CN107463486B - System performance analysis method and device and server - Google Patents
System performance analysis method and device and server Download PDFInfo
- Publication number
- CN107463486B CN107463486B CN201710541436.5A CN201710541436A CN107463486B CN 107463486 B CN107463486 B CN 107463486B CN 201710541436 A CN201710541436 A CN 201710541436A CN 107463486 B CN107463486 B CN 107463486B
- Authority
- CN
- China
- Prior art keywords
- variable
- determining
- target
- network topology
- target variable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3452—Performance evaluation by statistical analysis
Abstract
The invention provides a system performance analysis method, a device and a server, wherein the method comprises the following steps: acquiring historical operating data of a system and a target variable to be analyzed; performing correlation analysis on historical operating data of the system, and determining a variable set of which the correlation with a target variable meets a preset condition; learning the target variable and each variable in the variable set, and determining the network topology between the target variable and each variable in the variable set; determining an initial condition probability table among variables in the network topology according to the network topology and the historical operation data of the system; and determining a root variable corresponding to the target variable when the running data is abnormal according to the initial condition probability table. Therefore, the root variable corresponding to the target variable when the running data is abnormal is comprehensively determined according to the correlation analysis and the network topology, the accuracy of the analysis result is improved, and the reliability of the system is improved by optimizing and updating the system according to the root variable.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a system performance analysis method, a system performance analysis device and a server.
Background
With the development of science and technology and the improvement of the technical level of computers, computer application software is widely developed.
Generally, computer application software may cause performance degradation of an application system during operation for various reasons, thereby affecting the use of a user. Taking a website as an example, http response time is a main basis for measuring the performance of the website, and in an actual production environment, the http response time may be increased under the comprehensive influence of factors such as CPU utilization, session number, thread number, memory utilization and the like, so that a user cannot smoothly access the website.
Therefore, how to quickly find the root source problem influencing the system performance and determine the key influencing factors has important significance for the maintenance of the system and the improvement of the system performance.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, a first objective of the present invention is to provide a system performance analysis method, which achieves comprehensive determination of a root variable when operating data corresponding to a target variable is abnormal according to correlation analysis and network topology, improves accuracy of an analysis result, and improves reliability of a system by performing optimization update on the system according to the root variable.
A second object of the present invention is to provide a system performance analyzer.
A third object of the present invention is to provide a server.
A fourth object of the invention is to propose a computer-readable storage medium.
A fifth object of the invention is to propose a computer program product.
In order to achieve the above object, an embodiment of a first aspect of the present invention provides a system performance analysis method, including: acquiring historical operating data of a system and a target variable to be analyzed; performing correlation analysis on the historical operating data of the system, and determining a variable set of which the correlation with the target variable meets a preset condition, wherein the variable set comprises at least one non-target variable; learning the target variable and each variable in the variable set, and determining a network topology between the target variable and each variable in the variable set; determining an initial condition probability table among variables in the network topology according to the network topology and the historical system operation data; and determining a root variable corresponding to the target variable when the running data is abnormal according to the initial condition probability table.
In a possible implementation form of the present invention, after determining the variable set whose correlation with the target variable satisfies a preset condition, the method further includes:
and updating the variable set according to the attribute characteristics of the system.
In another possible implementation form of the present invention, the determining a root variable when the running data corresponding to the target variable is abnormal includes:
respectively determining each expected value of the target variable when each variable in the variable set is in a full load state according to the initial condition probability table;
and determining a root variable when the running data corresponding to the target variable is abnormal according to each expected value of the target variable.
In another possible implementation form of the present invention, each variable in the network topology corresponds to N states, the initial conditional probability table is composed of a conditional probability matrix among the variables, the conditional probability matrix among the variables includes N × N probability values, where N is a positive integer greater than 1;
the determining, when the variables in the variable set are in a full load state, each expected value of the target variable includes:
acquiring a first conditional probability matrix between a first variable in the variable set and the target variable;
and determining the expected value of the target variable when the first variable is in a full load state according to the first conditional probability matrix.
In another possible implementation form of the present invention, after determining the initial condition probability table among the variables in the network topology, the method further includes:
determining a pressure indicator of the system;
according to the pressure indexes, respectively expanding corresponding multiples of operation data corresponding to one or more variables in the network topology for updating;
determining an updated conditional probability table among variables in the network topology by using the updated data;
and determining the state of the system under the pressure index according to the updated conditional probability table.
In another possible implementation form of the present invention, after determining the root variable when the running data corresponding to the target variable is abnormal, the method further includes: and performing optimization updating on the system according to the root variable.
The system performance analysis method of the embodiment of the invention comprises the steps of firstly, obtaining historical operation data of a system and target variables to be analyzed; then, performing correlation analysis on historical operation data of the system, and determining a variable set of which the correlation with a target variable meets a preset condition; learning the target variable and each variable in the variable set, and determining the network topology between the target variable and each variable in the variable set; determining an initial condition probability table among variables in the network topology according to the network topology and the historical operation data of the system; and finally, according to the initial condition probability table, determining a root variable corresponding to the target variable when the running data is abnormal. Therefore, the root variable corresponding to the target variable when the running data is abnormal is comprehensively determined according to the correlation analysis and the network topology, the accuracy of the analysis result is improved, and the reliability of the system is improved by optimizing and updating the system according to the root variable.
In order to achieve the above object, a second embodiment of the present invention provides a system performance analysis apparatus, including: the acquisition module is used for acquiring historical operating data of the system and target variables to be analyzed; the first determination module is used for performing correlation analysis on the historical operating data of the system and determining a variable set of which the correlation with the target variable meets a preset condition, wherein the variable set comprises at least one non-target variable; the second determining module is used for learning the target variable and each variable in the variable set and determining the network topology between the target variable and each variable in the variable set; a third determining module, configured to determine an initial condition probability table among variables in the network topology according to the network topology and the historical system operation data; and the fourth determining module is used for determining the root variable corresponding to the target variable when the running data is abnormal according to the initial condition probability table.
In a possible implementation form of the present invention, the apparatus further includes:
and the first updating module is used for updating the variable set according to the attribute characteristics of the system.
In another possible implementation form of the present invention, the fourth determining module includes:
a first determining unit, configured to determine, according to the initial condition probability table, each expected value of the target variable when each variable in the variable set is in a full load state;
and the second determining unit is used for determining a root variable corresponding to the target variable when the running data corresponding to the target variable is abnormal according to each expected value of the target variable.
In another possible implementation form of the present invention, each variable in the network topology corresponds to N states, the initial conditional probability table is composed of a conditional probability matrix among the variables, the conditional probability matrix among the variables includes N × N probability values, where N is a positive integer greater than 1;
the first determining unit is specifically configured to:
acquiring a first conditional probability matrix between a first variable in the variable set and the target variable;
and determining the expected value of the target variable when the first variable is in a full load state according to the first conditional probability matrix.
In another possible implementation form of the present invention, the apparatus further includes:
a fifth determination module to determine a pressure indicator of the system;
the second updating module is used for respectively expanding corresponding multiples of the operation data corresponding to one or more variables in the network topology according to the pressure index to update;
a sixth determining module, configured to determine, by using the updated data, an updated conditional probability table among the variables in the network topology;
and the seventh determining module is used for determining the state of the system under the pressure index according to the updated conditional probability table.
In another possible implementation form of the present invention, the apparatus further includes:
and the third updating module is used for performing optimization updating on the system according to the root variable.
The system performance analysis device of the embodiment of the invention firstly obtains the historical operation data of the system and the target variable to be analyzed; then, performing correlation analysis on historical operation data of the system, and determining a variable set of which the correlation with a target variable meets a preset condition; learning the target variable and each variable in the variable set, and determining the network topology between the target variable and each variable in the variable set; determining an initial condition probability table among variables in the network topology according to the network topology and the historical operation data of the system; and finally, according to the initial condition probability table, determining a root variable corresponding to the target variable when the running data is abnormal. Therefore, the root variable corresponding to the target variable when the running data is abnormal is comprehensively determined according to the correlation analysis and the network topology, the accuracy of the analysis result is improved, and the reliability of the system is improved by optimizing and updating the system according to the root variable.
To achieve the above object, a third embodiment of the present invention provides a server, including:
a memory, a processor and a computer program stored on the memory and executable on the processor, the program when executed by the processor implementing the system performance analysis method according to the first aspect.
To achieve the above object, a fourth aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the system performance analysis method according to the first aspect.
To achieve the above object, an embodiment of a fifth aspect of the present invention provides a computer program product, which when executed by an instruction processor in the computer program product, performs the system performance analysis method according to the first aspect.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flow diagram of a system performance analysis method of one embodiment of the invention;
FIG. 1A is a network topology diagram of a Bayesian network of one embodiment of the present invention;
FIG. 2 is a flow chart of a system performance analysis method of another embodiment of the present invention;
FIG. 3 is a schematic diagram of a system performance analysis apparatus according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a system performance analysis apparatus according to another embodiment of the present invention;
fig. 5 is a schematic structural diagram of a server according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
Specifically, the embodiments of the present invention provide a system performance analysis method for solving the problem that the performance of an application system may be reduced due to various reasons during the running process of computer application software, thereby affecting the use of a user.
The system performance analysis method provided by the embodiment of the invention comprises the steps of firstly obtaining historical system operation data and target variables to be analyzed, then carrying out correlation analysis on the obtained historical system operation data, determining a variable set having certain influence on the target variables, then learning each variable in the variable set and the target variables, determining the network topology between the target variables and each variable in the variable set, and determining an initial condition probability table among each variable in the network topology, thereby determining root source variables corresponding to the target variables when the operation data is abnormal according to the initial condition probability table. The method and the device realize comprehensive determination of the root variables influencing the system performance according to the correlation analysis and the network topology, improve the accuracy of the analysis result, and improve the reliability of the system by optimizing and updating the system according to the root variables.
The following describes a system performance analysis method, apparatus, and server according to an embodiment of the present invention with reference to the drawings.
FIG. 1 is a flow chart of a system performance analysis method according to one embodiment of the invention.
As shown in fig. 1, the system performance analysis method includes:
The execution subject of the system performance analysis method provided by the embodiment of the present invention is the system performance analysis device provided by the embodiment of the present invention, and the device may be configured in any terminal, for example, in an application server, to implement analysis on system performance.
The target variable to be analyzed may be a variable that is a main basis for measuring the system performance. Such as http response time, which measures website performance, number of clicks, which measures web server processing power, etc.
102, performing correlation analysis on the historical operating data of the system, and determining a variable set of which the correlation with the target variable meets a preset condition, wherein the variable set comprises at least one non-target variable.
It can be understood that a large number of variables may be included in the system, and not all variables may affect the target variables, and in order to determine the key variables, i.e., the root variables, which affect the target variables, correlation analysis may be performed on historical operating data of the system, so as to determine variables having certain effects on the target variables in advance, thereby reducing the analysis range and speeding up the analysis efficiency.
Specifically, historical operating data corresponding to all system observation variables may be obtained first, and then correlations between the target variable and other system observation variables are analyzed according to the historical operating data, so that a variable set, the correlations of which with the target variable are greater than a preset range, is selected from the plurality of system observation variables according to the correlations.
For example, assume that the historical operating data of the system is X ═ X1，X2，...，XNIn which XiAnd historical operating data corresponding to the ith system observation variable.
First, the system historical operating data may be normalized according to the following formula.
Then, Pearson correlation analysis was performed on the normalized data to obtain the following correlation matrix.
| cpu | http_times | session_count | thread_count | used_memory | |
| cpu | 1 | 0.10547224 | 0.088306098 | 0.110819522 | 0.026441579 |
| http_times | 0.105472 | 1 | 0.567103592 | 0.484417275 | 0.639504609 |
| session_count | 0.088306 | 0.56710359 | 1 | 0.665938271 | 0.703912576 |
| thread_count | 0.11082 | 0.48441728 | 0.665938271 | 1 | 0.759832286 |
| used_memory | 0.026442 | 0.63950461 | 0.703912576 | 0.759832286 | 1 |
The http _ times is a target variable, and the cpu, the session _ count, the thread _ count and the used _ memory are system observation variables.
By sequencing the correlations, the correlation between each system observation variable and the target variable http _ times can be determined as
Assuming that the preset variable set includes system observation variables with correlation with the target variable greater than 0.4, according to the above analysis, because used _ memory, session _ count, thread _ count and http _ time have high linear correlation, http may be caused to be correspondingly too slow, and thus, it may be determined that each variable in the variable set is used _ memory, session _ count and thread _ count.
It will be appreciated that some important variables in the system, which may be of lesser relevance to the target variable, are not included in the set of variables determined from the relevance analysis. Then, in the embodiment of the present invention, a variable may also be added to the variable set as needed, that is, after step 102, the method may further include:
and updating the variable set according to the attribute characteristics of the system.
The attribute characteristics of the system may refer to the type of the system, the environment in which the system is located, and the like.
During specific implementation, operation and maintenance personnel can determine variables to be considered according to experience and attribute characteristics of the system, so that the variables to be considered are added to the determined variable set, and the variable set is updated.
And 104, determining an initial condition probability table among variables in the network topology according to the network topology and the historical system operation data.
In specific implementation, any model capable of expressing or reasoning uncertain knowledge can be adopted to learn the target variable and each variable in the variable set so as to determine the network topology between the target variable and each variable in the variable set. For example, a bayesian network, a hidden markov model, or the like may be used.
Taking a bayesian network as an example, a process of determining a network topology between a target variable and each variable in a variable set and determining an initial condition probability table between each variable in the network topology according to the network topology and historical operating data of the system will be described in detail below.
In particular, a bayesian network is a probabilistic graphical model that graphically represents the joint probability distribution function among a set of variables. The bayesian network comprises a set of conditional probability distribution functions associated with a structural model and a threshold. The structural model is a directed acyclic graph in which nodes represent random variables and describe the states of the variables. Edges represent probabilistic dependencies between variables. Each node in the probability map has a conditional probability distribution function for that node given its parent. Thus, a Bayesian network can be represented graphically and combine conditional probability functions associated with a series of nodes into an overall joint probability distribution function.
Bayesian networks have the following advantages:
1) the Bayesian network organically combines the directed acyclic graph with the probability theory, so that the method not only has a formal probability theory basis, but also has a more intuitive knowledge representation form.
2) The bayesian network is different from the general knowledge representation method in modeling the problem domain, and thus when conditions, behaviors, or the like change, the model is not corrected.
3) The Bayesian network can graphically represent joint probabilities among random variables, and thus can process various kinds of uncertainty information.
4) The Bayesian network has no determined input or output nodes, the nodes are mutually influenced, any node observation value or any node interference can influence other nodes, and estimation and prediction can be carried out by utilizing Bayesian network reasoning.
5) The Bayesian network reasoning is based on the Bayesian probability theory, does not need any external reasoning mechanism, not only has theoretical basis, but also combines knowledge representation and knowledge reasoning to form a unified whole.
Based on the advantages of the bayesian network, in the embodiment of the present invention, the bayesian network learning can be performed on the target variable and the variables in the variable set to determine the network topology between the target variable and the variables in the variable set, so as to determine the initial condition probability table between the variables in the network topology by using the network topology and the historical operating data of the system.
Specifically, bayesian network learning can be performed from multiple angles, and in the embodiment of the present invention, a method based on evaluation and search can be adopted to perform bayesian network learning on the target variable and each variable in the variable set, so as to determine the network topology between the target variable and each variable in the variable set.
For example, suppose Z is a set of n discrete variables, xiIs r isiThe possible values are:
d is a data set of m samples. B issIs a bayesian network that contains variables in Z. B issEach variable in the tree has a parent node set pii. By wijRepresents piiThe jth different state of (a). Suppose q is providediOne such different state. Definition of NijkIs the variable x in DiHas a value of vikAnd piiAt wijNumber of samples of the state.
The maximum likelihood of the data is represented by a bayesian dirichlet scoring function. The scoring function is as follows:
wherein, P (B)S) Is the prior probability of the network structure.
Specifically, to obtain the network structure with the maximum a posteriori probability, it can be assumed that Q is the set of all bayesian network structures formed by the variables in Z. The scale of Q increases rapidly with the scale of Z. Consider the following
And the | Y | is small and,
if Y is very significant, P (B)S| D) will be very close to the optimum value and the calculation will be very efficient.
The network topology shown in fig. 1A can be obtained by performing bayesian network learning on each variable session _ count, used _ memory, thread _ count, cpu and target variable http _ times in the variable set. And further, according to the network topology and the historical operation data of the system, an initial condition probability table among all variables can be determined.
And 105, determining a root variable corresponding to the target variable when the running data is abnormal according to the initial condition probability table.
Specifically, step 105 may include:
and 105a, respectively determining each expected value of the target variable when each variable in the variable set is in a full load state according to the initial condition probability table.
It can be understood that, when learning is performed by using the target variable and each variable in the variable set, the target variable and each variable in the variable set may be subjected to equal frequency discretization into N values, which represent N system states, so that each variable in the determined network topology corresponds to N states, respectively, and the initial conditional probability table is composed of a conditional probability matrix among the variables, where the conditional probability matrix among the variables includes N × N probability values, where N is a positive integer greater than 1.
Correspondingly, step 105a may specifically include:
acquiring a first conditional probability matrix between a first variable in the variable set and the target variable;
and determining the expected value of the target variable when the first variable is in a full load state according to the first conditional probability matrix.
For example, assuming that the target variable http _ times and the variables in the variable set, such as session _ count, used _ memory, thread _ count, and cpu, are frequency-discretized into three values of 0, 1, and 2, that is, each variable in the network topology corresponds to 3 states, respectively, a conditional probability matrix among the above 5 variables is formed in the initial conditional probability table, and the conditional probability matrix among the variables includes 9 probability values, respectively.
Assuming that the first variable is session _ count, a first conditional probability matrix C between the first variable and the target variable http _ times can be expressed as:
wherein, C11C12C13The conditional probabilities of the http _ times in the low state when the first variable session _ count is in the low, medium and high 3 states respectively are indicated; c21C22C23The conditional probabilities of the http _ times in the middle state when the first variable session _ count is in the low state, the middle state and the high state respectively are referred to respectively; c31C32C33The conditional probabilities of the http _ times in the high state when the first variable session _ count is in the low, middle and high 3 states respectively are referred to.
According to a first conditional probability matrix C, i.e. by 2C13+2*c23+2*c33And, the expected value of the target variable when the first variable session _ count is in the full load state (high state) is determined.
And the expected values of the target variables when the variables in the variable set are respectively in the full load state can be determined through the conditional probability matrix between the variables in the variable set and the target variables.
And 105b, determining a root variable corresponding to the target variable when the running data corresponding to the target variable is abnormal according to each expected value of the target variable.
Specifically, after each expected value of the target variable is determined, the variable with the largest expected value in the variable set may be determined as the root variable when the operation data corresponding to the target variable is abnormal.
For example, assuming that when the variables session _ count, used _ memory, thread _ count, and cpu in the variable set are in a full load state, respectively, the expected value of the target variable http _ times, that is, the elasticity risk of each variable is as follows, the root variable when the running data corresponding to the target variable is abnormal may be determined to be session _ count.
Specifically, after the root variable corresponding to the target variable when the operation data is abnormal is determined, the system can be maintained and optimized according to the root variable, so as to improve the performance of the system. That is, after step 105, the method may further include:
and performing optimization updating on the system according to the root variable.
The system performance analysis method of the embodiment of the invention comprises the steps of firstly, obtaining historical operation data of a system and target variables to be analyzed; then, performing correlation analysis on historical operation data of the system, and determining a variable set of which the correlation with a target variable meets a preset condition; learning the target variable and each variable in the variable set, and determining the network topology between the target variable and each variable in the variable set; determining an initial condition probability table among variables in the network topology according to the network topology and the historical operation data of the system; and finally, according to the initial condition probability table, determining a root variable corresponding to the target variable when the running data is abnormal. Therefore, the root variable corresponding to the target variable when the running data is abnormal is comprehensively determined according to the correlation analysis and the network topology, the accuracy of the analysis result is improved, and the reliability of the system is improved by optimizing and updating the system according to the root variable.
Through the analysis, the root variable when the running data corresponding to the target variable is abnormal can be comprehensively determined according to the correlation analysis of the historical running data of the system, the learning of the target variable and the variables in the variable set, the determined network topology, and the optimization and the updating of the system. In one possible implementation form, the system may also be subjected to a stress test according to a network topology, which is described in detail below with reference to fig. 2.
Fig. 2 is a flow chart of a system performance analysis method according to another embodiment of the present invention.
As shown in fig. 2, the method includes:
And 204, determining an initial condition probability table among variables in the network topology according to the network topology and the historical system operation data.
The implementation principle and process of the steps 201 to 204 may refer to the detailed description of the steps 101 to 104, which is not described herein again.
The pressure index refers to a pressure level for performing a pressure test on the system, and specifically, may be a user amount, a transaction amount, and the like for concurrently accessing the system.
And step 206, according to the pressure indexes, expanding corresponding multiples of the operation data corresponding to one or more variables in the network topology respectively for updating.
And step 208, determining the state of the system under the pressure index according to the updated conditional probability table.
It will be appreciated that the system may be stress tested in order to determine the bottleneck or maximum usage limit of the system. When a pressure test is usually carried out, a certain number of users can be simulated to access the system simultaneously, and the response time of the system is recorded and analyzed; or simulating that only one user accesses the system and uses the same operation, testing different data volumes, recording different data volumes and corresponding resource occupancy rates, and the like, to determine the performance of the system under extreme conditions. However, when the pressure test is performed in this way, the actual system may be directly subjected to the pressure test, which may cause some damage to the system and affect the performance of the system.
In the embodiment of the invention, the states of the system under different pressure indexes can be determined according to the determined target variable and the network topology among the variables in the variable set, so that the system is subjected to pressure test by utilizing theoretical analysis, the damage to the system caused by the test process is avoided, and the development time and labor cost are saved.
Specifically, the operation data corresponding to each variable in the network topology may be respectively expanded by corresponding multiples to be updated according to the pressure index, so that the updated conditional probability table in the network topology is determined by using the updated data.
For example, assume that the conditional probability table C of the variables session _ count and used _ memory under the current pressure is:
assuming that the determined pressure index of the system is 1.5 times of the current pressure, the operation data corresponding to each variable in the network topology may be expanded by 1.5 times according to the pressure index, so as to obtain updated data, and the condition probability table C is updated according to the following formula.
Wherein S isiIndicating the probability value of the variable used _ memory in the i state after the running data corresponding to the variable session _ count is enlarged by 1.5 times; cijWhen the variable session _ count is in the j state, the probability of the used _ memory in the i state is indicated; c'ijThe probability of used _ memory in the i state when the updated variable session _ count is in the j state.
Specifically, after the updated conditional probability table among the variables in the network topology is determined, the state of the system under the pressure index can be determined according to the updated conditional probability table.
For example, assume that the state of a variable in the system at the current pressure is:
j＝0 1 2
p＝0.332963436 0.319733649 0.347302943
assuming that the pressure index is 1.5 times of the current pressure, respectively expanding the operation data corresponding to each variable in the network topology by 1.5 times according to the pressure index for updating, and determining an updated conditional probability table among the variables in the network topology by using the updated data, so as to obtain the state of the variable under the pressure index:
j＝0 1 2
p′＝0.108047746 0.433293431 0.458658844
the method for analyzing the system performance of the embodiment of the invention comprises the steps of determining the pressure index of the system after determining the initial condition probability table among the variables in the network topology, expanding the corresponding times of the operation data corresponding to one or more variables in the network topology according to the pressure index for updating, and determining the updated condition probability table among the variables in the network topology by using the updated data, thereby determining the state of the system under the pressure index according to the updated condition probability table. Therefore, the system is subjected to pressure test according to the correlation analysis and the network topology, the damage to the system caused by the test process is avoided, the development time and the labor cost are saved, and the reliability of the system is improved.
Fig. 3 is a schematic structural diagram of a system performance analysis apparatus according to an embodiment of the present invention.
As shown in fig. 3, the system performance analysis apparatus includes:
the acquisition module 31 is used for acquiring historical operating data of the system and target variables to be analyzed;
a first determining module 32, configured to perform correlation analysis on the historical operating data of the system, and determine a variable set in which correlation with the target variable satisfies a preset condition, where the variable set includes at least one non-target variable;
a second determining module 33, configured to learn the target variable and each variable in the variable set, and determine a network topology between the target variable and each variable in the variable set;
a third determining module 34, configured to determine an initial condition probability table among variables in the network topology according to the network topology and the historical system operation data;
a fourth determining module 35, configured to determine, according to the initial condition probability table, a root variable when the running data corresponding to the target variable is abnormal.
Specifically, the system performance analysis apparatus provided in this embodiment may be configured in any terminal, and is configured to execute the system performance analysis method shown in the foregoing embodiment, so as to implement analysis on system performance.
In a possible implementation form of the embodiment of the present application, the fourth determining module 35 includes:
a first determining unit, configured to determine, according to the initial condition probability table, each expected value of the target variable when each variable in the variable set is in a full load state;
and the second determining unit is used for determining a root variable corresponding to the target variable when the running data corresponding to the target variable is abnormal according to each expected value of the target variable.
In a possible implementation form of the embodiment of the present application, each variable in the network topology corresponds to N states, the initial conditional probability table is composed of a conditional probability matrix among the variables, the conditional probability matrix among the variables includes N × N probability values, where N is a positive integer greater than 1;
the first determining unit is specifically configured to:
acquiring a first conditional probability matrix between a first variable in the variable set and the target variable;
and determining the expected value of the target variable when the first variable is in a full load state according to the first conditional probability matrix.
It should be noted that the foregoing explanation of the embodiment of the system performance analysis method is also applicable to the system performance analysis apparatus of this embodiment, and is not repeated here.
The system performance analysis device of the embodiment of the invention firstly obtains the historical operation data of the system and the target variable to be analyzed; then, performing correlation analysis on historical operation data of the system, and determining a variable set of which the correlation with a target variable meets a preset condition; learning the target variable and each variable in the variable set, and determining the network topology between the target variable and each variable in the variable set; determining an initial condition probability table among variables in the network topology according to the network topology and the historical operation data of the system; and finally, according to the initial condition probability table, determining a root variable corresponding to the target variable when the running data is abnormal. Therefore, the root variable corresponding to the target variable when the running data is abnormal is comprehensively determined according to the correlation analysis and the network topology, the accuracy of the analysis result is improved, and the reliability of the system is improved by optimizing and updating the system according to the root variable.
Fig. 4 is a schematic structural diagram of a system performance analysis apparatus according to another embodiment of the present invention.
As shown in fig. 4, the system performance analysis apparatus may further include, based on that shown in fig. 3:
a first updating module 41, configured to update the variable set according to the attribute characteristics of the system.
A fifth determination module 42 for determining a pressure indicator of the system;
a second updating module 43, configured to expand the operating data corresponding to one or more variables in the network topology by corresponding multiples respectively according to the pressure indicator to perform updating;
a sixth determining module 44, configured to determine, by using the updated data, an updated conditional probability table among the variables in the network topology;
a seventh determining module 45, configured to determine, according to the updated conditional probability table, a state of the system under the pressure indicator.
And a third updating module 46, configured to perform optimization updating on the system according to the root variable.
It should be noted that the foregoing explanation of the embodiment of the system performance analysis method is also applicable to the system performance analysis apparatus of this embodiment, and is not repeated here.
The system performance analysis device of the embodiment of the invention firstly obtains the historical operation data of the system and the target variable to be analyzed; then, performing correlation analysis on historical operation data of the system, and determining a variable set of which the correlation with a target variable meets a preset condition; learning the target variable and each variable in the variable set, and determining the network topology between the target variable and each variable in the variable set; determining an initial condition probability table among variables in the network topology according to the network topology and the historical operation data of the system; and finally, according to the initial condition probability table, determining a root variable corresponding to the target variable when the running data is abnormal. Therefore, the root variable corresponding to the target variable when the running data is abnormal is comprehensively determined according to the correlation analysis and the network topology, the accuracy of the analysis result is improved, and the reliability of the system is improved by optimizing and updating the system according to the root variable.
To achieve the above object, a third aspect of the present invention provides a server.
Fig. 5 is a schematic structural diagram of a server according to an embodiment of the present invention.
As shown in fig. 5, the server includes: a memory 51, a processor 52 and a computer program stored on the memory 51 and executable on the processor 52.
The processor 52 implements the application state monitoring method provided in the above-described embodiment when executing the program.
Further, the server further comprises:
a communication interface 53 for communication between the memory 51 and the processor 52.
A memory 51 for storing a computer program operable on the processor 52.
The memory 51 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The processor 52 is configured to implement the browser page testing method according to the foregoing embodiment when executing the program.
If the memory 51, the processor 52 and the communication interface 53 are implemented independently, the communication interface 53, the memory 51 and the processor 52 may be connected to each other through a bus and perform communication with each other. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (enhanced Industry Standard Architecture) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 5, but this does not mean only one bus or one type of bus.
Alternatively, in practical implementation, if the memory 51, the processor 52 and the communication interface 53 are integrated on one chip, the memory 51, the processor 52 and the communication interface 53 may complete communication with each other through an internal interface.
To achieve the above object, a fourth aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the system performance analysis method as in the previous embodiments.
To achieve the above object, an embodiment of a fifth aspect of the present invention provides a computer program product, which when executed by an instruction processor in the computer program product, performs a system performance analysis method as in the foregoing embodiments.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (8)
1. A method for analyzing system performance, comprising:
acquiring historical operating data of a system and a target variable to be analyzed;
performing correlation analysis on the historical operating data of the system, and determining a variable set of which the correlation with the target variable meets a preset condition, wherein the variable set comprises at least one non-target variable;
learning the target variable and each variable in the variable set, and determining a network topology between the target variable and each variable in the variable set;
determining an initial condition probability table among variables in the network topology according to the network topology and the historical system operation data;
and determining a root variable corresponding to a target variable when the running data is abnormal according to the initial condition probability table, wherein each expected value of the target variable when each variable in the variable set is in a full load state is determined according to the initial condition probability table, and the root variable corresponding to the target variable when the running data is abnormal is determined according to each expected value of the target variable.
2. The method of claim 1, wherein after determining the set of variables whose correlation with the target variable satisfies a preset condition, further comprising:
and updating the variable set according to the attribute characteristics of the system.
3. The method of claim 1, wherein each variable in the network topology corresponds to N states, and the initial conditional probability table is composed of a conditional probability matrix between each variable, where the conditional probability matrix between each variable includes N × N probability values, where N is a positive integer greater than 1;
the determining, when the variables in the variable set are in a full load state, each expected value of the target variable includes:
acquiring a first conditional probability matrix between a first variable in the variable set and the target variable;
and determining the expected value of the target variable when the first variable is in a full load state according to the first conditional probability matrix.
4. A method according to any of claims 1-3, wherein after determining the table of initial conditional probabilities between variables in the network topology, further comprising:
determining a pressure indicator of the system;
according to the pressure indexes, respectively expanding corresponding multiples of operation data corresponding to one or more variables in the network topology for updating;
determining an updated conditional probability table among variables in the network topology by using the updated data;
and determining the state of the system under the pressure index according to the updated conditional probability table.
5. The method according to any one of claims 1 to 3, wherein after determining the root variable when the running data corresponding to the target variable is abnormal, the method further comprises: and performing optimization updating on the system according to the root variable.
6. A system performance analysis apparatus, comprising:
the acquisition module is used for acquiring historical operating data of the system and target variables to be analyzed;
the first determination module is used for performing correlation analysis on the historical operating data of the system and determining a variable set of which the correlation with the target variable meets a preset condition, wherein the variable set comprises at least one non-target variable;
the second determining module is used for learning the target variable and each variable in the variable set and determining the network topology between the target variable and each variable in the variable set;
a third determining module, configured to determine an initial condition probability table among variables in the network topology according to the network topology and the historical system operation data;
a fourth determining module, configured to determine, according to the initial condition probability table, a root variable corresponding to the target variable when the operating data is abnormal;
the fourth determining module includes:
a first determining unit, configured to determine, according to the initial condition probability table, each expected value of the target variable when each variable in the variable set is in a full load state;
and the second determining unit is used for determining a root variable corresponding to the target variable when the running data corresponding to the target variable is abnormal according to each expected value of the target variable.
7. A server, comprising:
memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor implements the system performance analysis method according to any of claims 1-5 when executing the program.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the system performance analysis method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710541436.5A CN107463486B (en) | 2017-07-05 | 2017-07-05 | System performance analysis method and device and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710541436.5A CN107463486B (en) | 2017-07-05 | 2017-07-05 | System performance analysis method and device and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107463486A CN107463486A (en) | 2017-12-12 |
| CN107463486B true CN107463486B (en) | 2020-09-18 |
Family
ID=60544329
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710541436.5A Active CN107463486B (en) | 2017-07-05 | 2017-07-05 | System performance analysis method and device and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107463486B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109996269A (en) * | 2017-12-30 | 2019-07-09 | 中国移动通信集团湖北有限公司 | A kind of cordless communication network abnormal cause determines method, apparatus, equipment and medium |
| CN111327443B (en) * | 2018-12-17 | 2022-11-22 | 中国移动通信集团北京有限公司 | Fault root index determination method and device |
| CN112069052B (en) * | 2019-06-11 | 2024-04-12 | 腾讯科技（深圳）有限公司 | Abnormal object detection method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101814114A (en) * | 2010-01-07 | 2010-08-25 | 北京航空航天大学 | Method for performing fault diagnosis by using model conversion |
| CN105530122A (en) * | 2015-12-03 | 2016-04-27 | 国网江西省电力公司信息通信分公司 | Network failure diagnosis method based on selective hidden Naive Bayesian classifier |
| CN106124175A (en) * | 2016-06-14 | 2016-11-16 | 电子科技大学 | A kind of compressor valve method for diagnosing faults based on Bayesian network |
| CN106326585A (en) * | 2016-08-29 | 2017-01-11 | 东软集团股份有限公司 | Prediction analysis method based on bayesian network reasoning and device thereof |
| CN106570513A (en) * | 2015-10-13 | 2017-04-19 | 华为技术有限公司 | Fault diagnosis method and apparatus for big data network system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8489530B2 (en) * | 2007-12-03 | 2013-07-16 | Infosys Limited | System and method for root cause analysis of the failure of a manufactured product |
-
2017
- 2017-07-05 CN CN201710541436.5A patent/CN107463486B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101814114A (en) * | 2010-01-07 | 2010-08-25 | 北京航空航天大学 | Method for performing fault diagnosis by using model conversion |
| CN106570513A (en) * | 2015-10-13 | 2017-04-19 | 华为技术有限公司 | Fault diagnosis method and apparatus for big data network system |
| CN105530122A (en) * | 2015-12-03 | 2016-04-27 | 国网江西省电力公司信息通信分公司 | Network failure diagnosis method based on selective hidden Naive Bayesian classifier |
| CN106124175A (en) * | 2016-06-14 | 2016-11-16 | 电子科技大学 | A kind of compressor valve method for diagnosing faults based on Bayesian network |
| CN106326585A (en) * | 2016-08-29 | 2017-01-11 | 东软集团股份有限公司 | Prediction analysis method based on bayesian network reasoning and device thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107463486A (en) | 2017-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10552119B2 (en) | Dynamic management of numerical representation in a distributed matrix processor architecture | |
| CN108683530B (en) | Data analysis method and device for multi-dimensional data and storage medium | |
| JP6822509B2 (en) | Data processing methods and electronic devices | |
| CN110554958B (en) | Graph database testing method, system, device and storage medium | |
| CN107463486B (en) | System performance analysis method and device and server | |
| CN111122199A (en) | Boiler fault diagnosis method and device | |
| JP2017021772A (en) | Copula-theory based feature selection | |
| Chen et al. | Optimal variability sensitive condition-based maintenance with a Cox PH model | |
| US20140142903A1 (en) | Table model circuit simulation acceleration using model caching | |
| Balamurali et al. | Optimal design of repetitive group sampling plans for Weibull and gamma distributions with applications and comparison to the Birnbaum–Saunders distribution | |
| CN116089870A (en) | Industrial equipment fault prediction method and device based on meta-learning under small sample condition | |
| US9454457B1 (en) | Software test apparatus, software test method and computer readable medium thereof | |
| CN105447032A (en) | Method and system for processing message and subscription information | |
| CN112801315A (en) | State diagnosis method and device for power secondary equipment and terminal | |
| Chen et al. | Inference for a mean-reverting stochastic process with multiple change points | |
| Breitenbach et al. | A principal feature analysis | |
| JP2020181578A (en) | Method, device, and medium for data processing | |
| JP2014228975A (en) | Retrieval device, retrieval method and retrieval program | |
| Dannemann | Semiparametric hidden Markov models | |
| Poulos et al. | Exemplar-based failure triage for regression design debugging | |
| CN113357138B (en) | Method and device for predicting remaining service life of hydraulic pump and terminal equipment | |
| CN113887101A (en) | Visualization method and device of network model, electronic equipment and storage medium | |
| CN114021619A (en) | Classification method and device based on graph structure evolution | |
| CN113449382B (en) | Simulation method and system for whole vehicle emission result | |
| CN110796234B (en) | Method and device for predicting computer state |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |