CN105207843B - A kind of website visiting quality determining method based on historic test results - Google Patents
A kind of website visiting quality determining method based on historic test results Download PDFInfo
- Publication number
- CN105207843B CN105207843B CN201510535106.6A CN201510535106A CN105207843B CN 105207843 B CN105207843 B CN 105207843B CN 201510535106 A CN201510535106 A CN 201510535106A CN 105207843 B CN105207843 B CN 105207843B
- Authority
- CN
- China
- Prior art keywords
- test
- website
- tested
- historic
- factor
- 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.)
- Expired - Fee Related
Links
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention proposes a kind of website visiting quality determining methods.Comprising: carry out multiple random test in testing time section, it is accumulated by the historic test results of corresponding testing time section；The prior distribution and the historic test results specified based on user choose tested website and test path, with the validity test for next testing time section, can effectively select test object, optimal inspection strategy improves the efficiency of quality testing.
Description
Technical field
The present invention relates to network technique field more particularly to a kind of website visiting quality testings based on historic test results
Method.
Background technique
With the fast development of the network information technology, more and more websites use content distribution website CDN (Content
Delivery Network) technology accelerates content distribution and transmission speed, improve the access speed and quality of website.Thus it also urges
The manufacturer of many special offer CDN services has been provided, the acceleration node of these large size CDN manufacturers is all over the world, and network environment is complicated,
Website data is huge, and fast and effeciently detection accelerates node failure to become the important foundation for guaranteeing CDN service quality.
Testing time section property test CDN caching website visiting quality is that auxiliary detection CDN accelerates under different network environments
A kind of method of node failure.However, under the constraint of limit test resource, in face of complicated network environment and huge CDN
The quality testing effect of caching Websites quantity, simple sequential testing or random test is poor, it is more difficult in short time range
Complete loop test.Therefore, it is necessary to a kind of website visiting quality determining method based on historic test results, to solve existing skill
Above-mentioned technical problem present in art.
Summary of the invention
The present invention provides a kind of website visiting quality determining method based on historic test results, can effectively select test
Object, optimal inspection strategy improve the efficiency of quality testing.
The technical solution adopted by the present invention is that:
A kind of website visiting quality determining method comprising: multiple random test is carried out in testing time section, is added up
Obtain corresponding to the historic test results of testing time section；The prior distribution and the historic test results specified based on user,
Tested website and test path are chosen, with the validity test for next testing time section.
Preferably, the prior distribution specified based on user and the historic test results choose tested website and survey
Path is tried, to specifically include: according to the historic test results, determine in the validity test for next testing time section
Failure probability distribution and the Failure probability distribution for counting test factor in measurement circuit in failure set of websites, in conjunction with described
Prior distribution chooses tested website and test path.
Preferably, the prior distribution specified based on user and the historic test results choose tested website and survey
Path is tried, to choose the tested website P in the validity test for next testing time sectionuAnd test pathRespectively
Are as follows:
Pu=(1- α) P (u)+α Q (u)；
Wherein, α, β are damping factor, P (u), P (fi) it is the prior distribution that user specifies, Q (u) is the failure website
Failure probability distribution in set, Q (fi) be measurement circuit in test factor Failure probability distribution.
Preferably, described to carry out multiple random test in testing time section, it is accumulated by corresponding testing time section
Historic test results in, it is described to carry out multiple random test in testing time section, specifically include: from set of websites U
In randomly select tested website u, the test factor f in random arrangement measurement circuiti, carry out random test.
Preferably, the historic test results include failure set of websites HdWith normal set of websites Hn, failure set of websites
Middle Failure probability distribution Q (u), the Failure probability distribution Q (f of test factor in measurement circuiti) be respectively as follows:
Wherein, u is the tested website randomly selected from set of websites U, fiFor the survey in the measurement circuit of random arrangement
The factor is tried, the access quality that website is tested in failure set of websites is less than given threshold, is tested website in normal set of websites
It accesses quality and is greater than given threshold.
Preferably, prior distribution P (u), P (fi) be respectively with distribution.
By adopting the above technical scheme, the present invention at least has following effect:
Website visiting quality determining method provided by the invention based on historic test results, in the pact of limit test resource
Under beam, in face of the caching Websites quantity of complicated network environment and huge website distribution content CDN, historic test results are utilized
The prior distribution specified with user, effectively selects test object, and optimal inspection strategy improves the detection effect of website visiting quality
Rate.
Detailed description of the invention
Fig. 1 is the process of the website visiting quality determining method based on historic test results of first embodiment of the invention
Figure；
Fig. 2 is the schematic diagram of random test in website visiting quality determining method shown in FIG. 1；
Fig. 3 is the flow chart of the website visiting quality determining method of second embodiment of the invention.
Specific embodiment
Further to illustrate the present invention to reach the technical means and efficacy that predetermined purpose is taken, below in conjunction with attached drawing
And preferred embodiment, the present invention is described in detail as after.
Website visiting quality determining method provided by the invention can be improved tested website visiting quality verification and measurement ratio and by
The coverage at survey grid station.Website visiting quality determining method of the invention and its each step will be described in detail belows.
Network environment when being accessed due to website visiting quality with it is closely related, has used in the agreement present invention below
It closes term and symbol is as follows.
(1) it is tested website
Due to needing to detect access quality of a large amount of websites under different test environment, entire set of websites note
For U, one of them tested website is denoted as u.
(2) testing time section
Due to needing to carry out intermittent inspection ground survey grid station with testing time section during quality testing, even the same net
Stand also can repeatedly test in different time.The testing time section T tested between moment t0 and t1 is [t0, t1].
(3) environment (test path) is tested
Test environment refer to test access website when network environment, access website when as test, network provider ISP
Etc. environmental factors be denoted as f1,f2,f3,f4.Because needing to select test factor in test, test factor is also referred to as tested
The factor.Integrated testability environment is denoted as F.So have, F=(f1, f2, f3, f4).All values for testing environment F constitute the phase of test
Space.Because testing the combination that environment depends on different test factors, this, which is equivalent between different test factors, constitutes one
Test path, so test environment is also referred to as test path.
(5) test result
Refer to that tested website u in testing time section T, tests the access quality under environment F.Test result is denoted as r.Access matter
R value 1 when not less than preset threshold is measured, otherwise value 0.Repeatedly after test, the failure set of websites of tested website u is obtainedWith normal set of websites
First embodiment
As shown in Figure 1, the website visiting quality determining method in the present embodiment includes: step S10: in testing time section
It is interior to carry out multiple random test, it is accumulated by the historic test results of corresponding testing time section.In no historic test results
With reference under conditions of, multiple random test can be carried out in testing time section to be convenient for after getting historic test results
The validity test of next side test period.Validity test in next step refers to the prior distribution specified in conjunction with user and should
Historic test results obtained in step need object to be tested to obtain next testing time section: tested website and test road
Diameter.
Step S20: the historic test results that the prior distribution and previous step specified based on user are obtained are chosen by survey grid
It stands and test path, with the validity test for next testing time section.It is possible thereby to the prior distribution specified with reference to user and
The historic test results obtained before carry out the validity test of next testing time section.
As shown in Fig. 2, carrying out multiple random test in testing time section, it is accumulated by corresponding testing time section
In historic test results, multiple random test is carried out in a testing time section, is specifically included: random from set of websites U
Choose tested website u, the test factor f in random arrangement measurement circuiti, carry out random test.It can be seen from the figure that website
Set U is that CDN caches website library 1, and tested website u is selected at random, and side test path 2 is by test factor 3, test in figure
The factor 4, test factor 5 and test factor 6 form.As seen in Figure 2, the value difference of test factor can change test
Path, to obtain different test results 7, test result 7 includes normal set of websites and failure set of websites, even if thus
For the same random website u, the test path of selection is different, and the test result obtained is different.
Second embodiment
As shown in figure 3, choosing tested website and test road based on prior distribution and historic test results that user specifies
Diameter, to specifically include: according to historic test results, determine failure website in the validity test for next testing time section
Failure probability distribution and the Failure probability distribution for counting test factor in measurement circuit in set, in conjunction with the prior distribution
Choose tested website and test path.The principle of optimality for the Test Strategy that website visiting quality determining method of the invention uses is such as
Under: assuming that within a certain period of time, it is tested relatively fixed the case where the combination of website and different test factors is broken down.If
In the presence of the historic test results obtained before, then, the tested website of failure should preferentially be tested, and occur event in test path
The test factor of barrier preferentially selects.Therefore, the website visiting quality determining method of the invention based on historic test results is made
Test Strategy optimum theory can be stated are as follows: website test result be belonged in normal set of websites, in certain time
It should be tested less in section；Otherwise website test result belongs in failure set of websites, it should more tests.For in measurement circuit
Test factor, count failure ratio when different values, the high value of test factor failure ratio should test more, conversely, should
Few test.
As preferred embodiment, based on prior distribution and historic test results that user specifies, choose tested website and
Test path, to choose tested website P in the validity test for next testing time sectionuAnd test pathIt is respectively as follows:
Pu=(1- α) P (u)+α Q (u)；
Wherein, α, β are damping factor, P (u), P (fi) be the prior distribution specified of user, such as which website, test because
By user, rule of thumb judgement should be tested son more in advance.Q (u) is Failure probability distribution in failure set of websites, Q (fi) it is to survey
Try the Failure probability distribution of test factor in route.Q(u),Q(fi) it is the probability of malfunction point counted according to historic test results
Cloth.Damping factor α is introduced in formula, the purpose of β on the one hand can be to the more tests in place out of order in history, on the other hand
Test is avoided to fall into local Historic space again.Test on the one hand will place out of order to history " stroll " again, another party
Face also will carry out new " strolling " to the place that do not tested.Effect for damping factor, with test factor fiFor illustrate,
Sample testing factor fiDistribution be prior distribution P (fi) and Failure probability distribution Q (fi) by the mixing point after damping factor combination
Cloth.It is sampled thus in accordance with mixed distribution, so that the test object generated is to meet failure distribution a bit, some are completely new
Random distribution.This also to a certain extent, prevents overfitting historic test results.Finally to the measurement circuit of generation, detection
Whether it is busy, if it is, should be given up.According to the method described above, the measurement circuit of current site is regenerated.
If no historic test results can use when starting test, damping factor value zero can be allowed, indicate only
Rely on prior distribution, Lai Jinhang random test.Suitable damping can be set more than threshold value in historic test results until adding up
The factor.In view of testing the variation of environment, some time span threshold value is specified, the history test knot before this threshold value is given up
Fruit.The historic test results of usually next testing time section reference were the test results that the section of the upper testing time of neighbour obtains.
Further, historic test results include failure set of websites HdWith normal set of websites Hn, in failure set of websites
Failure probability distribution Q (u), the Failure probability distribution Q (f of test factor in measurement circuiti) be respectively as follows:
Wherein, u is the tested website randomly selected from set of websites U, fiFor the survey in the measurement circuit of random arrangement
The factor is tried, the access quality that website is tested in failure set of websites is less than given threshold, is tested website in normal set of websites
It accesses quality and is greater than given threshold.Wherein, it is the function for indicating set sizes that #, which is placed on before set,.
In addition, if user is without special requirement, prior distribution P (u), P (fi) be respectively with distribution.
The detailed process of website visiting quality determining method shown in Fig. 3 based on historic test results are as follows:
Start website visiting quality testing program first, in the case where no historic test results, according to prior distribution
Specify a random website u, and the test factor f in random arrangement measurement circuit1, f2, f3, f4, obtained after carrying out random test
To the step S100 of test result.Obtained test result can update the step S101 of historic test results, newest to obtain
Close historic test results.Next, the Failure probability distribution situation in statistical history test result (u, F, r), executes respectively
Step S200: Failure probability distribution and step S200 in failure set of websites are obtained: obtains the event of test factor in measurement circuit
Hinder probability distribution, the Failure probability distribution of test factor reflects test factor fiFailure of the different values in measurement circuit
Probability.Then, the test object that next testing time section carries out validity test is generated.The probability of malfunction obtained according to previous step
The prior distribution that distribution and user specify executes step S203 respectively: choosing test website and step S204: choosing test road
Diameter, to choose corresponding test factor f1, f2, f3, f4.Finally effectively surveyed according to the test object that previous step obtains
Examination generates step S100: executing step S101 after getting test result: historic test results is updated, in order to next survey again
Try the validity test of period.Repeatedly, realize testing time section intermittently to carry out validity test, and each test
Period all carrys out optimal inspection strategy according to the historic test results of upper testing time section.
By the explanation of specific embodiment, the present invention can should be reached technological means that predetermined purpose is taken and
Effect is able to more deeply and specifically understand, however appended diagram is only to provide reference and description and is used, and is not used to this
Invention limits.
Claims (4)
1. a kind of website visiting quality determining method characterized by comprising
Multiple random test is carried out in testing time section, is accumulated by the historic test results of corresponding testing time section；
The prior distribution and the historic test results specified based on user choose tested website and test environment, to be used for down
The validity test of one testing time section；
Wherein, the prior distribution specified based on user and the historic test results choose tested website and test environment,
To specifically include in the validity test for next testing time section:
According to the historic test results, the failure for counting test factor in the Failure probability distribution and test environment of website is general
Rate distribution, to choose tested website and test environment in conjunction with the prior distribution, the test environment includes multiple and different survey
Try the factor；
Wherein, the prior distribution specified based on user and the historic test results choose tested website and test environment,
To choose the probability distribution P of the tested website in the validity test for next testing time sectionuIt is tested in test environment
The probability distribution of the factorIt is respectively as follows:
Pu=(1- α) P (u)+α Q (u)；
Wherein, α, β are damping factor, and P (u) is the tested website prior distribution that user specifies, P (fi) it is the test that user specifies
Test factor prior distribution in environment, Q (u) are the Failure probability distribution of the website, Q (fi) be test environment in test because
The Failure probability distribution of son.
2. website visiting quality determining method according to claim 1, which is characterized in that described in testing time section
Multiple random test is carried out, is accumulated by the historic test results of corresponding testing time section, it is described in a testing time
Multiple random test is carried out in section, is specifically included:
Tested website u is randomly selected from set of websites U, random arrangement tests the test factor f in environmenti, surveyed at random
Examination.
3. website visiting quality determining method according to claim 2, which is characterized in that the historic test results include
Failure set of websites and normal set of websites, the Failure probability distribution Q (u) of website, the failure for testing test factor in environment are general
Rate is distributed Q (fi) be respectively as follows:
Wherein, # is " ... number ",For failure set of websites,For normal set of websites, u is from set of websites U
The tested website randomly selected, fiFor the test factor in the test environment of random arrangement, website is tested in failure set of websites
Access quality be less than given threshold, the access quality that website is tested in normal set of websites is greater than given threshold.
4. website visiting quality determining method according to claim 1, which is characterized in that the user specify by survey grid
Stand prior distribution P (u) and test environment that the user specifies in test factor prior distribution P (fi) it is to be uniformly distributed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510535106.6A CN105207843B (en) | 2015-08-27 | 2015-08-27 | A kind of website visiting quality determining method based on historic test results |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510535106.6A CN105207843B (en) | 2015-08-27 | 2015-08-27 | A kind of website visiting quality determining method based on historic test results |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105207843A CN105207843A (en) | 2015-12-30 |
| CN105207843B true CN105207843B (en) | 2019-05-28 |
Family
ID=54955294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510535106.6A Expired - Fee Related CN105207843B (en) | 2015-08-27 | 2015-08-27 | A kind of website visiting quality determining method based on historic test results |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105207843B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101262373A (en) * | 2008-04-18 | 2008-09-10 | 北京启明星辰信息技术股份有限公司 | A computer network instrument location system and method |
| CN101741643A (en) * | 2009-12-24 | 2010-06-16 | 北京世纪互联宽带数据中心有限公司 | Content delivery network node detecting method and system |
| CN101915234A (en) * | 2010-07-16 | 2010-12-15 | 西安交通大学 | Method for diagnosing compressor-associated failure based on Bayesian network |
| CN102238604A (en) * | 2011-08-18 | 2011-11-09 | 无锡儒安科技有限公司 | Wireless sensor network failure diagnosis method |
| CN102255764A (en) * | 2011-09-02 | 2011-11-23 | 广东省电力调度中心 | Method and device for diagnosing transmission network failure |
| CN102819239A (en) * | 2011-06-08 | 2012-12-12 | 同济大学 | Intelligent fault diagnosis method of numerical control machine tool |
| WO2014064396A2 (en) * | 2012-10-26 | 2014-05-01 | Snecma | System for monitoring a set of components of a device |
| CN104063586A (en) * | 2014-06-11 | 2014-09-24 | 西北工业大学 | Polymorphic failure tree-based bayesian network failure prediction method |
| CN104486113A (en) * | 2014-12-11 | 2015-04-01 | 重庆邮电大学 | Fault link positioning method based on active greed and passive greed in sensor network |
-
2015
- 2015-08-27 CN CN201510535106.6A patent/CN105207843B/en not_active Expired - Fee Related
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101262373A (en) * | 2008-04-18 | 2008-09-10 | 北京启明星辰信息技术股份有限公司 | A computer network instrument location system and method |
| CN101741643A (en) * | 2009-12-24 | 2010-06-16 | 北京世纪互联宽带数据中心有限公司 | Content delivery network node detecting method and system |
| CN101915234A (en) * | 2010-07-16 | 2010-12-15 | 西安交通大学 | Method for diagnosing compressor-associated failure based on Bayesian network |
| CN102819239A (en) * | 2011-06-08 | 2012-12-12 | 同济大学 | Intelligent fault diagnosis method of numerical control machine tool |
| CN102238604A (en) * | 2011-08-18 | 2011-11-09 | 无锡儒安科技有限公司 | Wireless sensor network failure diagnosis method |
| CN102255764A (en) * | 2011-09-02 | 2011-11-23 | 广东省电力调度中心 | Method and device for diagnosing transmission network failure |
| WO2014064396A2 (en) * | 2012-10-26 | 2014-05-01 | Snecma | System for monitoring a set of components of a device |
| CN104063586A (en) * | 2014-06-11 | 2014-09-24 | 西北工业大学 | Polymorphic failure tree-based bayesian network failure prediction method |
| CN104486113A (en) * | 2014-12-11 | 2015-04-01 | 重庆邮电大学 | Fault link positioning method based on active greed and passive greed in sensor network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105207843A (en) | 2015-12-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104185840B (en) | It is used for being prioritized the mthods, systems and devices of multiple tests in lasting deployment streamline | |
| CN111181801B (en) | Node cluster testing method and device, electronic equipment and storage medium | |
| CN110601900A (en) | Network fault early warning method and device | |
| CN103974139B (en) | Method and device for detecting network faults of terminal and system | |
| JP2011515978A (en) | Estimating available end-to-end bandwidth in packet-switched communication networks | |
| KR102061833B1 (en) | Apparatus and method for investigating cyber incidents | |
| CN105025141A (en) | Method and device for testing electricity consumption of terminal | |
| CN103024798B (en) | AP performance test methods and device | |
| CN104426881B (en) | A kind of method and device detecting malicious act | |
| CN109743411A (en) | A kind of method, apparatus and storage medium of the dynamic dispatching IP agent pool under distributed environment | |
| CN113574830A (en) | Measuring impact of network deployment | |
| CN107920115A (en) | A kind of City-level IP localization methods based on time delay and geographical consistency constraint | |
| CN106375156A (en) | Power network traffic anomaly detection method and device | |
| CN109032954A (en) | A kind of user's choosing method, device, storage medium and the terminal of A/B test | |
| CN113647055A (en) | Measuring impact of network deployment | |
| CN105207843B (en) | A kind of website visiting quality determining method based on historic test results | |
| CN105471674B (en) | The method and system of testing server performance | |
| CN110620671A (en) | Encryption algorithm evaluation method, medium, device and apparatus | |
| CN106790077A (en) | A kind of DNS full flows kidnap the detection method and device of risk | |
| CN104918297B (en) | A kind of cell accessing method and equipment | |
| CN104462116B (en) | Data selection method and device | |
| CN107704373A (en) | A kind of data processing method and device | |
| CN104346263B (en) | A kind of terminal capabilities arrangement method and device | |
| CN103116708A (en) | Device and method for evaluating website experience | |
| CN105515910B (en) | A kind of Group Delay Measurement method and apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190528 Termination date: 20210827 |