CN106951359A - A kind of system health degree determination method and device - Google Patents
A kind of system health degree determination method and device Download PDFInfo
- Publication number
- CN106951359A CN106951359A CN201710113911.9A CN201710113911A CN106951359A CN 106951359 A CN106951359 A CN 106951359A CN 201710113911 A CN201710113911 A CN 201710113911A CN 106951359 A CN106951359 A CN 106951359A
- Authority
- CN
- China
- Prior art keywords
- data
- health degree
- system health
- model
- degrees
- 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.)
- Pending
Links
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
- G06F11/3447—Performance evaluation by modeling
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a kind of system health degree determination method and device, the system health degree determination method includes:Set up system health degree model;According to the system health degree model, the system health degree in the first setting time interval is detected, the healthy degrees of data of the first system is obtained;According to the healthy degrees of data of the first system, the system health model is detected and adjusted, the degree of fitting of the system health degree model is reached predetermined value;According to the system health degree model after adjustment, the system health degree in the second setting time interval, generation detection and analysis report are detected;To user's displaying detection and analysis report.The present invention is by setting up system health degree model and by presetting method, the health degree of operational system can be tested and analyzed.
Description
Technical field
The present invention relates to computer application field, more particularly to a kind of system health degree determination method and device.
Background technology
National comprehensive securities broker company possesses substantial amounts of operation system and client, except the concentration related to core transaction
Outside the core transaction systems such as transaction system, fund-raising gap system, online transaction system, there are other peripheral system (data volumes
50 or so), these systems are disposed on a different server again, and with continuous hot, the stock of recent stock market's exchange quotation
The turnover and conclusion of the business stroke count in city are also continuous foundation new peak, the super trillion yuan of stock markets of Shanghai exchange hand of such as proximal segment time.And mesh
The O&M of preceding transaction system relies primarily on manual type progress, can not both meet the O&M requirement of the system of such vast number,
In the case of exchange hand is ever-increasing, running status of the system in certain time period can not also be detected, thus not
The health status of system is understood beneficial to user.
A series of above-mentioned reasons, result in maintenance work needs an intelligent O&M accessory system, and the system should
Can monitoring system certain time interval in running status, detecting system certain time interval in health degree, and by
This provides related advisory.
The content of the invention
For defect of the prior art, the present invention provides a kind of system health degree determination method and device, can
The health degree of operational system is tested and analyzed.
In a first aspect, the invention provides a kind of system health degree determination method, the system health degree detection point
Analysis method includes:
Set up system health degree model;
According to the system health degree model, the system health degree in the first setting time interval is detected, obtaining first is
The healthy degrees of data of system;
According to the healthy degrees of data of the first system, the system health model is detected and adjusted, makes the system
The degree of fitting of system health degree model reaches predetermined value;
According to the system health degree model after adjustment, the system health degree in the second setting time interval is detected, it is raw
Into detection and analysis report;
To user's displaying detection and analysis report.
Further, it is described to set up system health degree model, specifically include:
Collect first sample data, the first sample data include basic monitoring data and with the basic monitoring data
The healthy degrees of data of corresponding second system, the healthy degrees of data of the second system is the data measured in advance;It is default using first
Method learns the relation of the basic monitoring data and the healthy degrees of data of the second system;According to the basic monitoring data and
The relation of the healthy degrees of data of the second system sets up the system health degree model.
Further, it is described according to the system health degree model, detect the system health in the first setting time interval
Degree, obtains the healthy degrees of data of the first system, specifically includes:
Obtain multiple ongoing basis monitoring datas in the first setting time interval;
According to multiple ongoing basis monitoring in the system health degree model and first setting time interval
Data, the multiple system real time health degrees of data obtained in the first setting time interval are calculated using the second presetting method;
The system real time health number of degrees described in first in setting time interval are obtained using second presetting method calculating
According to variation tendency;
According to the variation tendency of the system real time health degrees of data in first setting time interval, described first is judged
System health degree in setting time interval, obtains the healthy degrees of data of the first system.
Further, in the system health degree model according to after adjustment, the second setting time of detection interval
System health degree, generation detection and analysis report, is specifically included:
Obtain the ongoing basis monitoring data in the second setting time interval;
It is default using described second according to the system health degree model and the ongoing basis monitoring data after adjustment
Method calculates the multiple system real time health degrees of data obtained in the second setting time interval;
The system real time health number of degrees obtained in the second setting time interval are calculated using second presetting method
According to variation tendency;
According to the variation tendency of the system real time health degrees of data in second setting time interval, described second is judged
System health degree in setting time interval, and generate detection and analysis report.
Further, the system includes core transaction system and peripheral system;The system health degree detection and analysis side
Method also includes:Set up core transaction system health degree model and basic monitoring data model;
The multiple ongoing basis monitoring datas obtained in the first setting time interval, or acquisition described second are set
The multiple ongoing basis monitoring datas fixed time in interval, are specifically included:
According to the core transaction system health degree model, when obtaining first setting using the calculating of the second presetting method
Between it is interval in or second setting time interval in multiple first core transaction system health degrees of data;
According to the multiple first core transaction system health degrees of data and the basic monitoring data model, using described
Second presetting method, which is calculated, to be obtained multiple ongoing basis monitoring datas in first setting time interval or described second sets
The multiple ongoing basis monitoring datas fixed time in interval.
Further, it is described to set up core transaction system health degree model, specifically include:
The second sample data is collected, second sample data includes basic monitoring data, application process data, daily record number
According to the second core transaction system health degrees of data;Wherein, the basic monitoring data, the application process data and the day
Will data are independent variable, and the second core transaction system health degrees of data is dependent variable;The second core transaction system is good for
Health degrees of data is the data measured in advance;Learn the pass between the independent variable and the dependent variable using the first presetting method
System;The core transaction system health degree model is set up according to the relation of the independent variable and the dependent variable.
Further, it is described to set up basic monitoring data model, specifically include:
Collect the 3rd sample data, the 3rd sample data include the second core transaction system health degrees of data and
The basic monitoring data;Learn the second core transaction system health degrees of data and the basis using the first presetting method
The relation of monitoring data;Set up according to the relation of the second core transaction system health degrees of data and the basic monitoring data
The basic monitoring data model.
Further, it is described to set up before system health degree model, also include:
The healthy degrees of data of legacy system is collected, according to the healthy degrees of data of the legacy system, analyzing influence system health degree
Influence factor and the influence factor weight;
According to the influence factor and the weight of the influence factor, it is determined that needing the class of system health degree model set up
Type;Wherein, the type of the system health degree model includes:Multivariate regression models, Logic Regression Models, neural network model.
Further, first presetting method is machine learning, the algorithm of the machine learning and the system health
Spend the type correspondence of model.
Second aspect, device, the system health degree detection are tested and analyzed present invention also offers a kind of system health degree
Analytical equipment includes:Model building module, data detection module, data analysis module, data display module;
The model building module, for setting up system health degree model;The data detection module, for according to described
System health degree in system health degree model, the first setting time of detection interval, obtains the healthy degrees of data of the first system;It is described
Data analysis module, for according to the healthy degrees of data of the first system, the system health model being detected and being adjusted,
The degree of fitting of the system health degree model is reached predetermined value, be also used for according to the system health degree model after adjustment,
Detect the system health degree in the second setting time interval, generation detection and analysis report;The data display module, for
The family displaying detection and analysis report.
As shown from the above technical solution, the present invention provides a kind of system health degree determination method and device, by building
Erection system health degree model and by presetting method, can monitor the running status of operational system, to system health degree model
Detection and adjustment, can be such that system health degree model more improves and accurate;The health degree of system is tested and analyzed and given birth to
Shown into detection and analysis report to user, allow users to be visually known the health status of system.
Brief description of the drawings
Fig. 1 shows the schematic flow sheet for the system health degree determination method that the present invention is provided.
Fig. 2 shows that the system health degree that the present invention is provided tests and analyzes the structural representation of device.
Embodiment
The embodiment of technical solution of the present invention is described in detail below in conjunction with accompanying drawing.Following examples are only used for
Clearly illustrate technical scheme, therefore be intended only as example, and the protection of the present invention can not be limited with this
Scope.
Embodiment one
Fig. 1 shows the schematic flow sheet for the system health degree determination method that the present invention is provided.As shown in figure 1, this
Invention provides a kind of system health degree determination method, including:
Step S1, sets up system health degree model;
Step S2, according to the system health degree model, tests and analyzes the system health degree in setting time interval, obtains
The first system health degrees of data;
Step S3, according to the healthy degrees of data of the first system, is detected and is adjusted to the system health model, made
The degree of fitting of the system health degree model is set to reach predetermined value;
Step S4, according to the system health degree model after adjustment, is tested and analyzed to the health degree of the system,
Generation detection and analysis report;
Step S5, is reported to user's displaying detection and analysis.
System described in the present embodiment includes core transaction system and peripheral system, wherein, core transaction system includes collection
Middle transaction system, fund-raising gap transaction system, online transaction system etc., system health degree can from the hardware utilization rate of machine and
Embodied in terms of the fault rate two of system, the hardware utilization rate of machine can be embodied from core transaction system, and core
The health degree of transaction system can be drawn from the O&M achievement data of each service link.The O&M achievement data includes
Basic monitoring data, application process data and the daily record data of system, wherein, the basic monitoring data is used including IT resources
Rate, is specifically included:CPU usage, memory usage, network traffics and disk utilization rate etc.;The application process data include
Port data, process status and application internal indicator etc.;The daily record data includes operating system daily record, using daily record and business
Daily record etc., wherein including other more specifically data again;The O&M achievement data is that advance monitoring and storage are good.It is existing
Operational system only save a small number of data, most data such as daily record data, history monitoring data, are not all protected
Stay, and system health degree determination method provided in an embodiment of the present invention before model is set up to the every operation/maintenance data of system
The monitoring of twenty four hours, and centralized collection and storage are carried out, so as to the health degree model for setting up correlation.
The technical scheme of the present embodiment is:
The healthy degrees of data of legacy system is collected, according to the healthy degrees of data of legacy system, the shadow of analyzing influence system health degree
The weight of the factor of sound and the influence factor, according to influence factor and its weight, it is determined that needing the system health degree model set up
Type, the type includes Logic Regression Models, multivariate regression models, neural network model, decision-tree model etc..Collect and
The O&M achievement data of history is stored, and according to the type determined, sets up multiple models for detecting system health degree, is wrapped
Include core transaction system health degree model, basic monitoring data model and system health degree model.Wherein, legacy system health degree
The O&M achievement data of data and history is the data measured beforehand through manual type, wherein, the core transaction system
Health degree model is used for the health degree for detecting the core transaction system, and acquisition can reflect that the core transaction system is good in real time
Kang Du the first core transaction system health degrees of data;The basic monitoring data model is used for according to first core transaction
System health degrees of data, obtains ongoing basis monitoring data corresponding with the first core transaction system health degrees of data;Institute
Stating system health degree model is used for according to the ongoing basis monitoring data, detects the system health degree in setting time interval.
Setting time interval can according to actual needs be set by user or operation maintenance personnel.
Wherein, set up system health degree model and specifically include procedure below:Collect first sample data, the first sample
Data include basic monitoring data and the healthy degrees of data of second system corresponding with the basic monitoring data, the second system
Healthy degrees of data is a part for legacy system health data, is the data for measuring and collecting in advance;Utilize the first presetting method
Learn the relation of the basic monitoring data and the healthy degrees of data of the second system;According to the basic monitoring data and described
The relation of second system health degrees of data sets up the system health degree model.Wherein, the quantity of the first sample data is
It is multiple, by the study to multiple data, system health degree model can be set up, data bulk is more, and results of learning are also better.
Preferably, first presetting method is the method for machine learning, other methods is can also be, in machine learning
It is preferably supervised study in a variety of learning skills, supervised study is a kind of skill in machine learning, can be provided by training
A pattern is acquired or set up in material, and pattern speculates new example according to this;Supervised study is provided to mistake in learning process
Indicate, machine is reduced error;The algorithm of machine learning has a variety of, including multiple regression analysis, logistic regression analysis, nerve net
Network analysis etc., it is different according to the type of identified system health degree model, different algorithms are accordingly selected, if for example,
The type of system health degree model is multivariate regression models, then multiple regression analysis is selected, if the class of system health degree model
Type is Logic Regression Models, then selects logistic regression analysis, if the type of system health degree model is neural network model,
Select analysis of neural network.
During system health degree model is set up, by first sample data-pushing to machine, it is allowed to learn each seed ginseng
Relation between number.For example, collecting first sample data, it includes following two groups of data:
IT resource CPU usages:50%, IT resource memory usage:60%, IT resource disk I/O:70%, IT resource network
Network flow:100M/s, system health degree was 51% at that time;
IT resource CPU usages:60%, IT resource memory usage:40%, IT resource disk I/O:70%, IT resource network
Network flow:10M/s, system health degree was 41% at that time;
By the study of the first sample data to this large amount of similar two groups of data, the recurrence of system health degree can be set up
Model.
Set up core transaction system health degree model and specifically include procedure below:Collect the second sample data, described second
Sample data includes the O&M achievement data and the second core transaction system health degrees of data;Learnt using the first presetting method
The relation of the O&M achievement data and the second core transaction system health degrees of data;According to the O&M achievement data and
The relation of the second core transaction system health degrees of data sets up the core transaction system health degree model.Wherein described fortune
Dimension indicator data are independent variable, and the second core transaction system health degrees of data is dependent variable;The second core transaction system
System data are the data measured in advance.Wherein, the second core transaction system health degrees of data includes and centralized transaction/trading system
The related data of health degree, fund-raising gap system health degree or online transaction system health degree.
During core transaction system health degree model is set up, the second sample data is pushed to machine, allows it to learn
Practise the relation between various parameters.For example, collecting the second sample data, it includes following two groups of data:
Centralized transaction/trading system exchange hour:11:08:, entrust stroke count at 07 point:1500 per second, inquire about Stock-operation:It is per second
10000, logon operation:100 per second, centralized transaction/trading system health degree at that time:50%;
Centralized transaction/trading system exchange hour:11:10:, entrust stroke count at 10 points:1200 per second, inquire about Stock-operation:It is per second
8000, logon operation:50 per second, centralized transaction/trading system health degree at that time:70%;
By the study to a large amount of the second similar sample datas, centralized transaction/trading system health degree and committee per second can be set up
Support stroke count, inquiry number per second, it is per second log in number or regression model, it is similar, can also set up through the above way concentration hand over
Easy system health degree, fund-raising gap system health degree or online transaction system health degree and the regression model of other parameters.
Set up basic monitoring data model and specifically include procedure below:Collect the 3rd sample data, the 3rd sample number
According to including the second core transaction system health degrees of data and the basic monitoring data;Learn institute using the first presetting method
State the relation of the second core transaction system health degrees of data and the basic monitoring data;According to the second core transaction system
The relation of healthy degrees of data and the basic monitoring data sets up the basic monitoring data model.
During basic monitoring data model is set up, the 3rd sample data is pushed to machine, makes its study various
Relation between parameter.For example, collecting the 3rd sample data, it includes following two groups of data:
Centralized transaction/trading system health index:50%, fund-raising gap system health index:60%, online transaction system health
Index:70%, IT resources CPU usage was 51% at that time;
Centralized transaction/trading system health index:30%, fund-raising gap system health index:10%, online transaction system health
Index:20%, IT resources memory usage was 20% at that time;
By the study to a large amount of the 3rd similar sample datas, IT resources CPU usage, internal memory can be set up respectively to be made
With the regression model of rate, network traffics, disk utilization rate etc., basic monitoring data model is belonged to.
By setting up above-mentioned model, operational system is set to detect its health degree, and exist due to setting up model
Before detection process, above three model is directly utilized in detection process so that system can more be rapidly performed by detection, saved
Go in detection process to set up the process of model in real time, also eliminated amount of calculation numerous and diverse in detection process.
Alternatively, system health degree model, core transaction system health degree model and basic monitoring data model are being set up
When, also including carrying out quality testing to first sample data, the second sample data and the 3rd sample data, to ensure the quality of data
Reliability, reduce and set up the error that exists during model.
Set up after model, the system health degree in the first setting time interval is detected using these models, first
Multiple ongoing basis monitoring datas in the first setting time interval are obtained, are specifically included:It is strong according to the core transaction system
Kang Du models, the multiple first core transaction systems obtained in the first setting time interval are calculated using the second presetting method
Healthy degrees of data;According to the multiple first core transaction system health degrees of data and the basis monitoring regression model, utilize
Second presetting method calculates the multiple ongoing basis monitoring datas obtained in the first setting time interval.Next basis
Multiple ongoing basis monitoring datas and the system health degree model in the first setting time interval, detection first are set
System health degree in time interval, obtains the healthy degrees of data of the first system, specifically includes:According to the system health degree model
With multiple ongoing basis monitoring datas in first setting time interval, calculated using second presetting method and obtain the
Multiple system real time health degrees of data in one setting time interval;Calculated using second presetting method and obtain described first
The variation tendency of system real time health degrees of data in setting time interval;According to the system in first setting time interval
The variation tendency of real time health degrees of data, judges the system health degree in the first setting time interval, obtains described first
System health degrees of data.Preferably, second presetting method, can be calculated using Spark internal memory Computational frame
The speed calculated is lifted, can also promote the speed of machine learning, second presetting method can be by preset rules to multiple
System real time health degrees of data is calculated, can such as calculate the extreme values of multiple system real time health degrees of data, average, median,
Variance etc..
Preferably, according to the healthy degrees of data of the first system, the system health model is detected and adjusted, with
Make the system health degree model more perfect, specifically, can be according to the healthy degrees of data of the first system, to the system
Health degree model carries out repeated detection and adjustment, until the degree of fitting of the system health degree model reaches predetermined value.
Preferably, after being adjusted to system health degree model, according to the system health degree model after adjustment, detection second is set
The system health degree fixed time in interval, generation detection and analysis report;Specifically, obtain first in the second setting time interval
Multiple ongoing basis monitoring datas, are specifically included:According to the core transaction system health degree model, the second presetting method is utilized
Calculate the multiple first core transaction system health degrees of data obtained in the second setting time interval;According to the multiple
One core transaction system health degrees of data and the basis monitoring regression model, are calculated using second presetting method and obtain institute
State multiple ongoing basis monitoring datas in the second setting time interval.Secondly according to many in second setting time interval
System health degree in individual ongoing basis monitoring data and the system health degree model, the second setting time of detection interval, it is raw
Into detection and analysis report, specifically include:According to multiple in the system health degree model and second setting time interval
Ongoing basis monitoring data, the multiple systems obtained in the second setting time interval is calculated using second presetting method real-time
Healthy degrees of data;The system real time health degree obtained in the second setting time interval is calculated using second presetting method
The variation tendency of data;According to the variation tendency of the system real time health degrees of data in second setting time interval, judge
System health degree in the second setting time interval, and generate detection and analysis report.Wherein, the detection and analysis report,
Can include multinomial content, such as reflection system real time health degree data, reflect general health degree level data, influence because
Element and its weight, the healthy suggestion of the lifting system etc..The influence factor can be system load, the IT utilizations of resources
One or more of rate, EMS memory occupation situation, user concurrent amount, system vulnerability etc., inspection is reflected in by the weight of these influence factors
Survey in analysis report, user can be made to be visually known the influence degree of each influence factor.
Alternatively, the first core transaction system health degrees of data, ongoing basis monitoring data and system real time health
Degrees of data, can be also used for that the core transaction system health model, basic monitoring data are updated and optimized as new parameter
Model and system health degree model, so as to above three model is kept good applicability, will not be because of long-term
Do not update and fail, while making detection more accurate and effective.
Finally, reported to user's displaying detection and analysis, it is preferable that the number of the real time health of the reflection system shown
According to or reflection system general health data shown in the form of percentage;If can not be shown in the form of percentage,
It can also be shown with word, such as " health status ", " hidden danger state " or " unhealthy condition ", so that user or O&M people
Member can be visually known the health status of system, and then can search failure for the health status and take corresponding arrange
Apply.When being shown with word, which kind of state belongs to health, hidden danger or unhealthy, can empirically be entered by operation maintenance personnel or user
Row setting, or set by corresponding with the numerical value of percents, such as under percents, it is with 90% and 60%
Critical value is divided, and when the first system health degrees of data is more than 90%, is considered as health status, during less than 60%, is considered as and is not good for
Health state, other situations belong to hidden danger state;Alternatively it is also possible to by way of normal distribution, the means extracted with machine
To set " health status ", " hidden danger state " or " unhealthy condition ".
Based on above content, the technique effect that the embodiment of the present invention one can be realized is:By setting up system health degree mould
Type is simultaneously calculated by presetting method, can monitor the running status of operational system, interval in setting time to operational system
Interior system health degree is tested and analyzed, and system health degree model is adjusted according to the data that detection and analysis is obtained, can
So that system health degree model is more perfect;Detection point is carried out to the health of system using the system health degree model after adjustment
Analysis, and generate detection and analysis report and shown to user, allow users to be visually known the health status of system operation, and according to
Factor of its health status further to failure situation and unhealthful situation is analyzed and judged, further to take phase
The measure answered.
Embodiment two
To the embodiment of the present invention one accordingly, Fig. 2 shows a kind of system health degree detection provided in an embodiment of the present invention
The structural representation of analytical equipment.As shown in Fig. 2 a kind of system health degree detection and analysis device, including:Model building module
101, data detection module 102, data analysis module 103, data display module 104.
Wherein, the model building module 101, for setting up system health degree model;The data detection module 102,
For according to the system health degree model, detecting the system health degree in setting time interval, obtaining the first system health degree
Data;The data analysis module 103, for according to the healthy degrees of data of the first system, entering to the system health model
Row detection and adjustment, make the degree of fitting of the system health degree model reach predetermined value, are also used for according to the system after adjustment
System health degree in system health degree model, the second setting time of detection interval, generation detection and analysis is reported;The data display
Module 104, for being reported to user's displaying detection and analysis.
Preferably, the model building module 101 is additionally operable to set up core transaction system health degree model and basis is monitored
Data model.Alternatively, the model building module 101 can be also used for using the first core transaction system health degrees of data,
Ongoing basis monitoring data and system real time health degrees of data to core transaction system health model, basic monitoring data model and
System health degree model is updated and optimized, so as to above three model is kept good applicability, will not
Because not updating and failing for a long time, while making detection more accurate and effective.
Preferably, the system health degree detection and analysis device also includes data collection module, the fortune for collection system
Dimension indicator data, and set up system health degree model, core transaction system health degree model and basic monitoring data model point
Not required first sample data, the second sample data and the 3rd sample data.
Alternatively, the data detection module 102 can be additionally used in set up system health degree model, core transaction system be good for
When Kang Du models and basic monitoring data model, to the first sample data, the second sample data and the 3rd sample data of collection
Quality testing is carried out, to ensure the reliability of the quality of data, reduces the error set up and existed during model.
Based on above content, what the embodiment of the present invention two can reach has the technical effect that:Built by model building module 101
Erection system health degree model, can make detection process have good basis, so as to quickly carry out health degree detection;The number
Calculated according to detection model 102 using presetting method, the running status of operational system can be monitored, to system in setting time
Health degree is tested and analyzed;The data display module 104 generation detection and analysis report shows to user, can make user or
Operation maintenance personnel is visually known the health status of system operation;The data analysis module 103 can be to system health degree model
Further detected and adjusted, so that system health degree model is more perfect and accurate, while can also be to being with after adjustment
System health degree model the health degree of system is tested and analyzed, generation detection and analysis report so that operation maintenance personnel can in time,
System health is visually known, to find and exclude in time hidden danger.
Finally it should be noted that:Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations;To the greatest extent
The present invention is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that:Its according to
The technical scheme described in foregoing embodiments can so be modified, or which part or all technical characteristic are entered
Row equivalent;And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme, it all should cover among the claim of the present invention and the scope of specification.
Claims (10)
1. a kind of system health degree determination method, it is characterised in that the system health degree determination method includes:
Set up system health degree model;
According to the system health degree model, the system health degree in the first setting time interval is detected, the first system is obtained and is good for
Health degrees of data;
According to the healthy degrees of data of the first system, the system health model is detected and adjusted, be good for the system
The degree of fitting of Kang Du models reaches predetermined value;
According to the system health degree model after adjustment, the system health degree in the second setting time interval, generation inspection are detected
Survey analysis report;
To user's displaying detection and analysis report.
2. system health degree determination method according to claim 1, it is characterised in that described to set up system health degree
Model, is specifically included:
First sample data are collected, the first sample data include basic monitoring data and corresponding with the basic monitoring data
The healthy degrees of data of second system, the healthy degrees of data of the second system is the data measured in advance;
Learn the relation of the basic monitoring data and the healthy degrees of data of the second system using the first presetting method;
The system health degree model is set up according to the relation of the basic monitoring data and the healthy degrees of data of the second system.
3. system health degree determination method according to claim 1, it is characterised in that described strong according to the system
System health degree in Kang Du models, the first setting time of detection interval, obtains the healthy degrees of data of the first system, specifically includes:
Obtain multiple ongoing basis monitoring datas in the first setting time interval;
According to multiple ongoing basis monitoring datas in the system health degree model and first setting time interval, utilize
Second presetting method calculates the multiple system real time health degrees of data obtained in the first setting time interval;
The system real time health degrees of data in the first setting time interval is obtained using second presetting method calculating
Variation tendency;
According to the variation tendency of the system real time health degrees of data in first setting time interval, first setting is judged
System health degree in time interval, obtains the healthy degrees of data of the first system.
4. system health degree determination method according to claim 1, it is characterised in that the institute according to after adjustment
State the system health degree in system health degree model, the second setting time of detection interval, generation detection and analysis report, specific bag
Include:
Obtain multiple ongoing basis monitoring datas in the second setting time interval;
According to the ongoing basis monitoring data in the system health degree model after adjustment and second setting time interval,
The multiple second system real time health number of degrees obtained in the second setting time interval are calculated using second presetting method
According to;
The system real time health degrees of data in the second setting time interval is obtained using second presetting method calculating
Variation tendency;
According to the variation tendency of the system real time health degrees of data in second setting time interval, second setting is judged
System health degree in time interval, and generate detection and analysis report.
5. the system health degree determination method according to claim 3 or 4, it is characterised in that the system includes core
Heart transaction system and peripheral system;The system health degree determination method also includes:Set up core transaction system health degree
Model and basic monitoring data model;
The multiple ongoing basis monitoring datas obtained in the first setting time interval, or when obtaining second setting
Between multiple ongoing basis monitoring datas in interval, specifically include:
According to the core transaction system health degree model, calculated using the second presetting method and obtain the first setting time area
Multiple first core transaction system health degrees of data in interior or described second setting time interval;
According to the multiple first core transaction system health degrees of data and the basic monitoring data model, described second is utilized
When presetting method calculates multiple ongoing basis monitoring datas or second setting obtained in the first setting time interval
Between multiple ongoing basis monitoring datas in interval.
6. system health degree determination method according to claim 5, it is characterised in that described to set up core transaction system
System health degree model, is specifically included:
Collect the second sample data, second sample data include basic monitoring data, application process data, daily record data and
Second core transaction system health degrees of data;Wherein, the basic monitoring data, the application process data and the daily record number
According to for independent variable, the second core transaction system health degrees of data is dependent variable;The second core transaction system health degree
Data are the data measured in advance;
Learn the relation between the independent variable and the dependent variable using the first presetting method;
The core transaction system health degree model is set up according to the relation of the independent variable and the dependent variable.
7. system health degree determination method according to claim 5, it is characterised in that the foundation basis monitoring number
According to model, specifically include:
The 3rd sample data is collected, the 3rd sample data includes the second core transaction system health degrees of data and described
Basic monitoring data;
Learn the pass of the second core transaction system health degrees of data and the basic monitoring data using the first presetting method
System;
The basis prison is set up according to the relation of the second core transaction system health degrees of data and the basic monitoring data
Control data model.
8. system health degree detection method according to claim 1, it is characterised in that described to set up system health degree model
Before, also include:
The healthy degrees of data of legacy system is collected, according to the healthy degrees of data of the legacy system, the shadow of analyzing influence system health degree
The weight of the factor of sound and the influence factor;
According to the influence factor and the weight of the influence factor, it is determined that needing the type of system health degree model set up;
Wherein, the type of the system health degree model includes:Multivariate regression models, Logic Regression Models, neural network model.
9. the system health degree determination method according to claim 2,6 or 7, it is characterised in that described first presets
Method is machine learning, and the algorithm of the machine learning is corresponding with the type of system health degree model, the system health degree mould
The type of type includes:Multivariate regression models, Logic Regression Models, neural network model.
10. a kind of system health degree tests and analyzes device, it is characterised in that the system health degree detection and analysis device includes:
Model building module, data detection module, data analysis module, data display module;
The model building module, for setting up system health degree model;
The data detection module, for according to the system health degree model, detecting the system in the first setting time interval
Health degree, obtains the healthy degrees of data of the first system;
The data analysis module, for according to the healthy degrees of data of the first system, being examined to the system health model
Survey and adjust, the degree of fitting of the system health degree model is reached predetermined value, be also used for being good for according to the system after adjustment
System health degree in Kang Du models, the second setting time of detection interval, generation detection and analysis report;
The data display module, for being reported to user's displaying detection and analysis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710113911.9A CN106951359A (en) | 2017-02-28 | 2017-02-28 | A kind of system health degree determination method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710113911.9A CN106951359A (en) | 2017-02-28 | 2017-02-28 | A kind of system health degree determination method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106951359A true CN106951359A (en) | 2017-07-14 |
Family
ID=59467041
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710113911.9A Pending CN106951359A (en) | 2017-02-28 | 2017-02-28 | A kind of system health degree determination method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106951359A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107545387A (en) * | 2017-07-18 | 2018-01-05 | 浙江百世技术有限公司 | A kind of express delivery website health degree detection method based on machine learning |
CN110489314A (en) * | 2019-07-05 | 2019-11-22 | 中国平安人寿保险股份有限公司 | Model method for detecting abnormality, device, computer equipment and storage medium |
CN110957044A (en) * | 2019-09-20 | 2020-04-03 | 上海派拉软件股份有限公司 | Health management method based on improved logistic regression model |
CN111209179A (en) * | 2020-04-23 | 2020-05-29 | 成都四方伟业软件股份有限公司 | Method, device and system for collecting and analyzing system operation and maintenance data |
CN112131071A (en) * | 2017-09-18 | 2020-12-25 | 华为技术有限公司 | Memory evaluation method and device |
CN112446570A (en) * | 2019-08-30 | 2021-03-05 | 顺丰科技有限公司 | Method, system, equipment and storage medium for constructing inventory health degree index |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120072781A1 (en) * | 2010-09-17 | 2012-03-22 | Oracle International Corporation | Predictive incident management |
CN102801548A (en) * | 2011-05-27 | 2012-11-28 | 腾讯科技(深圳)有限公司 | Intelligent early warning method, device and information system |
CN105335819A (en) * | 2015-10-21 | 2016-02-17 | 国家电网公司 | Information system risk early warning model construction method based on big data |
CN106021845A (en) * | 2016-04-18 | 2016-10-12 | 中国西安卫星测控中心 | Spacecraft health condition assessment method |
CN106095639A (en) * | 2016-05-30 | 2016-11-09 | 中国农业银行股份有限公司 | A kind of cluster subhealth state method for early warning and system |
CN106447210A (en) * | 2016-10-10 | 2017-02-22 | 国家电网公司 | Distribution network equipment health degree dynamic diagnosis method involving credibility evaluation |
-
2017
- 2017-02-28 CN CN201710113911.9A patent/CN106951359A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120072781A1 (en) * | 2010-09-17 | 2012-03-22 | Oracle International Corporation | Predictive incident management |
CN102801548A (en) * | 2011-05-27 | 2012-11-28 | 腾讯科技(深圳)有限公司 | Intelligent early warning method, device and information system |
CN105335819A (en) * | 2015-10-21 | 2016-02-17 | 国家电网公司 | Information system risk early warning model construction method based on big data |
CN106021845A (en) * | 2016-04-18 | 2016-10-12 | 中国西安卫星测控中心 | Spacecraft health condition assessment method |
CN106095639A (en) * | 2016-05-30 | 2016-11-09 | 中国农业银行股份有限公司 | A kind of cluster subhealth state method for early warning and system |
CN106447210A (en) * | 2016-10-10 | 2017-02-22 | 国家电网公司 | Distribution network equipment health degree dynamic diagnosis method involving credibility evaluation |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107545387A (en) * | 2017-07-18 | 2018-01-05 | 浙江百世技术有限公司 | A kind of express delivery website health degree detection method based on machine learning |
CN107545387B (en) * | 2017-07-18 | 2020-12-22 | 浙江百世技术有限公司 | Express delivery station health degree detection method based on machine learning |
CN112131071A (en) * | 2017-09-18 | 2020-12-25 | 华为技术有限公司 | Memory evaluation method and device |
CN112131071B (en) * | 2017-09-18 | 2024-05-17 | 华为技术有限公司 | Memory evaluation method and device |
CN110489314A (en) * | 2019-07-05 | 2019-11-22 | 中国平安人寿保险股份有限公司 | Model method for detecting abnormality, device, computer equipment and storage medium |
CN110489314B (en) * | 2019-07-05 | 2022-07-12 | 中国平安人寿保险股份有限公司 | Model anomaly detection method and device, computer equipment and storage medium |
CN112446570A (en) * | 2019-08-30 | 2021-03-05 | 顺丰科技有限公司 | Method, system, equipment and storage medium for constructing inventory health degree index |
CN112446570B (en) * | 2019-08-30 | 2024-04-19 | 顺丰科技有限公司 | Inventory health index construction method, system, equipment and storage medium |
CN110957044A (en) * | 2019-09-20 | 2020-04-03 | 上海派拉软件股份有限公司 | Health management method based on improved logistic regression model |
CN111209179A (en) * | 2020-04-23 | 2020-05-29 | 成都四方伟业软件股份有限公司 | Method, device and system for collecting and analyzing system operation and maintenance data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106951359A (en) | A kind of system health degree determination method and device | |
Wang et al. | Multi-criteria building energy performance benchmarking through variable clustering based compromise TOPSIS with objective entropy weighting | |
CN104573850B (en) | A kind of Power Plant Equipment state evaluating method | |
CN106447210B (en) | A kind of distribution net equipment health degree dynamic diagnosis method of meter and trust evaluation | |
CN106940678A (en) | A kind of system real time health degree analysis and assessment method and device | |
CN107451402A (en) | A kind of equipment health degree appraisal procedure and device based on alarm data analysis | |
CN108053110B (en) | PMU data-based transformer state online diagnosis method | |
CN106845526B (en) | A kind of relevant parameter Fault Classification based on the analysis of big data Fusion of Clustering | |
CN109540212A (en) | A kind of cable shaft status assessment early warning system and method based on big data | |
CN105095963A (en) | Method for accurately diagnosing and predicting fault of wind tunnel equipment | |
CN110865924B (en) | Health degree diagnosis method and health diagnosis framework for internal server of power information system | |
CN104218570A (en) | Method and system for online evaluating overall measuring errors of electric energy measuring device | |
CN116224137A (en) | UPS equipment fault early warning method and system | |
CN106777997A (en) | A kind of heath state of spacecraft appraisal procedure | |
CN112084220B (en) | Abnormality diagnosis method and device for electric energy metering device and readable storage medium | |
CN109597396A (en) | A kind of distribution transforming on-line fault diagnosis method based on high amount of traffic and transfer learning | |
CN111612371A (en) | Intelligent ammeter quality evaluation method based on analytic hierarchy process | |
CN113850475A (en) | Production equipment performance evaluation system based on big data | |
TW202100900A (en) | Intelligent learning mechanism and adaptive regulation system for liquid gas filling | |
CN118074126B (en) | Operation and maintenance monitoring method based on power grid information | |
CN117057644A (en) | Equipment production quality detection method and system based on characteristic matching | |
CN117933808A (en) | Production quality assessment method and system based on production process monitoring | |
CN110096723B (en) | High-voltage switch cabinet insulation state analysis method based on operation and maintenance detection big data | |
CN113868948A (en) | User-oriented dynamic threshold model training system and method | |
CN113033845B (en) | Construction method and device for power transmission resource co-construction and sharing |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170714 |