CN107276849A - The method for analyzing performance and device of a kind of cluster - Google Patents
The method for analyzing performance and device of a kind of cluster Download PDFInfo
- Publication number
- CN107276849A CN107276849A CN201710451908.8A CN201710451908A CN107276849A CN 107276849 A CN107276849 A CN 107276849A CN 201710451908 A CN201710451908 A CN 201710451908A CN 107276849 A CN107276849 A CN 107276849A
- Authority
- CN
- China
- Prior art keywords
- server
- monitored
- influence
- index
- default capabilities
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0817—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/04—Processing captured monitoring data, e.g. for logfile generation
- H04L43/045—Processing captured monitoring data, e.g. for logfile generation for graphical visualisation of monitoring data
Abstract
This application provides a kind of method for analyzing performance of cluster and device, factor of influence represents the index change of default capabilities index in the present invention, total factor of influence is to be added the factor of influence of the corresponding each server to be monitored of each default capabilities index, the index change of the performance indications of a cluster can be represented, the performance change using distributed type assemblies before and after updating can be accessed.The present invention can automatically analyze the index change of the performance indications of distributed type assemblies, no longer need manually to obtain this time applying the performance indications that updates and can influence according to reasoning from logic, and then go to analyze the change of those impacted performance indications correspondence numerical value, reduce human cost.
Description
Technical field
The present invention relates to distributed type assemblies field, in particular, it is related to the method for analyzing performance and device of a kind of cluster.
Background technology
Distributed type assemblies, to reach the purpose of Optimum distribution formula cluster, can include regularly using updating using renewal
Application configuration is adjusted or application parameter is adjusted etc..After renewal, it can analyze and become using the performance of distributed type assemblies before and after updating
Change.
In the prior art, the performance change of analysis distribution formula cluster is changing by analytical performance index correspondence numerical value
To the performance change of distributed type assemblies, in general, distributed type assemblies performance indications include：Read or write speed or handling capacity etc..Point
It is that the artificial performance for obtaining this time influenceing using updating according to reasoning from logic refers to when analysing the change of performance indications correspondence numerical value
Mark, and then go to analyze the change of those impacted performance indications correspondence numerical value, in this way, the workload of technical staff
It is larger, human cost can be improved.
Therefore, one kind is needed badly before and after distributed type assemblies application updates, during the performance change of analysis distribution formula cluster, reduction
The method of cost of labor.
The content of the invention
In view of this, the present invention provides a kind of method for analyzing performance and device of cluster, to solve in the prior art, dividing
Before and after cloth cluster application updates, during the performance change of analysis distribution formula cluster, the problem of cost of labor is higher.
In order to solve the above technical problems, present invention employs following technical scheme：
A kind of method for analyzing performance of cluster, including：
Obtain server name list, first time period and second time period；Wherein, in the server name list
Record the server name of each server to be monitored；
Obtain in the first time period multiple numerical value of the corresponding each default capabilities index of each server to be monitored, with
And in the second time period the corresponding each default capabilities index of each server to be monitored multiple numerical value；
According in the first time period multiple numerical value of the corresponding each default capabilities index of each server to be monitored, with
And in the second time period the corresponding each default capabilities index of each server to be monitored multiple numerical value, using default scoring
Algorithm, analysis obtains representing the score value of the index rate of change of the corresponding each default capabilities index of each server to be monitored；Wherein,
The index rate of change be the corresponding numerical value of the corresponding each default capabilities index of each server to be monitored in cluster application more
Intensity of variation before and after new；
The score value of the corresponding each default capabilities index of each server to be monitored is multiplied by corresponding default capabilities index
Default weight, calculates the factor of influence for the index change for obtaining representing the corresponding each default capabilities index of each server to be monitored；
The factor of influence of the corresponding each server to be monitored of each default capabilities index is added, each is obtained and presets
Total factor of influence of performance indications.
Preferably, according in the first time period the corresponding each default capabilities index of each server to be monitored it is multiple
Numerical value and in the second time period the corresponding each default capabilities index of each server to be monitored multiple numerical value, use
Default scoring algorithm, analysis obtains representing the scoring of the index rate of change of the corresponding each default capabilities index of each server to be monitored
Value, including：
According to multiple numerical value of the corresponding each default capabilities index of each server to be monitored in the first time period, adopt
With the default scoring algorithm, analysis obtains representing the first statistical number of the corresponding each default capabilities index of each server to be monitored
Value；
According to multiple numerical value of the corresponding each default capabilities index of each server to be monitored in the second time period, adopt
With the default scoring algorithm, analysis obtains representing the second statistical number of the corresponding each default capabilities index of each server to be monitored
Value；
The rate of change of first statistic and second statistic is calculated, obtains representing each server to be monitored
The score value of the index rate of change of corresponding each default capabilities index.
Preferably, the factor of influence by the corresponding each server to be monitored of each default capabilities index is added, and is obtained
To after total factor of influence of each default capabilities index, in addition to：
Filter out total factor of influence be on the occasion of default capabilities index；
According to total factor of influence be on the occasion of each default capabilities index total factor of influence numerical values recited, will it is described always
Factor of influence be on the occasion of default capabilities index, be ranked up from high to low；
Filter out the default capabilities index that total factor of influence is negative value；
, will be described total according to numerical values recited of total factor of influence for total factor of influence of each default capabilities index of negative value
Factor of influence is the default capabilities index of negative value, is ranked up from high to low.
Preferably, the score value by the corresponding each default capabilities index of each server to be monitored is multiplied by corresponding default
The default weight of performance indications, calculates what the index for obtaining representing the corresponding each default capabilities index of each server to be monitored changed
After factor of influence, in addition to：
The factor of influence of the corresponding each default capabilities index of each server to be monitored is added, each service to be monitored is obtained
Total factor of influence of device.
Preferably, the factor of influence of the corresponding each default capabilities index of each server to be monitored is added, respectively treated
After the total factor of influence for monitoring server, in addition to：
Filter out total factor of influence be on the occasion of server to be monitored；
According to total factor of influence be on the occasion of each server to be monitored total factor of influence numerical values recited, will it is described always
Factor of influence be on the occasion of server to be monitored, be ranked up from high to low；
Filter out the server to be monitored that total factor of influence is negative value；
, will be described total according to numerical values recited of total factor of influence for total factor of influence of each server to be monitored of negative value
Factor of influence is the server to be monitored of negative value, is ranked up from high to low.
Preferably, the factor of influence of the corresponding each server to be monitored of each default capabilities index is added, obtains every
After total factor of influence of one default capabilities index, in addition to：
The default weight of the corresponding each default capabilities index of each server to be monitored is adjusted, each server phase to be monitored is obtained
The new default weight for each default capabilities index answered.
A kind of performance evaluation device of cluster, including：
First acquisition unit, for obtaining server name list, first time period and second time period；Wherein, institute
State the server name that each server to be monitored is recorded in server name list；
Second acquisition unit, for obtaining the corresponding each default capabilities of each server to be monitored in the first time period
Multiple numerical value of index and in the second time period the corresponding each default capabilities index of each server to be monitored it is multiple
Numerical value；
Analytic unit, for according to the corresponding each default capabilities index of each server to be monitored in the first time period
Multiple numerical value and in the second time period the corresponding each default capabilities index of each server to be monitored many numbers
Value, using default scoring algorithm, analysis obtains representing the index change of the corresponding each default capabilities index of each server to be monitored
The score value of rate；Wherein, the index rate of change is that the corresponding each default capabilities index of each server to be monitored is corresponding
Intensity of variation of the numerical value before and after cluster application updates；
First computing unit, for the score value of the corresponding each default capabilities index of each server to be monitored to be multiplied by accordingly
Default capabilities index default weight, calculate the index for obtaining representing the corresponding each default capabilities index of each server to be monitored
The factor of influence of change；
Second computing unit, for by the factor of influence phase of the corresponding each server to be monitored of each default capabilities index
Plus, obtain total factor of influence of each default capabilities index.
Preferably, the analytic unit includes：
First analytic unit, for according to the corresponding each default capabilities of each server to be monitored in the first time period
Multiple numerical value of index, using the default scoring algorithm, analysis obtains representing the corresponding each default property of each server to be monitored
First statistic of energy index；
Second analytic unit, for according to the corresponding each default capabilities of each server to be monitored in the second time period
Multiple numerical value of index, using the default scoring algorithm, analysis obtains representing the corresponding each default property of each server to be monitored
Second statistic of energy index；
Numerical calculation unit, the rate of change for calculating first statistic and second statistic, is obtained
Represent the score value of the index rate of change of the corresponding each default capabilities index of each server to be monitored.
Preferably, in addition to：
First screening unit, for second computing unit by the corresponding each clothes to be monitored of each default capabilities index
The factor of influence of business device is added, after the total factor of influence for obtaining each default capabilities index, filters out total factor of influence for just
The default capabilities index of value；
First sequencing unit, for according to total factor of influence be on the occasion of each default capabilities index total factor of influence
Numerical values recited, by total factor of influence be on the occasion of default capabilities index, be ranked up from high to low；
Second screening unit, for filtering out the default capabilities index that total factor of influence is negative value；
Second sequencing unit, for being total factor of influence of each default capabilities index of negative value according to total factor of influence
Numerical values recited, the default capabilities index that total factor of influence is negative value is ranked up from high to low.
Preferably, in addition to：
3rd computing unit, for first computing unit by the corresponding each default capabilities index of each server to be monitored
Score value be multiplied by the default weight of corresponding default capabilities index, calculating obtains representing each server to be monitored corresponding each pre-
If after the factor of influence of the index change of performance indications, by the influence of the corresponding each default capabilities index of each server to be monitored
The factor is added, and obtains total factor of influence of each server to be monitored.
Preferably, in addition to：
Third filtering unit, refers to the corresponding each default capabilities of each server to be monitored for the 3rd computing unit
Target factor of influence is added, after the total factor of influence for obtaining each server to be monitored, filter out total factor of influence be on the occasion of treat
Monitor server；
3rd sequencing unit, for according to total factor of influence be on the occasion of each server to be monitored total factor of influence
Numerical values recited, by total factor of influence be on the occasion of server to be monitored, be ranked up from high to low；
4th screening unit, for filtering out the server to be monitored that total factor of influence is negative value；
4th sequencing unit, for being total factor of influence of each server to be monitored of negative value according to total factor of influence
Numerical values recited, the server to be monitored that total factor of influence is negative value is ranked up from high to low.
Preferably, in addition to：
Adjustment unit, for second computing unit by the corresponding each server to be monitored of each default capabilities index
Factor of influence be added, after the total factor of influence for obtaining each default capabilities index, adjust each server to be monitored corresponding
The default weight of each default capabilities index, obtains the new default power of the corresponding each default capabilities index of each server to be monitored
Weight.
Compared to prior art, the invention has the advantages that：
The invention provides a kind of method for analyzing performance of cluster and device, factor of influence represents default capabilities in the present invention
The index change of index, total factor of influence is by the factor of influence of the corresponding each server to be monitored of each default capabilities index
It is added, the index change of the performance indications of a cluster can be represented, can accesses using distributed type assemblies before and after updating
Performance change.The present invention can automatically analyze the index change of the performance indications of distributed type assemblies, it is no longer necessary to which artificial basis is patrolled
The performance indications that reasoning obtains this time influenceing using updating are collected, and then go to analyze the corresponding numerical value of those impacted performance indications
Change, reduce human cost.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
A kind of method flow diagram of the method for analyzing performance for cluster that Fig. 1 provides for the present invention；
The method flow diagram of the method for analyzing performance for another cluster that Fig. 2 provides for the present invention；
A kind of structural representation of the performance evaluation device for cluster that Fig. 3 provides for the present invention；
The structural representation of the performance evaluation device for another cluster that Fig. 4 provides for the present invention；
The structural representation of the performance evaluation device for another cluster that Fig. 5 provides for the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
The embodiments of the invention provide a kind of method for analyzing performance of cluster, reference picture 1, including：
S101, the list of acquisition server name, first time period and second time period；
Wherein, the server name of each server to be monitored is recorded in the server name list.Such as server A, clothes
Be engaged in device B, server C etc..In addition, first time period and second time period are the different periods.First time period is manually to exist
The period that the application of cluster is chosen before updating, second time period is artificial one chosen after the application of cluster updates
Period.Wherein, the performance of distributed type assemblies before and after first time period and second time period should update carrying out analysis application
Set before change.
S102, many numbers for obtaining the corresponding each default capabilities index of each server to be monitored in the first time period
Value and in the second time period the corresponding each default capabilities index of each server to be monitored multiple numerical value；
Wherein, default capabilities index includes read or write speed, caching accounting, network traffics and internal memory accounting etc..Default capabilities
Index contains each performance indications of server to be monitored.One default capabilities index of one server to be monitored is one
Multiple numerical value are had in the individual period, such as first time period is 9 points to 10 points of morning, if the data of collection in five minutes, that
Have 12 data.
The multiple numerical value of one default capabilities index of one server to be monitored in a period of time constitute a property
In energy curve, the present embodiment, multiple numerical value of the drawing instrument by a default capabilities index in a period of time can be used
Be depicted as a performance curve, so can become apparent from observe the number of a default capabilities index in a period of time
Value changes situation, because the quantity of a corresponding default capabilities index of server to be monitored is multiple, therefore can be constituted a plurality of
Performance curve.
Further, it is also possible to which performance curve is depicted as into chart by chart system, and then can more intuitively it observe
The numerical value change situation of one default capabilities index in a period of time.
It should be noted is that, the default capabilities index of each server to be monitored is identical, including read-write speed
Degree, caching accounting, network traffics and internal memory accounting etc..
S103, analysis obtain representing the scoring of the index rate of change of the corresponding each default capabilities index of each server to be monitored
Value；
Specifically, according in the first time period the corresponding each default capabilities index of each server to be monitored it is multiple
Numerical value and in the second time period the corresponding each default capabilities index of each server to be monitored multiple numerical value, use
Default scoring algorithm, analysis obtains representing the scoring of the index rate of change of the corresponding each default capabilities index of each server to be monitored
Value；
Wherein, the index rate of change is each server to be monitored corresponding numerical value of each default capabilities index accordingly
Intensity of variation before and after cluster application updates.
Wherein, according to many numbers of the corresponding each default capabilities index of each server to be monitored in the first time period
Value and in the second time period the corresponding each default capabilities index of each server to be monitored multiple numerical value, using pre-
If scoring algorithm, analysis obtains representing the scoring of the index rate of change of the corresponding each default capabilities index of each server to be monitored
Value, including：
1) according to multiple numerical value of the corresponding each default capabilities index of each server to be monitored in the first time period,
Using the default scoring algorithm, analysis obtains representing the first statistics of the corresponding each default capabilities index of each server to be monitored
Numerical value；
Wherein, the first statistic can be the corresponding each default property of each server to be monitored in the first time period
Average value, maximum, summation or variance of multiple numerical value of energy index etc..
2) according to multiple numerical value of the corresponding each default capabilities index of each server to be monitored in the second time period,
Using the default scoring algorithm, analysis obtains representing the second statistics of the corresponding each default capabilities index of each server to be monitored
Numerical value；
Wherein, the second statistic can be the corresponding each default property of each server to be monitored in the second time period
Average value, maximum, summation or variance of multiple numerical value of energy index etc..
3) rate of change of first statistic and second statistic is calculated, obtains representing each service to be monitored
The score value of the index rate of change of the corresponding each default capabilities index of device.
For example, the first statistic is 5, and the second statistic is 10, then the first statistic and second system
The rate of change of count value is (10-5)/5=1, then score value is 1.
Wherein, default capabilities index is different, presets scoring algorithm difference, for example, when default capabilities index is writing speed
When, it is mean variation rate to preset scoring algorithm, and when default capabilities index is request number, it is in the period to preset scoring algorithm
Total number rate of change.
After the score value of first default capabilities index of server A to be monitored is obtained, using same method, calculate
Obtain representing the score value of the index rate of change of the corresponding each default capabilities index of each server to be monitored.
S104, the score value of the corresponding each default capabilities index of each server to be monitored is multiplied by corresponding default capabilities referred to
Target presets weight, calculate the influence of the index change for obtaining representing the corresponding each default capabilities index of each server to be monitored because
Son；
Wherein, the default weight of each default capabilities index may be identical, it is also possible to differs, each default capabilities index
Initial preset weight it is identical, but in user's input server name list, first time period and second time period, if
Want to pay close attention to certain several default capabilities index, the default weight for these default capabilities indexs paid close attention to can be tuned up,
If not wanting the default capabilities index paid close attention to, now, the default weight of each default capabilities index is identical.
It is existing to be preset accordingly for example, the score value of the corresponding each default capabilities index of each server to be monitored is multiplied by
The default weight of performance indications, calculates what the index for obtaining representing the corresponding each default capabilities index of each server to be monitored changed
The process of factor of influence includes：
First server to be monitored in server name list is taken, such as server A to be monitored, by server to be monitored
The score value of A first default capabilities index is multiplied by the default weight of the default capabilities index, obtains representing service to be monitored
The factor of influence of the index change of device A first default capabilities index.
Equally, using this computational methods, calculating obtains representing the corresponding each default capabilities index of each server to be monitored
Index change factor of influence.
S105, the factor of influence addition by the corresponding each server to be monitored of each default capabilities index, obtain each
Total factor of influence of individual default capabilities index.
For example, as default capabilities index include read or write speed, caching accounting, network traffics and internal memory accounting this four
Kind, the factor of influence of the read or write speed of each server to be monitored is added, total factor of influence of read or write speed is obtained, total influence
The factor represents application and updates front and rear, the index change of the read or write speed of cluster.
Using identical method, caching accounting, network traffics and internal memory accounting these three default capabilities are calculated respectively and are referred to
The total factor of influence of target.
This example provides a kind of method for analyzing performance of cluster, and the index that factor of influence represents default capabilities index becomes
Change, total factor of influence is to be added the factor of influence of the corresponding each server to be monitored of each default capabilities index, Neng Goudai
The index change of the performance indications of one cluster of table, can access the performance change using distributed type assemblies before and after updating.This
Invention can automatically analyze the index change of the performance indications of distributed type assemblies, it is no longer necessary to manually obtain this according to reasoning from logic
Secondary application updates the performance indications that can be influenceed, and then goes to analyze the change of those impacted performance indications correspondence numerical value, reduction
Human cost.
In addition, being manually to obtain this according to reasoning from logic during the change of analytical performance index correspondence numerical value in the prior art
Secondary application updates the performance indications that can be influenceed, and then goes to analyze the change of those impacted performance indications correspondence numerical value, this
Method can ignore the index change of the artificial performance indications obtained without reasoning.And this programme be analyzed by computer it is each
The index change of individual performance indications, is compared and manual analysis, is capable of the index change of more comprehensive analytical performance index.
Optionally, in another embodiment of the present invention, reference picture 2 respectively treats each default capabilities index to supervise accordingly
The factor of influence for surveying server is added, after the total factor of influence for obtaining each default capabilities index, in addition to：
S206, filter out total factor of influence be on the occasion of default capabilities index；
It should be noted that total factor of influence of default capabilities index may be negative value, it is also possible to be on the occasion of when default property
Can index total factor of influence be on the occasion of when, represent the performance boost of the default capabilities index as on the occasion of when default capabilities index
Total factor of influence be negative value when, the performance boost for representing the default capabilities index is negative value.
S207, according to total factor of influence be on the occasion of each default capabilities index total factor of influence numerical values recited, will
Total factor of influence be on the occasion of default capabilities index, be ranked up from high to low；
Specifically, after being ranked up from high to low, before ranking ten default capabilities index is filtered out, this ten default capabilities
Index is the most performance indications of performance boost.
S208, filter out total factor of influence be negative value default capabilities index；
S209, according to total factor of influence for negative value each default capabilities index total factor of influence numerical values recited, will
Total factor of influence is the default capabilities index of negative value, is ranked up from high to low；
Specifically, being the numerical value of total factor of influence according to each default capabilities index when being ranked up from high to low
The size of absolute value is ranked up.After the completion of sequence, before ranking ten default capabilities index is filtered out, this ten default capabilities refer to
It is designated as the most performance indications of hydraulic performance decline.
In the present embodiment, after total factor of influence of each default capabilities index is obtained, by default capabilities index according to
The size of the numerical value of total factor of influence and it is positive and negative be ranked up, performance boost most default capabilities index and property can be obtained
The default capabilities index that can be reduced by up to, can make technical staff intuitively see that cluster application updates front and rear, change greatly
Default capabilities index.
Optionally, in another embodiment of the present invention, after step S105, in addition to：
The default weight of the corresponding each default capabilities index of each server to be monitored is adjusted, each server phase to be monitored is obtained
The new default weight for each default capabilities index answered.
Specifically, filter out total factor of influence be on the occasion of, ten default capabilities index and total factor of influence are before ranking
Negative value, before ranking after ten default capabilities index, this 20 performance indications of manual analysis, if finding some performance indications,
When being updated without application, the index change of itself is just manually entered very greatly, now the default weight of these performance indications, this
When system according to the default weight being manually entered, can just adjust each server to be monitored each default capabilities index default power
Weight, obtains the new default weight of the corresponding each default capabilities index of each server to be monitored.
It should be noted that the default weight of the corresponding each default capabilities index of each server to be monitored of adjustment, obtains each
The new default weight of the corresponding each default capabilities index of server to be monitored, is the feedback mechanism in the present embodiment, i.e., can
Part is adjusted according to sequence and changes irrational default capabilities index.
In the present embodiment, the default weight of the corresponding each default capabilities index of each server to be monitored is adjusted, is respectively treated
The new default weight of the corresponding each default capabilities index of server is monitored, part default capabilities index can be avoided due to service
Device self reason and influence that the sequence to whole default capabilities index is caused.
Optionally, in another embodiment of the present invention, after step S104, in addition to：
The factor of influence of the corresponding each default capabilities index of each server to be monitored is added, each service to be monitored is obtained
Total factor of influence of device；
Filter out total factor of influence be on the occasion of server to be monitored；
According to total factor of influence be on the occasion of each server to be monitored total factor of influence numerical values recited, will always influence
The factor be on the occasion of server to be monitored, be ranked up from high to low；
Filter out the server to be monitored that total factor of influence is negative value；
According to numerical values recited of total factor of influence for total factor of influence of each server to be monitored of negative value, will always it influence
The factor is the server to be monitored of negative value, is ranked up from high to low.
Specifically, total factor of influence of each performance indications in cluster is analyzed in above-described embodiment, the present embodiment analysis
Total factor of influence of each server to be monitored.Calculating the method for total factor of influence of each server to be monitored is, will be each
The factor of influence of the corresponding each default capabilities index of server to be monitored is added, obtain total influence of each server to be monitored because
Son.
Hereafter, it is ranked up according to the sort method of default capabilities index.
It should be noted that total factor of influence of server to be monitored is on the occasion of representing the performance of the server to be monitored
Be promoted on the occasion of, when server to be monitored total factor of influence be negative value when, represent the performance boost of the server to be monitored as
Negative value.
When by total factor of influence be on the occasion of server to be monitored, after being ranked up from high to low, filter out ten before ranking
Server to be monitored, this ten servers to be monitored are the most server to be monitored of performance boost.When by total factor of influence
For the server to be monitored of negative value, after being ranked up from high to low, before ranking ten server to be monitored is filtered out, this ten are treated
It is the most server to be monitored of hydraulic performance decline to monitor server.
In addition, the numerical value of total factor of influence of each server to be monitored is added, total factor of influence of cluster is obtained, this
Individual total factor of influence represents the clustering performance change that this application updates.
In the present embodiment, after total factor of influence of each server to be monitored is obtained, by server to be monitored according to total shadow
Ring the numerical value of the factor size and it is positive and negative be ranked up, obtain the most server to be monitored of performance boost and hydraulic performance decline most
Many servers to be monitored, can make technical staff intuitively see that cluster application updates front and rear, the clothes to be monitored changed greatly
Business device.
Optionally, in another embodiment of the present invention provide a kind of cluster performance evaluation device, reference picture 3, including：
First acquisition unit 101, for obtaining server name list, first time period and second time period；Wherein,
The server name of each server to be monitored is recorded in the server name list；
Second acquisition unit 102, it is corresponding each default for obtaining each server to be monitored in the first time period
Multiple numerical value of performance indications and the corresponding each default capabilities index of each server to be monitored in the second time period
Multiple numerical value；
Analytic unit 103, for according to the corresponding each default capabilities of each server to be monitored in the first time period
Multiple numerical value of index and in the second time period the corresponding each default capabilities index of each server to be monitored it is multiple
Numerical value, using default scoring algorithm, the index that analysis obtains representing the corresponding each default capabilities index of each server to be monitored becomes
The score value of rate；Wherein, the index rate of change is the corresponding each default capabilities index correspondence of each server to be monitored
Numerical value cluster application update before and after intensity of variation；
First computing unit 104, for the score value of the corresponding each default capabilities index of each server to be monitored to be multiplied by
The default weight of corresponding default capabilities index, calculates and obtains representing the corresponding each default capabilities index of each server to be monitored
The factor of influence of index change；
Second computing unit 105, for by the influence of the corresponding each server to be monitored of each default capabilities index because
Son is added, and obtains total factor of influence of each default capabilities index.
Optionally, in another embodiment of the present invention, analytic unit 103 includes：
First analytic unit, for according to the corresponding each default capabilities of each server to be monitored in the first time period
Multiple numerical value of index, using the default scoring algorithm, analysis obtains representing the corresponding each default property of each server to be monitored
First statistic of energy index；
Second analytic unit, for according to the corresponding each default capabilities of each server to be monitored in the second time period
Multiple numerical value of index, using the default scoring algorithm, analysis obtains representing the corresponding each default property of each server to be monitored
Second statistic of energy index；
Numerical calculation unit, the rate of change for calculating first statistic and second statistic, is obtained
Represent the score value of the index rate of change of the corresponding each default capabilities index of each server to be monitored.
This example provides a kind of performance evaluation device of cluster, and the index that factor of influence represents default capabilities index becomes
Change, total factor of influence is to be added the factor of influence of the corresponding each server to be monitored of each default capabilities index, Neng Goudai
The index change of the performance indications of one cluster of table, can access the performance change using distributed type assemblies before and after updating.This
Invention can automatically analyze the index change of the performance indications of distributed type assemblies, it is no longer necessary to manually obtain this according to reasoning from logic
Secondary application updates the performance indications that can be influenceed, and then goes to analyze the change of those impacted performance indications correspondence numerical value, reduction
Human cost.
It should be noted that in the present embodiment unit the course of work, refer to interior in the corresponding embodiments of Fig. 1
Hold, will not be repeated here.
Optionally, in another embodiment of the present invention, reference picture 4, in addition to：
First screening unit 106, respectively treats each default capabilities index for second computing unit 105 accordingly
The factor of influence for monitoring server is added, after the total factor of influence for obtaining each default capabilities index, filter out total influence because
Son be on the occasion of default capabilities index；
First sequencing unit 107, for according to total factor of influence be on the occasion of each default capabilities index total influence because
Son numerical values recited, by total factor of influence be on the occasion of default capabilities index, be ranked up from high to low；
Second screening unit 108, for filtering out the default capabilities index that total factor of influence is negative value；
Second sequencing unit 109, for according to total factor of influence for negative value each default capabilities index total influence because
The numerical values recited of son, the default capabilities index that total factor of influence is negative value is ranked up from high to low.
In the present embodiment, after total factor of influence of each default capabilities index is obtained, by default capabilities index according to
The size of the numerical value of total factor of influence and it is positive and negative be ranked up, performance boost most default capabilities index and property can be obtained
The default capabilities index that can be reduced by up to, can make technical staff intuitively see that cluster application updates front and rear, change greatly
Default capabilities index.
It should be noted that in the present embodiment unit the course of work, refer to interior in the corresponding embodiments of Fig. 2
Hold, will not be repeated here.
Optionally, in another embodiment of the present invention, in addition to：
Adjustment unit, for second computing unit 105 by the corresponding each clothes to be monitored of each default capabilities index
The factor of influence of business device is added, after the total factor of influence for obtaining each default capabilities index, adjusts each server phase to be monitored
The default weight for each default capabilities index answered, obtains the new of the corresponding each default capabilities index of each server to be monitored and presets
Weight.
In the present embodiment, the default weight of the corresponding each default capabilities index of each server to be monitored is adjusted, is respectively treated
The new default weight of the corresponding each default capabilities index of server is monitored, part default capabilities index can be avoided due to service
Device self reason and influence that the sequence to whole default capabilities index is caused.
It should be noted that in the present embodiment unit the course of work, refer to the content in above-described embodiment,
This is repeated no more.
Optionally, in another embodiment of the present invention, reference picture 5, in addition to：
3rd computing unit 111, for first computing unit 104 by the corresponding each default property of each server to be monitored
Can the score value of index be multiplied by the default weight of corresponding default capabilities index, calculating obtains representing each server to be monitored corresponding
Each default capabilities index index change factor of influence after, will the corresponding each default capabilities index of each server to be monitored
Factor of influence be added, obtain total factor of influence of each server to be monitored.
Third filtering unit 112, it is for the 3rd computing unit 111 that each server to be monitored is corresponding each default
The factor of influence of performance indications is added, after the total factor of influence for obtaining each server to be monitored, filters out total factor of influence for just
The server to be monitored of value；
3rd sequencing unit 113, for according to total factor of influence be on the occasion of each server to be monitored total influence because
Son numerical values recited, by total factor of influence be on the occasion of server to be monitored, be ranked up from high to low；
4th screening unit 114, for filtering out the server to be monitored that total factor of influence is negative value；
4th sequencing unit 115, for according to total factor of influence for negative value each server to be monitored total influence because
The numerical values recited of son, the server to be monitored that total factor of influence is negative value is ranked up from high to low.
In the present embodiment, after total factor of influence of each server to be monitored is obtained, by server to be monitored according to total shadow
Ring the numerical value of the factor size and it is positive and negative be ranked up, obtain the most server to be monitored of performance boost and hydraulic performance decline most
Many servers to be monitored, can make technical staff intuitively see that cluster application updates front and rear, the clothes to be monitored changed greatly
Business device.
It should be noted that in the present embodiment unit the course of work, refer to the content in above-described embodiment,
This is repeated no more.
The foregoing description of the disclosed embodiments, enables professional and technical personnel in the field to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (12)
1. a kind of method for analyzing performance of cluster, it is characterised in that including：
Obtain server name list, first time period and second time period；Wherein, recorded in the server name list
The server name of each server to be monitored；
Obtain multiple numerical value, the Yi Ji of the corresponding each default capabilities index of each server to be monitored in the first time period
Multiple numerical value of the corresponding each default capabilities index of each server to be monitored in the second time period；
According to multiple numerical value, the Yi Ji of the corresponding each default capabilities index of each server to be monitored in the first time period
Multiple numerical value of the corresponding each default capabilities index of each server to be monitored, are calculated using default scoring in the second time period
Method, analysis obtains representing the score value of the index rate of change of the corresponding each default capabilities index of each server to be monitored；Wherein, institute
Stating index rate of change, the corresponding numerical value of each default capabilities index updates in cluster application accordingly for each server to be monitored
Front and rear intensity of variation；
The score value of the corresponding each default capabilities index of each server to be monitored is multiplied by the default of corresponding default capabilities index
Weight, calculates the factor of influence for the index change for obtaining representing the corresponding each default capabilities index of each server to be monitored；
The factor of influence of the corresponding each server to be monitored of each default capabilities index is added, each default capabilities is obtained
Total factor of influence of index.
2. method for analyzing performance according to claim 1, it is characterised in that according to respectively waiting to supervise in the first time period
Survey the multiple numerical value and each server phase to be monitored in the second time period of the corresponding each default capabilities index of server
The multiple numerical value for each default capabilities index answered, using default scoring algorithm, analysis obtains representing each server to be monitored corresponding
Each default capabilities index index rate of change score value, including：
According to multiple numerical value of the corresponding each default capabilities index of each server to be monitored in the first time period, using institute
Default scoring algorithm is stated, analysis obtains representing the first statistic of the corresponding each default capabilities index of each server to be monitored；
According to multiple numerical value of the corresponding each default capabilities index of each server to be monitored in the second time period, using institute
Default scoring algorithm is stated, analysis obtains representing the second statistic of the corresponding each default capabilities index of each server to be monitored；
The rate of change of first statistic and second statistic is calculated, obtains representing each server to be monitored corresponding
Each default capabilities index index rate of change score value.
3. method for analyzing performance according to claim 1, it is characterised in that described that each default capabilities index is corresponding
The factor of influence of each server to be monitored be added, after the total factor of influence for obtaining each default capabilities index, in addition to：
Filter out total factor of influence be on the occasion of default capabilities index；
According to total factor of influence be on the occasion of each default capabilities index total factor of influence numerical values recited, by total influence
The factor be on the occasion of default capabilities index, be ranked up from high to low；
Filter out the default capabilities index that total factor of influence is negative value；
According to numerical values recited of total factor of influence for total factor of influence of each default capabilities index of negative value, by total influence
The factor is the default capabilities index of negative value, is ranked up from high to low.
4. method for analyzing performance according to claim 1, it is characterised in that described that each server to be monitored is each accordingly
The score value of default capabilities index is multiplied by the default weight of corresponding default capabilities index, and calculating obtains representing each service to be monitored
After the factor of influence of the index change of the corresponding each default capabilities index of device, in addition to：
The factor of influence of the corresponding each default capabilities index of each server to be monitored is added, each server to be monitored is obtained
Total factor of influence.
5. method for analyzing performance according to claim 4, it is characterised in that each server to be monitored is corresponding each pre-
If the factor of influence of performance indications is added, after the total factor of influence for obtaining each server to be monitored, in addition to：
Filter out total factor of influence be on the occasion of server to be monitored；
According to total factor of influence be on the occasion of each server to be monitored total factor of influence numerical values recited, by total influence
The factor be on the occasion of server to be monitored, be ranked up from high to low；
Filter out the server to be monitored that total factor of influence is negative value；
According to numerical values recited of total factor of influence for total factor of influence of each server to be monitored of negative value, by total influence
The factor is the server to be monitored of negative value, is ranked up from high to low.
6. method for analyzing performance according to claim 1, it is characterised in that each default capabilities index is each accordingly
The factor of influence of server to be monitored is added, after the total factor of influence for obtaining each default capabilities index, in addition to：
The default weight of the corresponding each default capabilities index of each server to be monitored is adjusted, each server to be monitored is obtained corresponding
The new default weight of each default capabilities index.
7. a kind of performance evaluation device of cluster, it is characterised in that including：
First acquisition unit, for obtaining server name list, first time period and second time period；Wherein, the clothes
The server name of each server to be monitored is recorded in business device name list；
Second acquisition unit, for obtaining the corresponding each default capabilities index of each server to be monitored in the first time period
Multiple numerical value and in the second time period the corresponding each default capabilities index of each server to be monitored many numbers
Value；
Analytic unit, for according in the first time period the corresponding each default capabilities index of each server to be monitored it is many
Individual numerical value and in the second time period the corresponding each default capabilities index of each server to be monitored multiple numerical value, adopt
Default scoring algorithm is used, analysis obtains representing commenting for the index rate of change of the corresponding each default capabilities index of each server to be monitored
Score value；Wherein, the index rate of change is that the corresponding numerical value of each default capabilities index exists each server to be monitored accordingly
Cluster application updates front and rear intensity of variation；
First computing unit, it is corresponding pre- for the score value of the corresponding each default capabilities index of each server to be monitored to be multiplied by
If the default weight of performance indications, the index change for obtaining representing the corresponding each default capabilities index of each server to be monitored is calculated
Factor of influence；
Second computing unit, for the factor of influence of the corresponding each server to be monitored of each default capabilities index to be added,
Obtain total factor of influence of each default capabilities index.
8. performance evaluation device according to claim 7, it is characterised in that the analytic unit includes：
First analytic unit, for according to the corresponding each default capabilities index of each server to be monitored in the first time period
Multiple numerical value, using the default scoring algorithm, analysis, which obtains representing the corresponding each default capabilities of each server to be monitored, to be referred to
The statistic of target first；
Second analytic unit, for according to the corresponding each default capabilities index of each server to be monitored in the second time period
Multiple numerical value, using the default scoring algorithm, analysis, which obtains representing the corresponding each default capabilities of each server to be monitored, to be referred to
The statistic of target second；
Numerical calculation unit, the rate of change for calculating first statistic and second statistic, is represented
The score value of the index rate of change of the corresponding each default capabilities index of each server to be monitored.
9. performance evaluation device according to claim 7, it is characterised in that also include：
First screening unit, for second computing unit by the corresponding each server to be monitored of each default capabilities index
Factor of influence be added, after the total factor of influence for obtaining each default capabilities index, filter out total factor of influence be on the occasion of
Default capabilities index；
First sequencing unit, for according to total factor of influence be on the occasion of each default capabilities index total factor of influence numerical value
Size, by total factor of influence be on the occasion of default capabilities index, be ranked up from high to low；
Second screening unit, for filtering out the default capabilities index that total factor of influence is negative value；
Second sequencing unit, for according to numerical value of total factor of influence for total factor of influence of each default capabilities index of negative value
Size, the default capabilities index that total factor of influence is negative value is ranked up from high to low.
10. performance evaluation device according to claim 7, it is characterised in that also include：
3rd computing unit, for first computing unit commenting the corresponding each default capabilities index of each server to be monitored
Score value is multiplied by the default weight of corresponding default capabilities index, and calculating obtains representing the corresponding each default property of each server to be monitored
After the factor of influence of the index change of energy index, by the factor of influence of the corresponding each default capabilities index of each server to be monitored
It is added, obtains total factor of influence of each server to be monitored.
11. performance evaluation device according to claim 10, it is characterised in that also include：
Third filtering unit, for the 3rd computing unit by the corresponding each default capabilities index of each server to be monitored
Factor of influence is added, after the total factor of influence for obtaining each server to be monitored, filter out total factor of influence be on the occasion of it is to be monitored
Server；
3rd sequencing unit, for according to total factor of influence be on the occasion of each server to be monitored total factor of influence numerical value
Size, by total factor of influence be on the occasion of server to be monitored, be ranked up from high to low；
4th screening unit, for filtering out the server to be monitored that total factor of influence is negative value；
4th sequencing unit, for according to numerical value of total factor of influence for total factor of influence of each server to be monitored of negative value
Size, the server to be monitored that total factor of influence is negative value is ranked up from high to low.
12. performance evaluation device according to claim 7, it is characterised in that also include：
Adjustment unit, for second computing unit by the shadow of the corresponding each server to be monitored of each default capabilities index
Ring the factor to be added, after the total factor of influence for obtaining each default capabilities index, adjust each server to be monitored corresponding each pre-
If the default weight of performance indications, the new default weight of the corresponding each default capabilities index of each server to be monitored is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710451908.8A CN107276849A (en) | 2017-06-15 | 2017-06-15 | The method for analyzing performance and device of a kind of cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710451908.8A CN107276849A (en) | 2017-06-15 | 2017-06-15 | The method for analyzing performance and device of a kind of cluster |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107276849A true CN107276849A (en) | 2017-10-20 |
Family
ID=60067231
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710451908.8A Pending CN107276849A (en) | 2017-06-15 | 2017-06-15 | The method for analyzing performance and device of a kind of cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107276849A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109951506A (en) * | 2017-12-20 | 2019-06-28 | 中移(苏州)软件技术有限公司 | A kind of appraisal procedure and equipment of storage cluster performance |
| WO2022001257A1 (en) * | 2020-06-28 | 2022-01-06 | 中兴通讯股份有限公司 | Data analysis method, electronic device, and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102624570A (en) * | 2012-04-27 | 2012-08-01 | 杭州东信北邮信息技术有限公司 | Monitoring system and method for detecting availability of web server |
| CN105260253A (en) * | 2015-09-06 | 2016-01-20 | 浪潮集团有限公司 | Server failure measurement and calculation method and device |
| US20160034492A1 (en) * | 2012-12-21 | 2016-02-04 | Zetta, Inc. | Systems and methods for on-demand data storage |
| CN105515812A (en) * | 2014-10-15 | 2016-04-20 | 中兴通讯股份有限公司 | Fault processing method of resources and device |
| CN105843724A (en) * | 2016-03-17 | 2016-08-10 | 杭州优云软件有限公司 | Monitoring state index compression analysis method of IT (Information Technology) system |
| CN105915405A (en) * | 2016-03-29 | 2016-08-31 | 深圳市中博科创信息技术有限公司 | Large-scale cluster node performance monitoring system |
| CN105975295A (en) * | 2016-04-25 | 2016-09-28 | 乐视控股（北京）有限公司 | Application updating method and apparatus used for host cluster and server device |
| CN106603299A (en) * | 2016-12-28 | 2017-04-26 | 北京奇艺世纪科技有限公司 | Service health index generation method and service health index generation device |
-
2017
- 2017-06-15 CN CN201710451908.8A patent/CN107276849A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102624570A (en) * | 2012-04-27 | 2012-08-01 | 杭州东信北邮信息技术有限公司 | Monitoring system and method for detecting availability of web server |
| US20160034492A1 (en) * | 2012-12-21 | 2016-02-04 | Zetta, Inc. | Systems and methods for on-demand data storage |
| CN105515812A (en) * | 2014-10-15 | 2016-04-20 | 中兴通讯股份有限公司 | Fault processing method of resources and device |
| CN105260253A (en) * | 2015-09-06 | 2016-01-20 | 浪潮集团有限公司 | Server failure measurement and calculation method and device |
| CN105843724A (en) * | 2016-03-17 | 2016-08-10 | 杭州优云软件有限公司 | Monitoring state index compression analysis method of IT (Information Technology) system |
| CN105915405A (en) * | 2016-03-29 | 2016-08-31 | 深圳市中博科创信息技术有限公司 | Large-scale cluster node performance monitoring system |
| CN105975295A (en) * | 2016-04-25 | 2016-09-28 | 乐视控股（北京）有限公司 | Application updating method and apparatus used for host cluster and server device |
| CN106603299A (en) * | 2016-12-28 | 2017-04-26 | 北京奇艺世纪科技有限公司 | Service health index generation method and service health index generation device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109951506A (en) * | 2017-12-20 | 2019-06-28 | 中移(苏州)软件技术有限公司 | A kind of appraisal procedure and equipment of storage cluster performance |
| CN109951506B (en) * | 2017-12-20 | 2021-11-30 | 中移(苏州)软件技术有限公司 | Method and equipment for evaluating performance of storage cluster |
| WO2022001257A1 (en) * | 2020-06-28 | 2022-01-06 | 中兴通讯股份有限公司 | Data analysis method, electronic device, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104636450B (en) | A kind of database table space monitoring method | |
| CN106600344B (en) | Method and device for acquiring active user data of target product | |
| CN103984726B (en) | A kind of local correction method of data base's implement plan | |
| CN106202444A (en) | A kind of implementation method of data base's O&M monitoring | |
| CN108984370A (en) | A kind of method and apparatus of determining monitoring threshold value | |
| CN104679590B (en) | Map optimization method and device in distributed computing system | |
| CN104217004B (en) | The monitoring method and device of a kind of database focus of transaction system | |
| CN107276849A (en) | The method for analyzing performance and device of a kind of cluster | |
| CN105930491A (en) | Monitoring data storage method based on time sequence database InfluxDB | |
| CN104881477B (en) | A kind of application data space uses evaluation method | |
| CN106502871A (en) | The alarm threshold dynamic configuration system of supervisory systems and method | |
| US20130263136A1 (en) | Information processing system and processing method for use therewith | |
| CN105867951A (en) | Data processing method and device as well as screen designer | |
| CN107147721A (en) | A kind of Audit data machining system of distributed deployment | |
| CN106951360B (en) | Data statistical integrity calculation method and system | |
| CN112398700A (en) | Service degradation method and device, storage medium and computer equipment | |
| CN106530107A (en) | Agricultural-big-data-based method and apparatus for generating growing progress of crops | |
| CN107193749A (en) | Method of testing, device and equipment | |
| CN107704371A (en) | A kind of management method, device and the equipment of storage medium and storage system | |
| CN116126415A (en) | Scene-based cloud component configuration recommendation method and system | |
| CN111861397A (en) | Intelligent scheduling platform for client visit | |
| CN106599005A (en) | Data archiving method and device | |
| CN107147542A (en) | A kind of information generating method and device | |
| CN108763273A (en) | A kind of Alpine Grasslands data processing method and management system | |
| CN101605049A (en) | The processing method of network management data statistical analysis index and device, data management system |
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 |
Application publication date: 20171020 |
|
| RJ01 | Rejection of invention patent application after publication |