current speed to obtain a new current position;

[0145] A convergence state determination unit for determining the convergence state of each subgroup according to each current position.

[0146] In a specific embodiment of the present application, the speed update unit includes:

[0147] A subgroup number acquisition sub-unit for acquiring the subgroup numbers of each subgroup obtained by initialization;

[0148] A random factor addition sub-unit for generating a random factor to be added according to the subgroup number;

[0149] A speed update sub-unit for updating the speed of each particle according to the current position, current speed and random factor of each particle.

[0150] In a specific embodiment of the present application, the position update unit includes:

[0151] A subgroup number obtaining sub-unit for numbering each subgroup to obtain each subgroup number;

[0152] A judgment sub-unit for judging whether there is a subgroup in each subgroup that is in the fastest convergence state continuously greater than or equal to a preset number of times within a preset time period;

[0153] A target number determination sub-unit for, when it is determined that there is a subgroup in each subgroup that is in the fastest convergence state continuously greater than or equal to a preset number of times within a preset time period, acquiring the subgroup number of the subgroup with the fastest convergence and determining the subgroup number of the subgroup with the fastest convergence as the target number;

[0154] A position update subunit, configured to update the current position according to the current position, the current speed, the number of subgroups, and the target number.

[0155] In a specific embodiment of the present application, the position update subunit is specifically a unit that obtains the iteration time of the current iteration; and updates the current position according to the current position, the current speed, the iteration time, the number of subgroups, and the target number.

[0156] In a specific embodiment of the present application, the system problem location module 34 includes:

[0157] A judgment sub-module, configured to judge whether there is a subgroup that is in the optimal convergence state when reaching each preset iteration number according to the convergence states of each subgroup;

[0158] A target subgroup determination sub-module, configured to, when it is determined that there is a subgroup that is in the optimal convergence state when reaching each preset iteration number among each subgroup, determine the subgroup that is in the optimal convergence state when reaching each preset iteration number as the target subgroup;

[0159] A target subgroup determination sub-module, configured to, when it is determined that there is no subgroup that is in the optimal convergence state when reaching each preset iteration number among each subgroup, determine the subgroup with the fastest convergence when reaching the first iteration number as the target subgroup;

[0160] A system problem location sub-module, configured to locate the system problem according to the target subgroup.

[0161] In a specific embodiment of the present application, the system problem location sub-module includes:

[0162] An iterative learning unit, configured to perform iterative learning by using the target subgroup according to the similarity dimension features corresponding to the target subgroup;

[0163] A system problem location unit, configured to, when reaching the second iteration number, locate the system problem according to the convergence states of each particle in the target subgroup.

[0164] For the description of the features in the embodiments corresponding to the system problem location device, reference may be made to the relevant descriptions in the embodiments corresponding to the system problem location method, which will not be elaborated here one by one.

[0165] An embodiment of the present application further provides a system problem location device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above embodiments of the system problem location method.

[0166] Embodiments of the present application further provide a computer-readable storage medium storing a computer program, where the computer program is configured to execute the steps in any of the above-described embodiments of the system problem localization method when running.

[0167] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: various media such as USB flash drives, read-only memories (ROM), random access memories (RAM), external hard drives, magnetic disks, or optical discs that can store computer programs.

[0168] Embodiments of the present application further provide a computer program product, where the computer program product includes a computer program, and the steps in any of the above-described embodiments of the system problem localization method are implemented when the computer program is executed by a processor.

[0169] Embodiments of the present application further provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, and the steps in any of the above-described embodiments of the system problem localization method are implemented when the computer program is executed by a processor.

[0170] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described according to their functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Skilled artisans can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0171] The above has introduced in detail a system problem localization method, device, readable storage medium, and program product provided by the present application. Specific examples are used herein to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application. It should be noted that for those of ordinary skill in the art in the technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A method for locating a system problem, characterized in that: include: Acquire current system features, and search for similar dimension features of the current system features from a pre-built problem knowledge base; Initialize the subgroups of the particle swarm according to the similar dimensional features found to obtain each subgroup; Each subgroup is used to perform iterative learning based on the characteristics of each similar dimension to obtain the convergence state of each subgroup; System problem location is performed based on the convergence status of each subgroup.

2. The system problem locating method according to claim 1, characterized in that: Get current system characteristics, including: Obtain one or more of the following features: problem phenomenon features, resource occupancy features, function abnormality features, and system status features of the current system; The acquired feature item is determined as the current system feature.

3. The system problem locating method according to claim 1 or 2, characterized in that: Find similar dimension features of the current system features from the pre-built problem knowledge base, including: Decomposing the current system feature to obtain sub-features; The similar dimension features of each sub-feature are searched from the problem knowledge base respectively.

4. The system problem locating method according to claim 1, characterized in that: Find similar dimension features of the current system features from the pre-built problem knowledge base, including: Find similar dimensional features of the current system features from the business layer, application layer, kernel layer and hardware layer included in the problem knowledge base.

5. The system problem locating method according to claim 1, characterized in that: After obtaining each subgroup, it also includes: The velocity of each subgroup is initialized to obtain the initial velocity corresponding to each particle in each subgroup.

6. The method for locating system problems according to claim 5, characterized in that: Use each subgroup to perform iterative learning based on similar dimensional features, including: Find the initial entropy value corresponding to each similar dimension feature from the feature entropy value table; Each subgroup is used to perform iterative learning according to each initial entropy value.

7. The method for locating a system problem according to claim 6, characterized in that: Each subgroup is used to perform iterative learning according to each initial entropy value to obtain the convergence state of each subgroup, including: Determine each initial speed as each current speed; Determine the initial position of each particle in each subgroup according to each initial entropy value, and determine each initial position as each current position; Update the speed of each particle according to its current position and current speed to obtain a new current speed; Update the current position according to the current position and the current speed to obtain a new current position; The convergence state of each subgroup is determined according to each current position.

8. The system problem locating method according to claim 7, characterized in that: Update the velocity of each particle according to its current position and velocity, including: Get the number of subgroups of each subgroup initialized; Generating a random factor to be added according to the number of subgroups; The speed of each particle is updated according to the current position, current speed and the random factor of each particle.

9. The method for locating system problems according to claim 7, characterized in that: Updating the current position according to the current position and the current speed includes: Number each subgroup to obtain the subgroup number; Determine whether there is a subgroup in each subgroup that is in the fastest convergence for a preset number of times in a preset time period; If yes, then obtain the subgroup number of the subgroup that converges fastest, and determine the subgroup number of the subgroup that converges fastest as the target number; The current position is updated according to the current position, the current speed, the number of subgroups and the target number.

10. The system problem locating method according to claim 9, characterized in that: The current position is updated according to the current position, the current speed, the number of subgroups and the target number, including: Get the iteration time of this iteration; The current position is updated according to the current position, the current speed, the iteration time, the number of subgroups and the target number.

11. The method for locating a system problem according to claim 1, characterized in that: Locate system problems based on the convergence status of each subgroup, including: According to the convergence state of each subgroup, it is determined whether there is a subgroup in each subgroup that is in optimal convergence when each preset number of iterations is reached; If so, the subgroup that is in optimal convergence when reaching each preset number of iterations is determined as the target subgroup; If not, the subgroup that converges fastest when reaching the first iteration number is determined as the target subgroup; System problem location is performed according to the target subgroup.

12. The system problem locating method according to claim 11, characterized in that: Locate system problems based on the target subgroups, including: Using the target subgroup to perform iterative learning according to similar dimensional features corresponding to the target subgroup; When the second iteration number is reached, the system problem is located according to the convergence state of each particle in the target subgroup.

13. A system problem location device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the system problem locating method as claimed in any one of claims 1 to 12 when executing the computer program.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the system problem locating method according to any one of claims 1 to 12 are implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the system problem locating method according to any one of claims 1 to 12 are implemented.

Images