CN102447593B - Test method and device - Google Patents

Test method and device Download PDF

Info

Publication number
CN102447593B
CN102447593B CN 201110372395 CN201110372395A CN102447593B CN 102447593 B CN102447593 B CN 102447593B CN 201110372395 CN201110372395 CN 201110372395 CN 201110372395 A CN201110372395 A CN 201110372395A CN 102447593 B CN102447593 B CN 102447593B
Authority
CN
China
Prior art keywords
test
index
matrix
corresponding
evaluation
Prior art date
Application number
CN 201110372395
Other languages
Chinese (zh)
Other versions
CN102447593A (en
Inventor
黄志忠
Original Assignee
福建星网锐捷网络有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 福建星网锐捷网络有限公司 filed Critical 福建星网锐捷网络有限公司
Priority to CN 201110372395 priority Critical patent/CN102447593B/en
Publication of CN102447593A publication Critical patent/CN102447593A/en
Application granted granted Critical
Publication of CN102447593B publication Critical patent/CN102447593B/en

Links

Abstract

本发明公开了一种测试方法及装置,用以解决现有技术中对网络设备测试的准确性低的问题。 The present invention discloses a testing method and apparatus to solve the prior art low accuracy network device testing problem. 该方法对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值,根据每个指标之间的相对重要性程度值构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集,并确定测试的每个指标在每个性能等级上的百分比,据此确定评判矩阵,将确定的权重集与评判矩阵相乘,得到该网络设备的性能评判集,并将其中最大的元素对应的性能等级确定为得到的测试结果。 The method for each indicator test multiple network devices to obtain a plurality of indicators each corresponding test values, the fuzzy evaluation matrix configuration based on the relative value between the degree of importance of each index, constructed in accordance with the fuzzy weight matrix determining evaluation weights set, and tested to determine the percentage of each indicator on each performance level, whereby evaluation matrix is ​​determined, multiplied by the determined set of weights and evaluation matrix, to obtain performance evaluation set of the network device, and wherein the largest element determining a corresponding performance level for the test results obtained. 通过上述方法,测试装置对每个指标都进行了测试,并综合了各个指标对网络设备的综合性能的影响,因此可以对网络设备的综合性能进行准确的测试。 By the above method, the test device for each indicator were tested, and the combination of overall performance impact indicators each network device, it can accurately test the overall performance of the network device.

Description

一种测试方法及装置 A testing method and apparatus

技术领域 FIELD

[0001] 本发明涉及通信技术领域,尤其涉及一种测试方法及装置。 [0001] The present invention relates to communication technologies, and particularly to a method and apparatus for testing.

背景技术 Background technique

[0002] 在通信网络中存在着各种各样的网络设备,网络设备综合性能的优劣往往决定了通信网络的质量。 [0002] There are various network devices, network equipment overall performance in the communications network often determines the quality of the merits of the communication network. 为了选用综合性能比较好的网络设备,以提高通信网络的质量,就需要对各个备选的网络设备进行测试。 In order to better overall performance selected network device, to improve the quality of the communications network, it is necessary to test each candidate network devices.

[0003] 以路由器为例进行说明。 [0003] to the router as an example. 影响路由器综合性能的指标存在很多种,例如路由器的吞吐量、时延等指标会影响路由器的综合性能。 Presence affect the overall performance of the router are many indicators, such as throughput, latency and other indicators of the router will affect the overall performance of the router.

[0004] 现有技术中对路由器进行测试的方法为:针对某一个或者某几个指标,对路由器进行测试,用户根据得到的测试数据,人为的判断路由器综合性能的优劣。 Method [0004] The prior art for testing the router is: for one or a few metrics, the router test, user test data obtained overall performance artificial router determines merits.

[0005] 然而,路由器的各个指标或多或少都会影响路由器的综合性能,只针对某一个或某几个指标进行测试必然不能准确的反映出路由器的综合性能。 [0005] However, the router's various indicators more or less affect the overall performance of the router, but can not necessarily be tested accurately reflect the overall performance router for one or a few indicators. 并且,各个用户判断路由器综合性能的优劣时所侧重的指标也因人而异,例如用户I侧重于路由器的吞吐量,如果路由器的吞吐量性能较好,则认为路由器的综合性能较好,而用户2侧重于路由器的时延,如果路由器的时延性能较好,则认为路由器的综合性能较好,但是无论是吞吐量还是时延,都是影响路由器综合性能的因素,只侧重于某一个指标所得到的测试结果也不能准确的反映出路由器的综合性能。 Then, each user is determined when the merits of the overall performance of the router focused indicators also vary, for example, a user I focused on the throughput of the routers, the router if better throughput performance, it is considered better overall performance of the router, 2 and user focused on the delay of the router, the router's performance is better if the delay is considered better overall performance of the router, but either throughput or latency, factors affecting the performance of the router are only focusing on a particular an indicator of the obtained test results do not accurately reflect the overall performance of the router. 因此,现有技术中对网络设备测试的准确性较低。 Thus, the lower the accuracy of the prior art network device testing.

发明内容 SUMMARY

[0006] 本发明实施例提供一种测试方法及装置,用以解决现有技术中对网络设备测试的准确性低的问题。 Embodiment [0006] The present invention provides a testing method and apparatus to solve the prior art low accuracy network device testing problem.

[0007] 本发明实施例提供的一种测试方法,包括: [0007] A test method according to an embodiment of the present invention, comprising:

[0008] 对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值; [0008] For each metric, the network device performs multiple tests to obtain a plurality of corresponding test values ​​of each indicator;

[0009] 根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性程度值,以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集; [0009] The plurality of test values ​​for each corresponding index obtained to determine the relative value between the degree of importance of each indicator, the relative degree of importance between each indicator element is configured to determine the value of the fuzzy evaluation matrix, the fuzzy evaluation matrix configuration is determined set of weights;

[0010] 针对每个指标,根据得到的该指标对应的多个测试值,以及设定的每个性能等级对应的指标测试值范围,确定测试的该指标在每个性能等级上的百分比,以测试的该指标在每个性能等级上的百分比为元素构造评判子集;其中,确定测试的该指标在每个性能等级上的百分比,具体包括:针对每个性能等级,确定得到的该指标对应的多个测试值中,属于该性能等级对应的指标测试值范围内的测试值的个数,将确定的个数与得到的该指标对应的多个测试值的总数的比值,确定为测试的该指标在该性能等级上的百分比;以测试的该指标在每个性能等级上的百分比为元素构造评判子集,具体包括:根据该指标在每个性能等级上的百分比,构造评判子集Yi = (yn,ii2, yi3...yip),其中,Yi为第i个指标的评判子集,P为性能等级的个数,对于Yi中的任意元素yiq,yiq为该 [0010] For each metric, in accordance with a plurality of test values ​​corresponding to the obtained index, and an index range for each test value of the performance level corresponding to the set, to determine the percentage of testing on each performance level indicator to the percentage of test indicator performance level on each element of a subset is configured judgment; index wherein the percentage is determined on each test of the performance level, it comprises: for each of the performance level is determined to obtain the index corresponding to the plurality of test values, which belong to a value within the test range of the test indicator performance level corresponding to the ratio of the total number of the plurality of test values ​​with the determined index number obtained corresponding to the determined test the indicator on the percentage level of performance; percentage indicator to the test performance level on each element is configured to judge subset comprises: percentage of the indicator in accordance with the level of each performance, configuration subset judgment Yi = (yn, ii2, yi3 ... yip), wherein, Yi judgment subset of the i-th index, P is the number of performance levels, for any element in yiq Yi, for YIQ i个指标在第q个性能等级上的百分比; I The percentage of indicators on the individual energy levels of q;

[0011] 根据分别针对每个指标确定的评判子集,确定评判矩阵,将确定的权重集与评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0011] The evaluation of subsets being determined for each metric evaluation matrix is ​​determined, the weight and the determined weight set evaluation matrix multiplied by the performance evaluation of the set of network devices, the performance evaluation of the element corresponding to the maximum concentration performance is rated as the test results obtained.

[0012] 本发明实施例提供的一种测试方法,包括: [0012] A test method according to an embodiment of the present invention, comprising:

[0013] 对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值; [0013] Each secondary indicators for multiple network test equipment, to obtain a plurality of test values ​​for each corresponding secondary indexes;

[0014] 针对所述网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要程度值,以确定的各二级指标之间的相对重要程度值为元素构造该一级指标对应的二级模糊评判矩阵,根据构造的二级模糊评判矩阵,确定该一级指标对应的二级权重集;[0015] 针对该一级指标下的每个二级指标,根据得到的该二级指标对应的多个测试值,以及设定的每个性能等级对应的二级指标测试值范围,确定测试的该二级指标在每个性能等级上的百分比,以测试的该二级指标在每个性能等级上的百分比为元素构造该二级指标对应的二级评判子集; [0014] metrics for each one of the network device, in accordance with a plurality of test values ​​of the two indexes in an index corresponding to the obtained, to determine the relative index between the two at the level indicators importance values ​​to relative importance between the two indexes to determine the elements of the value of the index corresponding to a configuration of two fuzzy evaluation matrix, according to two fuzzy evaluation matrix configuration, determines that an index corresponding to two level weight set; [0015] for each of the two indexes in an index, in accordance with a plurality of test values ​​of the two corresponding to the obtained index, and two indicator test performance level for each range of values ​​corresponding to a set , determine the percentage of the two indicator test on each performance level to the percentage of the two indicator test performance level on each of the two construction elements two evaluation indexes corresponding to the subset;

[0016] 根据分别针对该一级指标下的各二级指标构造的二级评判子集,确定该一级指标对应的二级评判矩阵,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,得到该一级指标对应的一级评判集; [0016] Evaluation for each subset of two secondary indicators each configured in one of the index, an index corresponding to the determined evaluation matrix according to two, the two re-set a weight corresponding to the index a stage two evaluation indexes corresponding to matrix multiplication, to obtain a set of evaluation index corresponding to the one;

[0017] 根据分别针对所述网络设备的每个一级指标确定的一级评判集,确定一级评判矩阵; [0017] The evaluation set separately determined for each one index of said one network device, determines an evaluation matrix;

[0018] 确定所述网络设备的每个一级指标之间的相对重要程度值,以每个一级指标之间的相对重要程度值为元素构造一级模糊评判矩阵,根据构造的一级模糊评判矩阵确定一级权重集,将确定的一级权重集与一级评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0018] determining the relative importance of each of an index value between the network device, fuzzy evaluation matrix to relative importance between an index value of each element of a structure, according to the configuration of a fuzzy an evaluation matrix set of weights is determined, the determined weight set a right and a matrix multiplication evaluation, performance evaluation set to give the network device, the performance evaluation of the concentration of elements corresponding to the maximum performance level is determined to be obtained Test Results.

[0019] 本发明实施例提供的一种测试装置,包括: [0019] A test apparatus according to an embodiment of the present invention, comprising:

[0020] 测试模块,用于对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值; [0020] The testing module, each indicator for multiple network test equipment, to obtain a plurality of corresponding test values ​​of each indicator;

[0021] 权重集确定模块,用于根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性程度值,以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集; The relative degree of importance of each metric value between [0021] the weight set determining module, a plurality of test values ​​for each indicator according to the obtained corresponding value to determine the relative importance degree between each of indicators to determine the fuzzy evaluation matrix element constructed according to the determined weights fuzzy evaluation matrix configuration weight set;

[0022] 评判子集确定模块,用于针对每个指标,根据得到的该指标对应的多个测试值,以及设定的每个性能等级对应的指标测试值范围,确定测试的该指标在每个性能等级上的百分比,以测试的该指标在每个性能等级上的百分比为元素构造评判子集;其中,确定测试的该指标在每个性能等级上的百分比,具体包括:针对每个性能等级,确定得到的该指标对应的多个测试值中,属于该性能等级对应的指标测试值范围内的测试值的个数,将确定的个数与得到的该指标对应的多个测试值的总数的比值,确定为测试的该指标在该性能等级上的百分比;以测试的该指标在每个性能等级上的百分比为元素构造评判子集,具体包括:根据该指标在每个性能等级上的百分比,构造评判子集Yi = (yn»Yi2^yi3-..yip),其中,Yi为第i个指标的评判子集,P为性能等级的个数,对于Yi [0022] Evaluation subset determination module configured for each indicator, the indicator in accordance with a plurality of corresponding test values ​​obtained, and the performance index for each test value range corresponding to a set level, the indicator is determined in each test the percentage of the energy level of the individual, in order to test the indicator on the percentage of each configuration element performance level for a subset of evaluation; wherein the index is determined on each test performance level percentage, comprises: for each performance level, a plurality of test values ​​of the index corresponding to the determined obtained, which belong to the test value corresponding to the level of the performance index of the test range, the plurality of test values ​​with the determined index number corresponding to the obtained the ratio of the total number, as determined in a test on the percentage of the level of performance metrics; as a percentage of the index test performance level in each element is configured to judge subset comprises: at each level based on the performance index percentage, configured to judge subset Yi = (yn »Yi2 ^ yi3 - .. yip), where, Yi is the i-th subset evaluation index, P is the number of performance levels for Yi 的任意元素yiq,yiq为该第i个指标在第q个性能等级上的百分比; Any element yiq, yiq percentages for the i-th index in the q-th individual energy levels;

[0023] 测试结果确定模块,用于根据分别针对每个指标确定的评判子集,确定评判矩阵,将确定的权重集与评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0023] The test result determination means for determining based on the judging of subsets for each index, evaluation matrix is ​​determined, the determined set of weights and evaluation matrix multiplied by the set of the performance evaluation of the network device, the performance Evaluation of the concentration of elements corresponding to the maximum level of performance test results obtained for the determination.

[0024] 本发明实施例提供的一种测试装置,包括: [0024] A test apparatus according to an embodiment of the present invention, comprising:

[0025] 测试模块,用于对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值; [0025] The testing module, two indicators for each network device performs multiple tests to obtain a plurality of test values ​​for each corresponding secondary indexes;

[0026] 二级权重集确定模块,用于针对所述网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要程度值,以确定的各二级指标之间的相对重要程度值为元素构造该一级指标对应的二级模糊评判矩阵,根据构造的二级模糊评判矩阵,确定该一级指标对应的二级权重集; [0026] The two sets of weights determining module for each of the metrics for a network device, a corresponding plurality of test values ​​based on each of the two indexes in an index which is obtained, an indicator is determined in this the relative importance between the respective values ​​of the relative importance between the respective two indicators to determine the secondary structure elements of the value of the index corresponding to index a secondary fuzzy evaluation matrix, according to a configuration of two fuzzy evaluation matrix , it is determined that the two weights corresponding to a weight set index;

[0027] 二级评判子集确定模块,用于针对该一级指标下的每个二级指标,根据得到的该二级指标对应的对个测试值,以及设定的每个性能等级对应的二级指标测试值范围,确定测试的该二级指标在每个性能等级上的百分比,以测试的该二级指标在每个性能等级上的百分比为元素构造该二级指标对应的二级评判子集; [0027] Evaluation two subset determination module, configured to, for each index in the two-level indicators, based on test values, and each of the performance level is set corresponding to the two index corresponding to the obtained two indicator test range, determine the percentage of the two indicator test on each level of performance, the percentage of the secondary indicators to test the performance level on each of two construction elements evaluation index corresponding to the two Subset;

[0028] 一级评判集确定模块,用于根据分别针对该一级指标下的各二级指标构造的二级评判子集,确定该一级指标对应的二级评判矩阵,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,得到该一级指标对应的一级评判集; [0028] Evaluation of a set determination module, for respectively two for two evaluation index subset for each configuration at the level indicators to determine an index corresponding to the two evaluation matrix according to the level indicators two set of weights corresponding to the two evaluation indexes corresponding to a matrix multiplication, to give an evaluation of the set corresponding to an index;

[0029] 一级评判矩阵确定模块,用于根据分别针对所述网络设备的每个一级指标确定的一级评判集,确定一级评判矩阵; [0029] matrix determining a judgment means for judgment according to each set of level indicators determined for each of said one network device, determines an evaluation matrix;

[0030] 测试结果确定模块,用于确定所述网络设备的每个一级指标之间的相对重要程度值,以每个一级指标之间的相对重要程度值为元素构造一级模糊评判矩阵,根据构造的一级模糊评判矩阵确定一级权重集,将确定的一级权重集与一级评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0030] The test result of the determination means for the relative importance of each of an index value is determined between the network device, to the relative importance of each of an index value between a fuzzy evaluation matrix elements configured the configuration of a fuzzy evaluation matrix to determine a set of weights, the weight is multiplied by a set of weights determined with an evaluation matrix, to obtain performance evaluation set of the network device, the performance evaluation of the element corresponding to the maximum concentration performance is rated as the test results obtained.

[0031] 本发明实施例提供一种测试方法及装置,该方法对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值,根据每个指标之间的相对重要性程度值构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集,并确定测试的每个指标在每个性能等级上的百分比,据此确定评判矩阵,将确定的权重集与评判矩阵相乘,得到该网络设备的性能评判集,并将其中最大的元素对应的性能等级确定为得到的测试结果。 Example embodiments provide a method and apparatus for testing [0031] the present invention, the method for each indicator test multiple network devices to obtain a plurality of corresponding test values ​​of each indicator, the relative importance of each metric according to the degree of the fuzzy evaluation matrix configuration value, determined according to weight the fuzzy set evaluation matrix structure weight, and the percentage of each indicator is determined on each test performance level, whereby evaluation matrix is ​​determined, multiplied by the determined set of weights and evaluation matrix, performance evaluation set to give the network device, and wherein the elements corresponding to the maximum level of performance test results obtained for the determination. 通过上述方法,测试装置对每个指标都进行了测试,并综合了各个指标对网络设备的综合性能的影响,因此可以对网络设备的综合性能进行准确的测试。 By the above method, the test device for each indicator were tested, and the combination of overall performance impact indicators each network device, it can accurately test the overall performance of the network device.

附图说明 BRIEF DESCRIPTION

[0032] 图1为本发明实施例提供的测试过程; [0032] FIG 1 Test procedure according to an embodiment of the present invention;

[0033] 图2为本发明实施例提供的另一种测试过程; [0033] FIG 2 Another test procedure according to an embodiment of the present invention;

[0034] 图3为本发明实施例提供的与图1所示的测试方法对应的测试装置结构示意图; [0034] Fig 3 a schematic view of a test device configuration provided corresponding to the test method of the first embodiment of the present invention, FIG;

[0035] 图4为本发明实施例提供的与图2所示的另一种测试方法对应的测试装置结构示意图。 [0035] Fig 4 a schematic view of a test apparatus corresponding to the structure provided with another embodiment of the test method shown in FIG. 2 embodiment of the present invention. 具体实施方式 Detailed ways

[0036] 由于网络设备的各个指标或多或少都会影响网络设备的综合性能,因此只根据对某一个或某几个指标的测试值,并通过人为的判断网络设备综合性能的优劣,所得到的测试结果必然是不准确的,根据这样的测试结果选择使用的网络设备,会导致选择的网络设备在实际应用中表现出来的综合性能并不尽如人意,使网络质量下降。 [0036] Since the respective indices are hindering the overall performance of the network device the network device, and therefore only based on one or a few test value index, and overall performance is determined by artificially merits network device, the the resulting test result will be inaccurate, according to test results so choose to use the network equipment, will result in the selection of network equipment manifested in practical applications overall performance is not satisfactory, the network quality. 本发明实施例提供一种测试方法,该方法测试装置对网络设备的每个指标进行多次测试,并综合各个指标对网络设备综合性能的影响,以此得出测试结果,可以对网络设备的综合性能进行准确的测试。 Embodiments provide a testing method of the present invention, the method of testing apparatus for each of the multiple network devices index test, and synthesize various indicators affect the overall performance of the network device, in order to obtain test results, the network equipment can overall performance for accurate testing.

[0037] 下面结合说明书附图,对本发明实施例进行详细描述。 [0037] the following description in conjunction with the accompanying drawings, embodiments of the present invention will be described in detail.

[0038] 图1为本发明实施例提供的测试过程,具体包括以下步骤: [0038] FIG 1 Test procedure according to an embodiment of the present invention includes the following steps:

[0039] SlOl:对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值。 [0039] SlOl: each indicator test multiple network devices to obtain a plurality of index values ​​for each corresponding test.

[0040] 在本发明实施例中,测试装置对网络设备的每个指标都进行了多次测试,具体测试的次数可以根据需要进行设定,例如100次。 [0040] In an embodiment of the present invention, the network testing apparatus for each indicator device have carried out several tests, the number of specific tests may be set as needed, for example, 100 times.

[0041] 例如,网络设备的每个指标如表1所示: [0041] For example, each network device indicator is shown in Table 1:

[0042] [0042]

Figure CN102447593BD00141

[0043]表1 [0043] TABLE 1

[0044] 如表1所示,该网络设备的指标包括:指标1、指标2、指标3、指标4,这四个指标都会影响该网络设备的综合指标,因此,测试装置对这四个指标分别进行多次测试,例如分别对这四个指标进行100次测试,得到这四个指标分别对应的100个测试值。 The indicators of the network device [0044] As shown in Table 1 comprising: Index 1, Index 2, Index 3, 4 indicators, four indicators comprehensive index will affect the network device, and therefore, these four indicators test apparatus several tests were carried out, for example, each of these four indicators 100 tests, which give four indicators respectively corresponding to the 100 test values.

[0045] S102:根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性 [0045] S102: The plurality of test values ​​for each corresponding index obtained, determining the relative importance of each indicator

程度值。 The degree of value.

[0046] 在本发明实施例中,测试装置可以根据得到的测试值,确定每个指标之间的相对重要性程度值,相对重要性程度值表示:将该网络设备的每个指标进行两两比较,其中一个指标影响网络设备综合性能的程度相比于另一个指标影响网络设备综合性能的程度的比较值。 [0046] In an embodiment of the present invention, the test device may be a value obtained according to the test, to determine the relative value between the degree of importance of each indicator, the relative value represents the degree of importance: the network device of each indicator pairwise comparison, the more value compared to impact the overall performance of network equipment to another indicator of the degree of an indicator of the overall performance impact of network devices degree.

[0047] S103:以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集。 [0047] S103: In the relative degree of importance between each metric determined configuration fuzzy evaluation matrix elements of the value, the fuzzy set evaluation matrix configured to determine the weight of the weight.

[0048] 测试装置确定了每个指标之间的相对重要性程度值后,则以每个相对重要性程度值为元素构造模糊评判矩阵,并据此确定权重集,该权重集中的每个元素表示:综合考虑了每个指标之间的相对重要性程度值后,相应的每个指标影响网络设备综合性能的程度值。 Each element of the [0048] test device to determine the relative value between the degree of importance of each indicator, the relative degree of importance of each of places is constructed by fuzzy evaluation matrix elements, and accordingly determine the set of weights, the weights concentrated He said: after considering the relative degree of importance between each index value, the corresponding value of the degree of influence of each indicator of the overall performance of network equipment.

[0049] S104:针对每个指标,根据得到的该指标对应的多个测试值,以及设定的每个性能等级对应的指标测试值范围,确定测试的该指标在每个性能等级上的百分比,以测试的该指标在每个性能等级上的百分比为元素构造评判子集。 [0049] S104:, a plurality of test values ​​corresponding to the obtained index, and an index range of values ​​for each test performance level corresponding to a set, a test to determine the percentage indicator for each indicator on each level of performance to the percentage of target test performance level in each element is configured to judge subset.

[0050] 在本发明实施例中,可以预先设定各个性能等级,例如设定为四个性能等级,分别为:第一等级、第二等级、第三等级、第四等级。 [0050] In an embodiment of the present invention, each pre-set level of performance, for example, set to four performance levels, namely: a first level, second level, third level, fourth level. 并可以针对每个指标,设定每个性能等级对应的指标测试值范围。 And for each index can be set for each range index test performance level corresponding. 测试装置则可以针对每个指标,确定该指标对应的测试值中,分别落在每个性能等级对应的指标测试值范围内的测试值的数量,进而确定该指标在每个性能等级上的百分比,并据此构造评判子集。 Testing means for each indicator may be, determines the value of the index corresponding to the test, the falling number of test values ​​are within specification values ​​for each test performance level corresponding to the range, then determine the percentage of the indicator in each level of performance and accordingly construct judge subset.

[0051] S105:根据分别针对每个指标确定的评判子集,确定评判矩阵。 [0051] S105: The evaluation of subsets being determined for each indicator, determining evaluation matrix.

[0052] 在本发明实施例中,分别确定了每个指标的评判子集后,则以各评判子集中的元素确定评判矩阵,该评判矩阵中包含了各评判子集中的所有元素。 After [0052] In an embodiment of the present invention, were determined for each evaluation index subset, places each element of the subset judgment evaluation matrix is ​​determined, the evaluation matrix contains all the elements of each subset judgment.

[0053] S106:将确定的权重集与评判矩阵相乘,得到该网络设备的性能评判集,将该性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0053] S106: The determination of weight sets and weight matrix multiplication evaluation, performance evaluation set to give the network device, and the performance evaluation of the concentration of elements corresponding to the maximum level of performance test results obtained for the determination.

[0054] 由于权重集中的每个元素表示每个指标影响网络设备综合性能的程度值,评判矩阵中的每个元素表示某个指标在某个性能等级上的百分比,因此将权重集与评判矩阵相乘,所得到的性能评判集中的每个元素即为网络设备的综合性能在每个性能等级上的百分t匕,因此将性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0054] Since each element represents the weight concentration of each element of the degree of influence of each index value of the overall performance of the network device, evaluation matrix represents a certain percentage of an indicator on the performance level, and therefore the evaluation matrix and weight set multiplying the resulting concentration of each element performance evaluation is the overall performance of the network device at each percent t dagger performance level, and therefore judge the performance of the element corresponding to the maximum concentration level of performance test results obtained for the determination .

[0055] 在上述过程中,测试装置对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值,根据每个指标之间的相对重要性程度值构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集,并确定测试的每个指标在每个性能等级上的百分比,据此确定评判矩阵,将确定的权重集与评判矩阵相乘,得到该网络设备的性能评判集,并将其中最大的元素对应的性能等级确定为得到的测试结果。 [0055] In the above process, the test apparatus for each indicator test multiple network devices to obtain a plurality of indicators each corresponding test values, the fuzzy evaluation matrix configuration based on the relative value between the degree of importance of each indicator, the percentage of each indicator is determined fuzzy evaluation matrix configuration set of weights, and tested to determine performance at each level, whereby evaluation matrix is ​​determined, multiplied by the determined set of weights and evaluation matrix, to obtain the performance evaluation of the network device set, and wherein the elements corresponding to the maximum performance level is determined as the test results obtained. 通过上述方法,测试装置对每个指标都进行了测试,并综合了各个指标对网络设备的综合性能的影响,因此可以对网络设备的综合性能进行准确的测试。 By the above method, the test device for each indicator were tested, and the combination of overall performance impact indicators each network device, it can accurately test the overall performance of the network device.

[0056] 在图1所示的步骤SlOl中,测试装置根据设定次数,对网络设备的每个指标进行设定次数的测试,得到每个指标对应的多个测试值。 [0056] In step SlOl of FIG. 1, the test apparatus according to the set number of times, each indicator for testing network equipment set number, to obtain a plurality of index values ​​for each corresponding test. 其中,针对每个指标,得到的该指标对应的多个测试值的数量与该设定次数相等。 Wherein the number of the plurality of test values ​​for each indicator, the indicator obtained corresponding to the set count is equal. 并且,为了进一步提高测试的准确性,该设定次数可以设定的较大,较佳的,该设定次数不小于网络设备的指标的个数。 And the number, in order to further improve the accuracy of the test, the set number of times can be set larger, preferably, not less than the set number of network devices indicators.

[0057] 在图1所示的步骤S102中,测试装置根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性程度值的方法为:以得到的每个指标对应的多个测试值为元素构造第一测试矩阵: [0057] In the method step shown in FIG. 1 S102, the test device of the plurality of corresponding test values ​​for each metric obtained to determine the relative value between the degree of importance for each indicator: In each metric obtained a plurality of test elements of the value corresponding to the first test the matrix configuration:

Figure CN102447593BD00151

[0059] 其中,η为所述设定次数,m为所述网络设备的指标的个数,Xij为对m个指标中的第j个指标进行的第i次测试所得到的测试值;根据构造的第一测试矩阵确定每个指标之间的相对重要性程度值。 [0059] wherein, η is the set number, m is the index of the number of network devices, Xij is the value of the test to test the m i in the j-th index index of the obtained; according the first test matrix configuration determine the relative value between the degree of importance of each indicator.

[0060] 在根据上述第一测试矩阵确定每个指标之间的相对重要性程度值时,首先要对第一测试矩阵进行归一化处理,得到第二测试矩阵: [0060] When the degree of relative importance between each first value of the index is determined according to a test matrix, the first test must first normalized matrix, to obtain a second test matrix:

Figure CN102447593BD00161

[0062] 其中,为第一测试矩阵中的Xu的归一化值。 [0062] wherein, in a first test matrix Xu normalized value. 具体归一化的方法可以为:对于第一测试矩阵中的元素Xu,确定元素Xu所在的一列中所有元素中的最大元素max以及最小元素min ;由于Xij为对第i个指标的测试值,因此考虑第i个指标与网络设备综合性能 DETAILED normalization method may be as follows: For element Xu first test matrix to determine the maximum element max is an element Xu where all the elements and the smallest element min; since Xij is the test value for the i-th index, So consider the i-th overall performance indicators and network equipment

的关系,如果该第i个指标的测试值越大,综合性能越好,则采用公式 The relationship, if the test value of the i-th index greater the better the overall performance, the use of formula

Figure CN102447593BD00162

确定Xi,j的归一化值Xij,如果该第i个指标的测试值越小,综合性能越好,则采用公式m£i?c— X.- OK Xi, j normalized value Xij, if the test value of the i-th index of the smaller, the better the overall performance, the use of formula m £ i? C- X.-

Figure CN102447593BD00163

确定Xij的归一化值X'ij。 Determining the normalized value of X'ij Xij. 当然,还可以采用其他方法对第一测试矩 Of course, other methods may also be employed for the first test moment

阵进行归一化处理,例如直接将Xij与max的比值作为归一化值X'ij Array for normalization processing, for example, directly with the ratio max Xij as the normalization value X'ij

[0063] 得到归一化处理后的第二测试矩阵后,将第二测试矩阵的转置矩阵与该第二测试矩阵相乘,得到第三测试矩阵: [0063] After obtaining a second test matrix normalized after the treatment, the transposed matrix of the second test matrix and the second matrix multiplication test, the third test matrix is ​​obtained:

[0064] [0064]

Figure CN102447593BD00164

[0065] 根据为每个指标设定的基准值,将每个基准值相应的添加到所述第三测试矩阵中,得到第四测试矩阵: [0065] The index set for each reference value, the reference value corresponding to each is added to the third test matrix, to obtain a fourth test matrix:

[0066] [0066]

Figure CN102447593BD00165

[0067] 其中,β m+1;i为m个指标中第i个指标的基准值,为每个指标设定的基准值可以根据需要进行设定。 [0067] where, β m + 1; i is a reference index value of m i-th index, for each index set reference value may be set as needed.

[0068] 对所述第四测试矩阵进行归一化处理,得到第五测试矩阵: [0068] the fourth test matrix is ​​normalized to give a fifth test matrix:

[0069] [0069]

Figure CN102447593BD00166

[0070] 其中,a' ij为第四测试矩阵中的Ciij的归一化值,β'm+1,i为第四测试矩阵中的β'm+1,i的归一化值。 [0070] wherein, a 'ij is the normalized value of the fourth test matrix of Ciij, β'm + 1, i is the fourth test matrix β'm + 1, i of the normalized value. 对第四测试矩阵进行归一化处理的方法具体可以为:对于第四测试矩阵中的元素,确定元素CiuK在的一列中所有元素中的最大元素max以及最小元素.t *.~~ The method of the fourth test matrix normalization process may specifically be: For the fourth test matrix element, determining a maximum element max CiuK element in all elements of one and the smallest element .t * ~~

min,并米用公式 min, and using the formula m

Figure CN102447593BD00171

厂确定Ctij的归一化值α 1 iJO当然,也可以米用公式 Plant Ctij determined normalized value α 1 iJO course, may be formulated m

Figure CN102447593BD00172

或者其他方法进行归一化处理。 Or other methods for normalization.

[0071] 根据所述第五测试矩阵,釆用下述公式确定各指标与各指标的基准值的距离: [0071] According to the fifth test matrix, Bian determine the index of each index by the following equation from a reference value:

Figure CN102447593BD00173

[0073] 其中,i = 1,2,3...m,di;ffl+1为m个指标中的第i个指标与该第i个指标的基准值的距离; [0073] where, i = 1,2,3 ... m, di; ffl + 1 m for the indicators in the i-th index and the distance from the i-th reference index value;

[0074] 针对任意两个指标,根据下述公式确定该两个指标之间的相对重要性程度值: [0074] for any two indicators, to determine the relative importance degree index value between the two according to the following equation:

Figure CN102447593BD00174

[0076] 其中,为m个指标中的第i个指标相对于第j个指标的相对重要性程度值。 [0076] where m is the number of indices i-th index value of the relative degree of importance of the j-th index. 由此可以看出,对于网络设备的任意两个指标,第i个指标相对于第j个指标的相对重要性程度值ru,与第j个指标相对于第i个指标的相对重要性程度值的和值为1,并且第i个指标相对于其自身(第i个指标)的相对重要性程度值为0.5。 It can be seen that for any two network devices indicator, the relative degree of importance with respect to the j-th index in the i-th index value Ru, and the j-th index value of the relative degree of importance of the i-th index and a value of 1, and the i-th index relative to its (the i-th index) their relative importance degree is 0.5.

[0077] 另外,考虑到实际应用场景的差异,网络设备的每个指标之间的相对重要性程度值还可以进行人为的调整,以使得最后的测试结果是针对与某个特定应用场景的测试结果,调整的相对重要性程度值仍然需要满足与的和值为1,且rii的值为0.5。 [0077] Further, considering the difference between the actual application scenarios, the relative value between the degree of importance of each network device further indicators may be artificially adjusted so that the final test results are tested for a specific application scenario and As a result, the degree of adjustment of the relative importance value and still need to satisfy the value of 1, and rii is 0.5.

[0078] 采用上述方法确定了每个指标之间的相对重要性程度值之后,在图1所示的步骤S103中,以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵的方法具体为,构造下述矩阵作为模糊评判矩阵: [0078] After determining the degree of relative importance between the value of each indicator using the above method, at step S103 shown in FIG. 1, the relative degree of importance between each of the elements of the value of the indicator to determine the configuration Fuzzy Evaluation matrix is ​​specifically constructed as a fuzzy evaluation matrix of the following matrix:

Figure CN102447593BD00175

[0080] 其中,rij为第i个指标相对于第j个指标的相对重要性程度值,与I^ji的和值为1,且b的值为0.5,也即,模糊评判矩阵R中的对角线元素的值均为0.5,任意两个相对于对角线对称的元素的和值为I。 [0080] wherein, Rij is the i-th index value of the relative degree of importance of the j-th index, and I ^ ji and a value of 1, and b value of 0.5, i.e., the fuzzy evaluation matrix R values ​​are 0.5 diagonal elements, with respect to any two of the diagonal elements of the symmetric value and I.

[0081] 根据上述模糊评判矩阵R确定权重集的方法具体为:对构造的模糊评判矩阵R进行模糊一致化处理,得到模糊一致矩阵M: [0081] The method for determining said fuzzy evaluation matrix R is specifically set of weights: the structure of the fuzzy evaluation matrix R fuzzy unification process, to obtain fuzzy consistent matrix M:

Figure CN102447593BD00176

[0083] 其中,对于模糊一致矩阵M中的任意元素ξ ij; [0083] wherein, for the fuzzy consistent matrix M is any element ξ ij;

Figure CN102447593BD00181

rik为模糊评判矩阵中第i行第k列的元素,rJk为模糊评判矩阵中第j行第k列的元素; rik fuzzy evaluation element in row i and column k of the matrix, rJk fuzzy evaluation element j th row of k-column matrix;

[0084] 根据模糊一致矩阵M,采用下述公式确定每个指标对应的权重: [0084] The fuzzy consistent matrix M, using the following formula to determine the weight of each index corresponding weight:

Figure CN102447593BD00182

[0086] 其中,ω i为第i个指标对应的权重,即为综合考虑了每个指标之间的相对重要性程度值后,该第i个指标影响网络设备综合性能的程度值。 After [0086] where, [omega] i is the i-th index corresponding weight, i.e. the degree of importance of considering the relative values ​​between each of the indexes, the i-th index value of the degree of influence the overall performance of the network device.

[0087] 以确定的每个指标对应的权重为元素构成权重集:ω = (ω。ω2, ω 3...ωπ)。 [0087] The weight of each index to determine a corresponding weight of the weight element constituting the weight set: ω = (ω.ω2, ω 3 ... ωπ).

[0088] 在图1所示的步骤S104中,确定测试的该指标在每个性能等级上的百分比的方法具体为:针对每个性能等级,确定得到的该指标对应的多个测试值中,属于该性能等级对应的指标测试值范围内的测试值的个数,将确定的个数与得到的该指标对应的多个测试值的总数的比值,确定为测试的该指标在该性能等级上的百分比。 [0088] In step S104 shown in FIG. 1, the percentage of the test method for determining the performance level indicator on each specifically: a plurality of test values ​​of the index corresponding to each of the performance level for determining obtained, the number of test values ​​within a range of target test belonging to the performance level corresponding to the ratio of the total number of the determined plurality of test values ​​corresponding to the obtained index, the index is determined on the test performance level percentage.

[0089] 其中,该指标针对每个性能等级对应的指标测试值范围可以根据测试日志中记录的之前对该指标每次进行测试得到的各测试值确定。 [0089] wherein the index index for each test performance level corresponding to the range of values ​​each may test each test values ​​obtained prior to the test is determined according to the index recorded in the log. 例如,将测试记录中记录的之前对该指标每次进行测试所得到的各测试值由大到小排序,将排在前25%的测试值对应的测试值范围确定为该指标针对第一等级对应的指标测试值范围,将排在25%~50%的测试值对应的测试值范围确定为该指标针对第二等级对应的指标测试值范围,以此类推,确定每个性能等级对应的指标测试值范围。 For example, the previously recorded test recording each time the index value for each test of the obtained test descending order, the top 25% of the test value corresponding to the range of test values ​​for the metrics for determining a first level test index value range corresponding to the row in the 25% to 50% of the test value corresponding to the range of test values ​​for the metrics for determining the test range indicator corresponding to the second level, and so on, to determine the performance level corresponding to each index test range.

[0090] 另外,该指标在每个性能等级上的百分比还可以采用其他方式确定,例如统计每个用户对该指标的评价,针对每个性能等级,确定将该指标评价为该性能等级的个数,将确定的个数与统计的评价的总数的比值,确定为该指标在该性能等级上的百分比。 [0090] In addition, the percentage of the indicator in each of the performance level may also be determined in other ways, such as statistical evaluation of the indexes for each user, for each of the performance level, the evaluation index for determining a level of performance number, the ratio of the total number of statistical evaluation of the determined indicators for determining the percentage level of performance.

[0091] 以测试的该指标在每个性能等级上的百分比为元素构造评判子集的方法具体为:根据该指标在每个性能等级上的百分比,构造下述评判子集: [0091] In this test the percentage indicator on each performance level for the sub-element constructor evaluation methodology specifically: percentage on each level of performance, the following configuration based on the evaluation index subset:

[0092] Yi = (yn, ji2, yi3...yip), [0092] Yi = (yn, ji2, yi3 ... yip),

[0093] 其中,Yi为第i个指标的评判子集,ρ为性能等级的个数,对于Yi中的任意元素yiq,yiq为该第i个指标在第q个性能等级上的百分比。 [0093] where, Yi is the i-th subset evaluation index, [rho] is the number of performance levels, for any element in yiq Yi, YIQ percentages for the i-th index in the individual energy levels of the q.

[0094] 采用上述方法针对每个指标都确定了相应的评判子集后,在图1所示的步骤S105中,确定的评判矩阵具体为:根据分别针对每个指标确定的评判子集,确定下述评判矩阵: [0094] After the evaluation are to determine the appropriate subset for each index using the above method, in the step shown in FIG. 1 S105, the evaluation matrix is ​​determined specifically as follows: The judgment of subsets being determined for each indicator, determined The following evaluation matrix:

Figure CN102447593BD00183

[0096] 其中,对于Y中的任意元素yiq,yiq为第i个指标在第q个性能等级上的百分比。 [0096] wherein, for any element in yiq Y, YIQ as a percentage of the i-th index in the individual energy levels of the q. [0097] 在图1所示的步骤S106中,测试装置则可以将权重集ω = (ω1, ω2, ω 3...ωn) [0097] In step S106 shown in FIG. 1, the test device may be weight set ω = (ω1, ω2, ω 3 ... ωn)

与评判矩阵 And evaluation matrix

Figure CN102447593BD00191

相乘,得到性能评判集: Multiplied by the performance evaluation set:

[0099] 其中,性能评判集S中的每个元素即为网络设备的综合性能在每个性能等级上的百分比,假设性能评判集S中的元素Sq为最大的元素,则将该元素Sq对应的第q个性能等级确定为得到的测试结果。 [0099] wherein, performance evaluation set S is the percentage of each element in the overall performance of the network device at each performance level, Sq is assumed that the elements of the performance evaluation set S is the largest element, then the element corresponding to Sq q the first character can be rated as the test results obtained.

[0100] 下面以上述表1为例说明测试过程。 [0100] In the above Table 1 below illustrate an example test procedure.

[0101] 该网络设备的指标包括:指标1、指标2、指标3、指标4,测试装直对这四个指标分别进行100次测试,得到这四个指标分别对应的100个测试值。 [0101] Indicators of the network device comprising: index 1, index 2, index 3, 4 indicators, four of these straight test apparatus 100 tests index, respectively, to obtain the four indicators 100 corresponding to each test values. 则构造第一测试矩阵 The first configuration of the test matrix

Figure CN102447593BD00192

,对第一测试矩阵进行归一化处理得到第二测试矩 , The first test matrix is ​​normalized to obtain the second test moment

阵,并将第二测试矩阵的转置矩阵与第二测试矩阵相乘,得到第三测试矩阵,假设针对这四个指标设定的基准值分别为β51,β52,β53,β54,则添加这四个基准值,并进行归一化 Array, and the second transposed matrix multiplication of the second test matrix test matrix, to obtain a third test matrix, it is assumed for these four indicators are set reference value β51, β52, β53, β54, which is added four reference values, and normalized

处理得到的第五测试矩阵为 Treating the resulting matrix for the fifth test

Figure CN102447593BD00193

,假设根据第五测试矩阵得到 , Is assumed to give a test matrix according to a fifth

的模糊评判矩阵为 The fuzzy evaluation matrix

Figure CN102447593BD00194

,则经过模糊一致化处理得到的模糊一 , The same processing to be blurred blur a

致矩阵为 Matrix induced

Figure CN102447593BD00195

进而得到的权重集为ω = Further weight is set to give a weight ω =

(0.2667,0.3083,0.2417,0.1833),假设设定的性能等级分别为:第一等级、第二等级、第三等级、第四等级,根据针对这四个指标确定的在每个性能等级上的百分比确定的评判矩阵为 (0.2667,0.3083,0.2417,0.1833), that the set performance levels are as follows: a first level, second level, third level, fourth level, in accordance with the determined indicators for the four on each performance level the percentage is determined by evaluation matrix

Figure CN102447593BD00201

其中,评判矩阵中的第I行第I列的值0.1为指标 Wherein the value of column I of the I row matrix evaluation index is 0.1

I在第一等级上的百分比,第I行第2列的值0.1为指标I在第二等级上的百分比,第2行第I列的值O为指标2在第一等级上的百分比,以此类推,则将权重集与评判矩阵相乘得到的性能评判集为S= ω XY = (0.0875,0.155,0.66,0.0975),可见性能评判集中的元素0.66为最大元素,其对应的性能等级为第三等级,因此将该第三等级作为对网络设备进行测试所得到的测试结果,该测试结果即为测试的该网络设备的综合性能。 The percentage on the first level I, the value of the second row I and column index I 0.1 as a percentage on the second level, the value at row I, column 2 O 2 percentage of the indexes as the first level, to such push performance evaluation set of weights will be obtained by multiplying the evaluation matrix is ​​S = ω XY = (0.0875,0.155,0.66,0.0975), the performance evaluation can be seen as the maximum concentration of elements in element 0.66, which corresponds to the level of performance third level, and therefore the level of the third network device as the obtained result of the test, the overall performance of the network device is the test results of this test.

[0102] 在本发明实施例中,考虑到实际应用中还可能针对网络设备的每个指标,将该指标再细分为更细粒度的下级指标。 [0102] In an embodiment of the present invention, considering the actual application is also possible metrics for each network device, the indicator subdivided into more fine-grained lower index. 例如,路由器的指标包括吞吐量、时延等,吞吐量和时延为路由器的一级指标。 For example, a router indicators include throughput, and delay, throughput and delay for the router level indicator. 吞吐量又分为:设备吞吐量和端口吞吐量,设备吞吐量和端口吞吐量共同影响着路由器的吞吐量,因此这两个指标是吞吐量下的二级指标。 Throughput is divided into: equipment throughput and port throughput, device throughput and port throughput combined effect of the throughput of the router, so these two indicators is the second indicator in throughput. 时延又分为:最大时延、最小时延、平均时延,这三个指标共同影响着路由器的时延,是时延下的二级指标。 Delay is divided into: maximum delay, minimum delay, average delay, the combined effect of these three indicators delay router is secondary indicators under the delay. 因此,基于与图1所示的测试方法同样的思路,本发明还提供了另一种测试方法,用于对细分为一级和二级指标的网络设备的综合性能进行测试。 Thus, based on the same test method shown in FIG. 1 and FIG idea, the present invention also provides another method for testing for the overall performance of the network device is subdivided into primary and secondary indicators tested.

[0103] 图2为本发明实施例提供的另一种测试过程,具体包括以下步骤: [0103] FIG 2 Another test procedure according to an embodiment of the present invention includes the following steps:

[0104] S201:对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值。 [0104] S201: secondary indicators for each network device performs multiple tests to obtain a corresponding plurality of test values ​​for each two indexes.

[0105] 在本发明实施例中,测试装置对每个二级指标进行测试的次数也可以根据需要进行设定,例如100次。 [0105] In an embodiment of the present invention, the number of test apparatus for testing each of two indicators may be set as required, for example 100 times.

[0106] 其中,该网络设备的一级指标和二级指标的层级关系如表2所示。 Hierarchy [0106] wherein an indicator of the network device and secondary indicators as shown in Table 2.

[0107] [0107]

Figure CN102447593BD00202

[0108] [0108]

Figure CN102447593BD00211

[0109]表 2 [0109] TABLE 2

[0110] 如表2所示,该网络设备的一级指标包括:指标1、指标2、指标3、指标4,每个一级指标下又包括若干个二级指标。 [0110] As shown in the table, an index of the network device 2 includes: index 1, index 2, index 3, 4 indicators, the indicators turn each comprise a plurality of secondary indicators. 以指标I为例,指标I下包括的各二级指标分别为:指标 In Case I indicators, each of the two indicators include indicators I are: Index

I1、指标I2、指标I3、指标I4,这四个二级指标共同影响着指标1,其他一级指标和二级指标的关系也与上述关系基本相同。 I1, index I2, index I3, I4 indicators, four indicators of the combined effect of the two index 1, two level indicators and relationships with other indicators are also substantially the same as the above relationship. 因此,测试装置对每个二级指标分别进行多次测试,例如分别对每个二级指标进行100次测试,得到每个二级指标分别对应的100个测试值。 Thus, the test device for each test several secondary indicators separately, for example, 100 were tested for each two indexes, each two indicators were obtained 100 corresponding test values.

[0111] S202:针对该网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要性程度值。 [0111] S202: an indicator for each of the network device, in accordance with a plurality of test values ​​of the two indices of an index obtained at the corresponding index is determined between the two at the level indicators The relative importance of the degree of value.

[0112] 在本发明实施例中,测试装置可以针对每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要性程度值。 [0112] In an embodiment of the present invention, the test device may be an index for each, in accordance with a plurality of test values ​​of the two indexes in an index obtained corresponding to each of the two to determine an index in the the relative degree of importance between the index value. 该一级指标下的各二级指标之间的相对重要性程度值表示:将该一级指标下的每个二级指标进行两两比较,其中一个二级指标影响该一级指标的程度相比于另一个及指标影响该一级指标的程度的比较值。 Degree of relative importance between the two index values ​​in an indicator that indicates: the index in each of two level indicators pairwise comparison, wherein a degree of influence of the two indexes relative to an index comparative value ratio on the level indicators and targets to another level.

[0113] S203:以确定的各二级指标之间的相对重要性程度值为元素构造该一级指标对应的二级模糊评判矩阵,根据构造的二级模糊评判矩阵,确定该一级指标对应的二级权重集。 [0113] S203: In the relative importance degree among the two indexes to determine the elements of the value of the index corresponding to a configuration of two fuzzy evaluation matrix, according to a configuration of two fuzzy evaluation matrix, an index corresponding to the determined the two sets of weights.

[0114] 测试装置确定的该一级指标下的各二级指标之间的相对重要性程度值后,则以各二级指标之间的每个相对重要性程度值为元素构造模糊评判矩阵,并据此确定该一级指标对应的二级权重集,该一级指标对应的二级权重集中的每个元素表示:考虑了该一级指标下的各二级指标之间的相对重要性程度值后,该一级指标下的每个二级指标影响该一级指标的程度值。 After [0114] the degree of relative importance between the two index values ​​in an indicator that the test apparatus determines, places relative degree of importance of each index value between the two elements of the fuzzy evaluation matrix configuration, and accordingly determines the index corresponding to a set of two weights, two weights each element of the index corresponding to a set of weight represented by: considering the relative importance degree among the two indexes in an index after the value of each of the two indexes in an index value of the degree of influence of the level indicators.

[0115] S204:针对该一级指标下的每个二级指标,根据得到的该二级指标对应的多个测试值,以及设定的每个性能等级对应的二级指标测试值范围,确定测试的该二级指标在每个性能等级上的百分比,以测试的该二级指标在每个性能等级上的百分比为元素构造该二级指标对应的二级评判子集。 [0115] S204: For each index in the two-level indicators, in accordance with a plurality of test values ​​of the two corresponding to the obtained index, and two indicator test performance level for each range of values ​​corresponding to a set, determined the percentage of secondary indexes on each test performance level to the percentage of the two indicator test performance level on each of the two construction elements corresponding to two evaluation index subset.

[0116] 在本发明实施例中,可以预先限定各个性能等级,例如设定为四个性能等级,分别为:第一等级、第二等级、第三等级、第四等级。 [0116] In an embodiment of the present invention, each predefined performance levels, for example, set to four performance levels, namely: a first level, second level, third level, fourth level. 并可以针对该一级指标下的每个二级指标,设定每个性能等级对应的指标测试值范围。 And for each two indexes can be at the level indicators, test indicators set range corresponding to each of the performance level. 测试装置则可以针对该一级指标下的每个二级指标,确定该二级指标对应的测试值中,分别落在每个性能等级对应的指标测试值范围内的测试值的数量,进而确定该二级指标在每个性能等级上的百分比,并据此构造该二级指标对应的二级评判子集。 Testing means for each two indexes can be under the level indicators to determine the value of the two test indicator corresponding to, respectively, the number of test values ​​falling within a range of values ​​for each test indicator corresponding to the level of performance, and to determine the on each of the two performance level percentage indicator, and configured accordingly two evaluation index subset corresponding to two.

[0117] S205:根据分别针对该一级指标下的各二级指标构造的二级评判子集,确定该一级指标对应的二级评判矩阵,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,得到该一级指标对应的一级评判集。 [0117] S205: Evaluation for each subset of two secondary indicators each configured in one of the index, an index corresponding to the determined evaluation matrix according to two, the two set of weights corresponding to the level indicators an index corresponding to the two evaluation matrix multiplication, to give an evaluation set corresponding to the level indicators.

[0118] 在本发明实施例中,分别确定了该一级指标下的每个二级指标对应的二级评判子集后,则以各二级评判子集中的元素确定该一级指标对应的二级评判矩阵,该二级评判矩阵中包含了该一级指标下的每个二级指标对应的二级评判子集中的元素。 After [0118] In an embodiment of the present invention, each of the two indexes were determined two evaluation corresponding to the subset in which an indicator, two places each element of the subset judgment determines that an index corresponding to two evaluation matrix, the matrix contains two evaluation indexes for each two in this indicator corresponding to a subset of secondary judgment elements.

[0119] 由于该一级指标对应的二级权重集中的每个元素表示该一级指标下的每个二级指标影响该一级指标的程度值,该一级指标对应的二级评判矩阵中的每个元素表示该一级指标下的某个二级指标在某个性能等级上的百分比,因此将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,所得到的该一级指标对应的一级评判集中的每个元素即为:该一级指标在每个性能等级上的百分比。 [0119] Since each element of the two weights corresponding to a weight set index represents the index of each two-level indicators of the degree of influence of the value of an index, an index corresponding to the two evaluation matrix each element represents a percentage of the two indices at a certain performance level in the indicator, and thus the weight of a weight set index corresponding to two multiplied by the indicator corresponding to a two evaluation matrix the obtained an index corresponding to each element of a set is the evaluation: a percentage of the index at each level of performance.

[0120] S206:根据分别针对该网络设备的每个一级指标确定的一级评判集,确定一级评判矩阵。 [0120] S206: The evaluation set separately determined for each one of the indicators of a network device, a determination evaluation matrix.

[0121] 在本发明实施例中,测试装置分别针对每个一级指标都确定了该一级指标对应的一级评判集后,则以各一级评判集中的元素确定一级评判矩阵,该一级评判矩阵中包含了各一级评判集中的所有元素。 After [0121] In an embodiment of the present invention, a test apparatus one for each of the indicators to determine an evaluation index corresponding to one of the set, are concentrated in a set of elements is determined to judge an evaluation matrix, which an evaluation matrix contains all the elements of each one judge set.

[0122] S207:确定该网络设备的每个一级指标之间的相对重要性程度值,以每个一级指标之间的相对重要性程度值为元素构造一级模糊评判矩阵,根据构造的一级模糊评判矩阵 [0122] S207: the value of the degree of relative importance between each of the network device determines an index of the degree of importance relative to an index value between each element of a matrix is ​​configured fuzzy evaluation, constructed in accordance with a fuzzy evaluation matrix

确定一级权重集。 Determining a set of weights.

[0123] 在本发明实施例中,测试装置还要确定网络设备的每个一级指标之间的相对重要性程度值,每个一级指标之间的相对重要性程度值表示:将该网络设备的每个一级指标进行两两比较,其中一个一级指标影响网络设备综合性能的程度相比于另一个指标影响网络设备综合性能的程度的比较值。 [0123] In an embodiment of the present invention, the test device but also to determine the relative value between the degree of importance of each network device of an indicator, the degree of relative importance between each value represents an index: The Network each device-level indicators of pairwise comparison, a degree-level indicators which affect the overall performance of network equipment compared to the degree of influence of another indicator of the overall performance of network equipment comparison value.

[0124] 确定了每个一级指标之间的相对重要性程度值后,则以每个相对重要性程度值为元素构造一级模糊评判矩阵,并据此确定一级权重集,该一级权重集中的每个元素表示:综合考虑了每个一级指标之间的相对重要性程度值后,相应的每个一级指标影响网络设备综合性能的程度值。 After [0124] determining the relative value between the degree of importance of each of an index, the relative degree of importance of each of the fuzzy places element is configured an evaluation matrix, and accordingly determines a set of weights, the one each element of the weight concentrated, said: considering the relative value between the degree of importance of each level indicators, the corresponding index for each one degree affect the value of the overall performance of network equipment.

[0125] S208:将确定的一级权重集与一级评判矩阵相乘,得到该网络设备的性能评判集,将该性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0125] S208: The determination of a set of weights with an evaluation matrix multiplied by the performance evaluation set of the network device, and the performance evaluation of the concentration of elements corresponding to the maximum level of performance test results obtained for the determination.

[0126] 由于一级权重集中的每个元素表示每个一级指标影响网络设备综合性能的程度值,一级评判矩阵中的每个元素表示某个一级指标在某个性能等级上的百分比,因此将一级权重集与一级评判矩阵相乘,所得到的性能评判集中的每个元素即为网络设备的综合性能在每个性能等级上的百分比,因此将性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0126] Since each element of a set of weights to represent each an index value of the degree of influence the overall performance of network equipment, judge each element of the matrix represents the percentage of a certain level indicators on the performance level of a , therefore a set of weights with an evaluation matrix multiplication, the resulting performance evaluation of each element is the percentage of the overall performance of the centralized network device at each level of performance, and therefore judge the performance of the element corresponding to the maximum concentration performance is rated as the test results obtained.

[0127] 在上述过程中,测试装置对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值,针对每个一级指标,根据该一级指标下的每个二级指标之间的相对重要性程度值构造该一级指标对应的二级模糊评判矩阵,根据构造的二级模糊评判矩阵确定该一级指标对应的二级权重集,并确定测试的该一级指标下的每个二级指标在每个性能等级上的百分比,据此确定该一级指标对应的二级评判矩阵,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相等,得到该一级指标对应的一级评判集,根据分别针对每个一级指标得到的一级评判集确定一级评判矩阵,根据每个一级指标之间的相对重要性程度值构造一级模糊评判矩阵,据此确定一级权重集,将一级权重集与一级评判矩阵相乘得到该网络设备的性能评判集,将其中最 [0127] In the above process, the test apparatus for each two indexes multiple test network device to obtain a plurality of test values ​​for each corresponding secondary indicators, one for each index, an index based on the lower the relative importance degree between each of the two index values ​​of the index corresponding to a configuration of two fuzzy evaluation matrix, an index corresponding to the determined set of weights according to two secondary fuzzy evaluation matrix configuration, and determine a test the percentage of each of the two indicators under the level indicators in each performance level, whereby a determination of the indicators corresponding to two evaluation matrix, the weight set two weights level indicators corresponding to the one index is equal to the corresponding two evaluation matrix, to obtain a set of evaluation index corresponding to the one of an evaluation matrix are determined for each of a set of evaluation index obtained according to a, in accordance with an index between each degree of relative importance value a fuzzy evaluation matrix structure, whereby to determine a set of weights, the one set of weights and a performance evaluation matrix obtained by multiplying the evaluation set of the network device, in which the most 大的元素对应的性能等级确定为得到的测试结果。 Large performance level determination element corresponding to test results obtained. 通过上述方法,测试装置对每个二级指标都进行了测试,并综合了各二级指标对其所属的一级指标的影响,以及各一级指标对网络设备的综合性能的影响,逐级向上确定网络设备的综合性能,因此可以对网络设备的综合性能进行准确的测试。 By the above method, the test apparatus for each two indexes were tested, and the combination of the respective two Effect index belongs to its level indicators, as well as a comprehensive indicator of the performance of network devices, step by step up determine the overall performance of network devices, so you can accurately test the integrated performance of network devices.

[0128] 当网络设备的指标细分为三级、四级或者更多级时,图2所示的方法仍然适用。 When [0128] the network device when the index is divided into three, four or more stages, the method shown in FIG. 2 still apply. 以三级为例,可以采用与图2类似的方法,对每个三级指标进行多次测试,并综合各三级指标对其所属的二级指标的影响、各二级指标对其所属的一级指标的影响、各一级指标对网络设备的综合性能的影响,逐级向上确定网络设备的综合性能,这里就不再一一赘述。 In an example three, may be employed for each of the three indicators of multiple test and a method similar to FIG. 2, and the combined effects of each of the three indicators of its two index belongs, belongs to its respective secondary indicators the impact level indicators, the impact of each level indicators of the overall performance of network equipment, network equipment up level by level to determine the overall performance, it is not going to repeat them here.

[0129] 在图2所示的步骤S201中,测试装置根据设定次数,对网络设备的每个二级指标进行设定次数的测试,得到每个二级指标对应的多个测试值。 [0129] In step S201 shown in FIG. 2, the test apparatus according to the set number of times, two indicators for each network device to test a set number, to obtain a plurality of test values ​​for each corresponding secondary indicators. 其中,针对每个二级指标,得到的该二级指标对应的多个测试值的数量与该设定次数相等。 Wherein the number of the plurality of test values ​​for each of the two indexes, the two resulting index corresponding to the set count is equal. 并且,为了进一步提高测试的准确性,该设定次数可以设定的较大,较佳的,该设定次数不小于网络设备的二级指标的个数。 Preferably the number and, in order to further improve the accuracy of the test, the set number of times can be set larger, not less than the set number of secondary indicators of network devices.

[0130] 在图2所示的步骤S202中,测试装置针对网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要程度值的方法具体为:针对网络设备的每个一级指标,以得到的该一级指标下的每个二级指标对应的多个测试值为元素构造第一测试矩阵: [0130] In the step shown in FIG. 2 S202, the test device for each network device level indicators, a plurality of test values ​​of the two indices of an index obtained at the corresponding level indicators to determine the the method of relative importance between the two index values ​​in particular: a metrics for each network device, a plurality of test elements of the value of each index in the two-level indicators corresponding to afford construction of the first test matrix:

Figure CN102447593BD00231

[0132] 其中,n为所述设定次数,m为该一级指标下的二级指标的个数,xgu为对第g个一级指标下的第j个二级指标进行的第i次测试所得到的测试值;根据构造的第一测试矩阵确定该一级指标下的各二级指标之间的相对重要性程度值。 [0132] wherein, for said n-set number, m the number of secondary indexes for the next level indicators, the i-th to j-th xgu secondary indicators of the indicators of a g th value test obtained; degree of relative importance between the two index values ​​in an indicator that the first test the matrix configuration is determined according to.

[0133] 在根据上述第一测试矩阵确定该一级指标下的各二级指标之间的相对重要性程度值时,首先要对第一测试矩阵进行归一化处理,得到第二测试矩阵::'¾! ΛΈη …xSlm ^ [0133] When the first test matrix to determine the relative value between the degree of importance of each of the two indexes in an index in accordance with the first test first normalized matrix, to obtain a second test matrix: : '¾ ΛΈη ... xSlm ^!

[0134] [0134]

Figure CN102447593BD00232

[0135] 其中,xg, u为第一测试矩阵中的Xgij的归一化值。 [0135] where, xg, u is the first test matrix Xgij normalized value. 具体归一化的方法可以为:对于第一测试矩阵中的元素Xgu,确定元素Xgu所在的一列中所有元素中的最大元素max以及最小元素min ;由于Xgij为对第g个一级指标下的第j个二级指标进行的第i次测试所得到的测试值,因此考虑该第j个二级指标与第g个一级指标的关系,如果该第j个 DETAILED normalization method may be as follows: For the first element Xgu test matrix, determining a maximum element in all elements max and a minimum element min Xgu element is located; Xgij due to the g-th to the next level indicators i-j-th test indicators for the two test values ​​obtained, so consider the relationship between the j-th index and the g-th two-level indicators, if the j-th

二级指标的测试值越大,第g个一级指标越好,则釆用公式 The larger the value of the two test index, the better the g level indicators, with the formula preclude

Figure CN102447593BD00233

确定Xgij的归一化值Xg' ij,如果该第j个指标的测试值越小,第g个一级指标越好,则采用公式 Xgij determined normalized value Xg 'ij, if the j-th index test value, the better the g level indicators, then using the equation

Figure CN102447593BD00241

确定Xgij的归一化值xg' ij。 Determining the normalized value Xgij xg 'ij. 当然,还可以采用其他方法对第一测试 Of course, other methods may also be employed for the first test

矩阵进行归一化处理,例如直接将Xgij与max的比值作为归一化值Xg' iJO Matrix normalization processing, for example, directly with the ratio Xgij max as the normalization value Xg 'iJO

[0136] 得到归一化处理后的第二测试矩阵后,将第二测试矩阵的转置矩阵与第二测试矩阵相乘,得到第三测试矩阵: [0136] After obtaining a second test matrix normalized after the treatment, the transposed matrix of the second test matrix and a second matrix multiplication the test, the third test matrix is ​​obtained:

Figure CN102447593BD00242

[0138] 根据为该一级指标下的每个二级指标设定的基准值,将每个基准值相应的添加到所述第三测试矩阵中,得到第四测试矩阵: [0138] According to each of the two indexes in index for a set reference value, the reference value corresponding to each is added to the third test matrix, to obtain a fourth test matrix:

Figure CN102447593BD00243

[0140] 其中,β gffl+1;i为该第g个一级指标下的第i个二级指标的基准值,为每个二级指标设定的基准值可以根据需要进行设定。 [0140] where, β gffl + 1; i-th reference value under the two indexes i for the g level indicators, may be set for each of the two reference index setting value as needed.

[0141] 对所述第四测试矩阵进行归一化处理,得到第五测试矩阵: [0141] the fourth test matrix is ​​normalized to give a fifth test matrix:

Figure CN102447593BD00244

[0143] 其中,ag' u为第四测试矩阵中的agij的归一化值,i3g'm+li为第四测试矩阵中的β δω+Μ的归一化值。 [0143] wherein, ag 'u fourth test matrix agij normalized value, i3g'm + li fourth test matrix β δω + Μ normalized value. 对第四测试矩阵进行归一化处理的方法具体可以为:对于第四测试矩阵中的元素agu,确定元素Cigu所在的一列中所有元素中的最大元素max以及最小元 The method of the fourth test matrix normalization process may specifically be: For the fourth test agu matrix element, determining a list of elements where the largest element Cigu all elements max and a minimum element

素min,并釆用公式 Su min, and preclude the use of the formula

Figure CN102447593BD00245

确定agij的归一化值ag' iJO当然,也可以釆 Determining the normalized value agij ag 'iJO course, and also can be

用公式 Using the formula

Figure CN102447593BD00246

,或者其他方法进行归一化处理。 Or other methods normalized.

[0144] 根据所述第五测试矩阵,采用下述公式确定该一级指标下的各二级指标与各二级指标的基准值的距离: [0144] According to the fifth test matrix, using the following formula to determine the distance of each two indexes in the index with a reference index for each two values:

[0145] [0145]

Figure CN102447593BD00247

[0146] 其中,i = 1,2,3...m,dgi,m+1为该第g个一级指标下的第i个二级指标与该第i个二级指标的基准值的距离; [0146] where, i = 1,2,3 ... m, dgi, m + 1 for i-g-th secondary indicators in an index value with reference to the i-th secondary indicators of distance;

[0147] 针对该一级指标下的任意两个二级指标,根据下述公式确定该两个二级指标之间的相对重要性程度值: [0147] for any two indices at the two-level indicators to determine the relative value between the degree of importance of the two secondary indicators according to the formula:

[0148] [0148]

Figure CN102447593BD00251

[0149] 其中,rgu为该第g个一级指标下的第i个二级指标相对于该第g个一级指标下的第j个二级指标的相对重要性程度值。 [0149] wherein, rgu for the i-th index in the two-level indicators of g relative degree of importance of the j-th value in the two indicators g of level indicators. 由此可以看出,对于该第g个一级指标下的任意两个二级指标,第i个二级指标相对于第j个二级指标的相对重要性程度值rgu,与第j个二级指标相对于第i个二级指标的相对重要性程度值rgu的和值为1,并且第i个二级指标相对于其自身(第i个二级指标)的相对重要性程度值为0.5。 It can be seen, for any two of the two indexes in the g level indicators, two i-th index j with respect to the secondary indicators of the relative degree of importance RGU value, and the j-th two level indicators relative degree of importance of the i-th index value rgu two and a value of 1, and the i-th secondary indicators relative to its own (the i-th secondary indicators) relative importance degree is 0.5 .

[0150] 另外,考虑到实际应用场景的差异,该一级指标下的每个二级指标之间的相对重要性程度值还可以进行人为的调整,以使得最后的测试结果是针对与某个特定应用场景的测试结果,调整的相对重要性程度值仍然需要满足rgu与1^4的和值为1,且rgii的值为0.5。 [0150] Further, considering the difference between the actual application scenarios, the relative degree of importance of each of the two index values ​​between the lower-level indicators may also be artificially adjusted so that the final result for the test of a the test results of a particular application scenario, the degree of adjustment of the relative importance value and still need to meet rgu 1 ^ 4 and a value of 1, and a value of 0.5 rgii.

[0151] 采用上述方法确定了该一级指标下的每个二级指标之间的相对重要性程度值之后,在图2所示的步骤S203中,以确定的各二级指标之间的相对重要性程度值为元素构造该一级指标对应的二级模糊评判矩阵的方法具体为,构造下述矩阵为该一级指标对应的二级模糊评判矩阵: The relative index between the two [0151] After determining the value of the degree of relative importance between each of the two indexes in an index using the above method, at step S203 shown in FIG. 2, to determine the method two fuzzy evaluation matrix elements of the value of the importance degree of an index corresponding to the configuration specifically, the following matrix configuration for a secondary fuzzy evaluation matrix corresponding to the index:

Figure CN102447593BD00252

[0153] 其中,rgij为该第g个一级指标下的第i个二级指标相对于该第g个一级指标下的第j个二级指标的相对重要性程度值,rgiJ与rgji的和值为1,且rgii的值为0.5,也即,该第g个一级指标对应的二级模糊评判矩阵Rg中的对角线元素的值均为0.5,任意两个相对于对角线对称的元素的和值为I。 [0153] wherein, rgij for the i-th index in the two-level indicators of g relative degree of importance of the j-th value in the two indicators g of level indicators, rgiJ with the rgji and a value of 1, and rgii value of 0.5, i.e., the value of the diagonal elements of the g-th fuzzy evaluation matrix Rg two level indicators corresponding to 0.5 are, with respect to any two diagonals and symmetric element values ​​I.

[0154] 根据上述二级模糊评判矩阵Rg,确定该一级指标对应的二级权重集的方法具体为:对构造的二级模糊评判矩阵进行模糊一致化处理,得到二级模糊一致矩阵Mg: [0154] The above-described method two fuzzy evaluation matrix Rg, determining the two weights corresponding to a weight set index is specifically: of two fuzzy judgment matrix configuration is consistent with the fuzzy processing, to obtain two fuzzy consistent matrix Mg:

Figure CN102447593BD00253

[0156] 其中,对于二级模糊一致矩阵Mg中的任意元素ξ giJ, [0156] wherein, for any two elements of fuzzy consistent matrix ξ and Mg gij,

Figure CN102447593BD00254

rgik为二级模糊评判矩阵中第i行第k列的元素,I^k为二级模糊评判矩阵中第j行第k列的元素; rgik fuzzy evaluation of two elements in row i and column k of the matrix, I ^ k fuzzy evaluation of two elements j th row of k-column matrix;

[0157] 根据二级模糊一致矩阵Mg,采用下述公式确定该一级指标下的每个二级指标对应的权重: [0157] The two fuzzy consistent matrix Mg, using the following formula to determine each of the two indexes in the index corresponds to a weight:

Figure CN102447593BD00261

[0159] 其中,Cogi为该第g个一级指标下的第i个二级指标对应的权重; [0159] wherein, Cogi weight for the i-th index in the two-level indicators of g corresponding to the weight;

[0160] 以确定的该一级指标下的每个二级指标对应的权重为元素构成该一级指标对应的二级权重集:ω g = ((Og1, Qg2, Qg3...ω gm)。 [0160] Each secondary index in the index to determine a corresponding weight of an element constituting the two indicators corresponding to a set of weights: ω g = ((Og1, Qg2, Qg3 ... ω gm) .

[0161] 在图2所示的步骤S204中,确定测试的该二级指标在每个性能等级上的百分比的方法具体为:针对每个性能等级,确定得到的该二级指标对应的多个测试值中,属于该性能等级对应的指标测试值范围内的测试值的个数,将确定的个数与得到的该二级指标对应的多个测试值的总数的比值,确定为测试的该二级指标在该性能等级上的百分比。 [0161] In the step shown in FIG. 2 S204, the percentage of the two method indicator is determined on each test performance level is specifically: the two corresponding performance metrics for each level, a plurality of determination obtained test value, the test values ​​which belong to the range of values ​​of the index test performance level corresponding to the ratio of the total number of the plurality of test values ​​of the determined number of the corresponding secondary indicators obtained, it is determined that the test the percentage of secondary indexes on the level of performance. 其中,该二级指标针对每个性能等级对应的指标测试值范围可以根据测试日志中记录的之前对该二级指标每次进行测试得到的各测试值确定,这里就不再一一赘述。 Wherein the two metrics for each range index value test performance level corresponding to each index may be performed for each of the two test values ​​determined in accordance with the test result recorded in the log prior to testing, not further described here.

[0162] 另外,该二级指标在每个性能等级上的百分比还可以采用其他方式确定,例如统计每个用户对该二级指标的评价,针对每个性能等级,确定将该二级指标评价为该性能等级的个数,将确定的个数与统计的评价的总数的比值,确定为该二级指标在该性能等级上的百分比。 [0162] In addition, the percentage of secondary indexes on each performance level may also be determined in other ways, such as statistical evaluation of every user of the secondary indicators for each performance level, the two evaluation index determined the number for the level of performance, the ratio of the total number of evaluation will determine the number of statistics, determined as the percentage of secondary indexes on the level of performance.

[0163] 以测试的该二级指标在每个性能等级上的百分比为元素构造该二级指标对应的二级评判子集的方法具体为:根据该二级指标在每个性能等级上的百分比,构造下述二级评判子集: [0163] In this test the percentage of secondary indicators on each level of performance evaluation method of a subset of two elements of the two corresponding indicator configured specifically as follows: The percentage of each performance level on the basis of the two indicator , the following two judgment subset configured:

[0164] Ygi = (ygn, ygi2, ygi3...ygip), [0164] Ygi = (ygn, ygi2, ygi3 ... ygip),

[0165] 其中,Ygi为第g个一级指标下的第i个二级指标的二级评判子集,P为性能等级的个数,对于Ygi中的任意元素ygiq,yglq为该g个一级指标下的第i个二级指标在第q个性能等级上的百分比。 [0165] wherein, for the two evaluation YGI i-th subset of the secondary indicators of level indicators g, P is the number of performance levels, for any element in ygiq Ygi, yglq for one-g the percentage of the i-th index in the two-level metrics on individual energy levels of the q.

[0166] 采用上述方法针对该一级指标下的每个二级指标都确定了相应的二级评判子集后,在图2所示的步骤S205中,确定的该一级指标对应的二级评判矩阵具体为:根据分别针对该一级指标下的各二级指标确定的二级评判子集,确定下述二级评判矩阵: After the [0166] method described above for each index in the two indicators to determine a corresponding secondary judgment subset indicators corresponding to the step shown in FIG. 2 S205, it is determined that one of the two specifically evaluation matrix: Evaluation the two subsets are determined for each of the two indexes in an index, determined by the following two evaluation matrix:

Figure CN102447593BD00262

[0168] 其中,对于Yg中的任意元素ygiq,ygiq为该第g个一级指标下的第i个二级指标在第q个性能等级上的百分比。 [0168] wherein, for any element in ygiq Yg, ygiq percentages for the i-th index in the two-level indicators of g in the first energy level q personality. [0169] 确定了上述该一级指标对应的二级评判矩阵后,测试装置则将该一级指标对应的二级权重集Og= (COg1, Qg2, COg3...(Ogm)与该一级指标对应的二级评判矩阵 After the [0169] determined that an index corresponding to the above-mentioned two evaluation matrix, then the two test apparatus Og a set of weights corresponding to the index = (COg1, Qg2, COg3 ... (Ogm) with the one evaluation matrix index corresponding to two

Figure CN102447593BD00271

相乘,得到该一级指标对应的一级评判集为: Multiplying the one set to obtain an evaluation index corresponding to:

Figure CN102447593BD00272

其中,Sg为第g个一级指标对应的一级评判集,Wg= (Og1, «g2» Wg3...Wgm)为确定的该第g个一级指标对应的二级权重集,Sg1, Sg2, Sg3...Sgp为该第g个一级指标分别对应P个性能等级的评判权重。 Wherein, Sg is the g-th an evaluation index corresponding to a current collector, Wg = (Og1, «g2» Wg3 ... Wgm) set two weights is the weight of the g-th level indicators corresponding to the determined, Sg1, Sg2, Sg3 ... Sgp an index for the g-th P corresponding to individual can judge right level weight, respectively.

[0170] 测试装置针对每个一级指标都确定了相应的一级评判集后,确定下述矩阵作为一级评判矩阵: After [0170] the test apparatus are identified by a corresponding set of evaluation index, for each one, the following matrix is ​​determined as an evaluation matrix:

Figure CN102447593BD00273

[0172] 其中,S为确定的一级评判矩阵,h为所述网络设备的一级指标的个数,对于一级评判矩阵S中的任意元素sgq,Sgq为所述网络设备的h个一级指标中的第g个一级指标对应第q个性能等级的评判权重。 [0172] where, S is the determination of an evaluation matrix, h is an index number of the network device, for an evaluation matrix sgq any element of S, the network device SGQ is the h a level indicators of level indicators g of q corresponding to individual can judge right level weight.

[0173] 在图2所示的步骤S207中,测试装置还要确定网络设备的每个一级指标之间的相 [0173] with an index between each step shown in FIG. 2 S207, the test device but also to determine the network device

对重要性程度值,并据此构造一级模糊评判矩阵: The importance degree value, and accordingly construct a fuzzy evaluation matrix:

Figure CN102447593BD00274

[0175] 其中,h为网络设备的一级指标的个数,rij为网络设备的h个一级指标中的第i个一级指标相对于第j个一级指标的相对重要性程度值,与I^i的和值为1,且的值为0.5。 [0175] where, h is an index number of network devices, level indicators Rij is h network devices in an i-th index value of the relative degree of importance of the j-th level indicators, I ^ i and the value of 1, and a value of 0.5.

[0176] 并且,针对网络设备的任意两个一级指标,第i个一级指标相对于第j个一级指标的相对重要性程度值的确定方法可以为:确定第i个一级指标与该第i个一级指标下的各二级指标的运算关系,将每次测试的该第i个一级指标下的各二级指标的测试值按照确定的运算关系进行运算,得到第i个一级指标对应每次测试的测试值;采用上述方法确定了每个一级指标对应每次测试的测试值后,可以采用与确定某个一级指标下的各二级指标之间的相对重要性程度值类似的方法,根据每个一级指标对应每次测试的测试值,确定每个一级指标之间的相对重要性程度值,并基于每个一级指标之间的相对重要性程度值构造一级模糊评判矩阵。 Method for determining [0176] and any two metrics for a network device, the i-th level indicators with respect to a j-th index value relative importance degree may be: determining an i-th index and test value of each index in the two i-th index calculating an index of relationship of the two at the i th level indicators, each test computes the calculation according to the determined relationship to obtain the i an index value corresponding to each test test; determined after each corresponding to an index value for each test using the test method described above, may be employed between the two relative importance index and an index determination under one degree values ​​in a similar manner in accordance with an index corresponding to each of the test values ​​for each test, to determine the relative value between the degree of importance of each of an index, and based on the relative importance degree between each level indicators a matrix configuration fuzzy evaluation value.

[0177] 例如,一级指标Al下的各二级指标为:al和a2,该一级指标与各二级指标的运算关系为:Al=al+a2,则将第一次对al测试得到的测试值,与第一次对a2测试得到的测试值的和值,作为得到的一级指标Al对应第一次测试的测试值,以此类推,也即将每次对al测试得到的测试值,与每次对a2测试得到的测试值的和值,作为得到的一级指标Al对应每次测试的测试值。 [0177] For example, each of the two indexes in an index for the Al: al and a2, the relationship between an index and operation of each of two indicators: Al = al + a2, then get on the first test al the test value, the test value and the first value a2 obtained test as an index to obtain a corresponding test value Al of the first test, and so on, will soon give each test value of the test al with each test value and the test value a2 obtained as an index corresponding to Al obtained test values ​​for each test. 确定了每个一级指标对应每次测试的测试值后,根据每个一级指标对应每次测试的测试值,确定每个一级指标之间的相对重要性程度值,并构造一级模糊评判矩阵。 After determining the test values ​​for each test indicator corresponding to each one, according to an index corresponding to each of the test values ​​for each test, to determine the relative value between the degree of importance of each of an index, and a configuration Fuzzy evaluation matrix.

[0178] 还可以根据需要,人为设定每个一级指标之间的相对重要性程度值,并基于人为设定的每个一级指标之间的相对重要性程度值,构造一级模糊评判矩阵。 [0178] may also be required, the relative degree of importance of the artificially set values ​​between each of an index value based on the relative importance degree between each of a set of indicators artificial construct a fuzzy evaluation matrix.

[0179] 其中,无论采用哪种方法确定每个一级指标之间的相对重要性程度值,确定的相对重要性程度值均需要满足:第i个一级指标相对于第j个一级指标的相对重要性程度值,与第j个一级指标相对于第i个一级指标的相对重要性程度值的和值为1,且第i个一级指标相对于其自身(第i个一级指标)的相对重要性程度值为0.5。 [0179] where, no matter what method of determining the relative degree of importance of each of an index value between the relative importance of the extent required to meet all values ​​determined using: an i-th index with respect to the j-th level indicators degree of relative importance value, a j-th index with respect to an i-th index value and the relative importance degree is 1, and the i-th level indicators relative to its own (a i-th level indicator) the relative importance degree is 0.5.

[0180] 测试装置确定了一级模糊评判矩阵R后,确定一级权重集的方法具体为: After [0180] the testing device determines a fuzzy evaluation matrix R, a method of determining the set of weights is specifically:

[0181] 对构造的一级模糊评判矩阵进行模糊一致化处理,得到一级模糊一致矩阵M: [0181] The configuration of a fuzzy evaluation matrix fuzzy unification process, to give a fuzzy consistent matrix M:

Figure CN102447593BD00281

[0183] 其中,对于一级模糊一致矩阵M中的任意元素ξ ij? [0183] wherein a fuzzy consistent for any element of the matrix M ξ ij?

Figure CN102447593BD00282

rik为一级模糊评判矩阵中第i行第k列的元素,rjk为一级模糊评判矩阵中第j行第k列的元素; rik fuzzy evaluation matrix element in row i and column k a, rjk is the fuzzy evaluation matrix element in the j-th row of a k-th column;

[0184] 根据一级模糊一致矩阵M,釆用下述公式确定每个一级指标对应的权重: [0184] According to a fuzzy consistent matrix M, an index corresponding to preclude determined for each weight by the following equation:

Figure CN102447593BD00283

[0186] 其中,COi为该第i个一级指标对应的权重; [0186] wherein, for the right COi level indicators corresponding to the i-th weight;

[0187] 以确定的每个一级指标对应的权重为元素构成一级权重集:ω = (CO1, ω2,O 3...0 h) ο [0187] each of a weight index to determine a corresponding weight of an element constituting the weight set: ω = (CO1, ω2, O 3 ... 0 h) ο

[0188] 可见,根据一级模糊评判矩阵确定一级权重集的方法,与根据某个一级指标对应的二级模糊评判矩阵确定该一级指标对应的二级权重集的方法基本相同。 [0188] visible, an evaluation matrix method of determining a fuzzy set of weights in accordance with an index corresponding to a secondary fuzzy evaluation matrix is ​​determined according to substantially the same procedure two weights corresponding to a weight set index.

[0189] 测试装置确定了一级评判矩阵S以及一级权重集ω后,在图2所示的步骤S208中,测试装置则可以将一级权重集ω = (ωι,ω2,ω3...ω,)与一级评判矩阵 [0189] Evaluation of a test apparatus to determine the weight matrix S, and a rear set of weight [omega], in the step shown in FIG. 2 S208, the test device may be a weight set ω = ... (ωι, ω2, ω3 ω,) and an evaluation matrix

Figure CN102447593BD00291

相乘,得到性能评判集:T = ω XS = (t1; t2, Multiplied by the performance evaluation set: T = ω XS = (t1; t2,

t3...tp),其中,性能评判集τ中的每个元素即为网络设备的综合性能在每个性能等级上的百分比,假设性能评判集T中的元素t,为最大元素,则将该元素t,对应的第q个性能等级确定为得到的测试结果。 t3 ... tp), wherein the performance evaluation set τ is the percentage of each element in the overall performance of the network device at each level of performance, the performance evaluation set is assumed that the elements T t, is the largest element, then the element t, q corresponding to the first energy level determination individual test results obtained.

[0190] 下面以表2为例说明测试过程。 [0190] In Table 2 below an example test procedure.

[0191] 该网络设备的一级指标包括:指标1、指标2、指标3、指标4,每个一级指标下包括若干个二级指标,测试装置对每个二级指标分别进行100次测试。 [0191] an indicator of the network device comprising: indicator 1, the next 2, index 3, 4 indicators, each indicator comprises an index number of secondary indexes, each of two test apparatus 100 tests index, respectively, . 其中,指标I下的二级指标为:指标I1、指标I2、指标I3、指标14,则测试装置以得到的该指标I下的每个二级指标对应的多个测试值为元素构造的第一测试矩阵为 Wherein two of the index I to index: index I1, index I2, index I3, indicators 14, a first plurality of test elements of the value corresponding to the configuration of each of the two indexes in the index I is the test apparatus to afford a test matrix

Figure CN102447593BD00292

[0192] 对第一测试矩阵进行归一化处理得到第二测试矩阵,并将第二测试矩阵的转置矩阵与第二测试矩阵相乘,得到第三测试矩阵,假设针对指标I1、指标I2、指标I3、指标I4分别设定的基准值为β151,β152,β153,β I54,则添加这四个基准值,并进行归一化处理得到的 [0192] The first test matrix is ​​normalized to obtain the second test matrix, and a second transposed matrix multiplication of the second test matrix test matrix, to obtain a third test matrix, it is assumed for the index I1, I2 index , index I3, I4 indicators are set reference value β151, β152, β153, β I54, add these four reference value, and normalized to give the

第五测试矩阵为別 The fifth test matrix is ​​not

Figure CN102447593BD00293

[0193] 假设根据第五测试矩阵得到的指标I对应的二级模糊评判矩阵为 [0193] The hypothesis index I corresponding to the fifth test matrix obtained two fuzzy evaluation matrix

Figure CN102447593BD00294

,则经过模糊一致化处理得到的该指标I对应的二级模糊-致矩阵为 The two fuzzy index I corresponding to the same treatment obfuscated obtained - induced matrix

Figure CN102447593BD00301

进而翻的该指标m应的 Further turning of the indicator corresponding to m

二级权重集为ω I = (0.2667,0.3083,0.2417,0.1833)。 Two sets of weights ω I = (0.2667,0.3083,0.2417,0.1833).

[0194] 假设设定的性能等级分别为:第一等级、第二等级、第三等级、第四等级,根据针对指标I下的四个二级指标确定的在每个性能等级上的百分比确定的二级评判矩阵为 [0194] provided that the set performance levels are as follows: a first level, second level, third level, fourth level, determined in accordance with the determined two indicators for the four index I in percentage on each level of performance secondary evaluation matrix

Figure CN102447593BD00302

其中,二级评判矩阵中的第I行第I列的值0.1为.JL U.Z5 U.JL , Wherein the value of column I I two row matrix evaluation to 0.1 .JL U.Z5 U.JL,

指标11在第一等级上的百分比,第I行第2列的值0.1为指标I1在第二等级上的百分比,第2行第I列的值O为指标I2在第一等级上的百分比,以此类推,则将指标I对应的二级权重集与指标I对应的二级评判矩阵相乘得到指标I对应的一级评判集SI = ω IXYl =(0.0875,0.155,0.66,0.0975)。 The percentage indicator 11 on the first level, the value of the second row I, column 0.1 percentage index I1 is at the second level, the value at row I, column 2 O as a percentage of the index I2 in the first level, so, then the index I corresponding to the weight set with two weights corresponding to two judgment index I is obtained by multiplying a matrix index I corresponding to a set of evaluation SI = ω IXYl = (0.0875,0.155,0.66,0.0975).

[0195] 测试装置针对指标2、指标3、指标4都进行上述处理,得到相应的一级评判集,假设针对指标2得到的一级评判集为S2 = ω2ΧΥ2 = (0.5241,0.1717,0.163,0.1477),针对指标3得到的一级评判集为S3 = ω3ΧΥ3 = (0.1525,0.2475,0.1525,0.4475),针对指标4得到的一级评判集为S4 = ω4ΧΥ4 = (0.09,0.155,0.3,0.455),则确定的一级评判矩阵 [0195] 2 test apparatus for index, index 3, 4 indicators are above process, to give a corresponding set of evaluation, it is assumed for the index 2 is obtained an evaluation set S2 = ω2ΧΥ2 = (0.5241,0.1717,0.163,0.1477 ), for a set of evaluation index 3 is obtained S3 = ω3ΧΥ3 = (0.1525,0.2475,0.1525,0.4475), indexes for evaluation of a set of 4 is obtained S4 = ω4ΧΥ4 = (0.09,0.155,0.3,0.455), it is determined an evaluation matrix

即为 That is

Figure CN102447593BD00303

[0196] 测试装置确定指标1、指标2、指标3、指标4之间的相对重要性程度值,并构造一级模糊评判矩阵,假设根据构造的一级模糊评判矩阵确定的一级权重集ω = (0.2333, [0196] Testing apparatus 1 determines, 2 index, index 3, the degree of relative importance between the 4 index value index, and a fuzzy evaluation matrix configuration, assuming the weight ω according to a set of weights a fuzzy evaluation matrix configuration determined = (0.2333,

0.275,0.2333,0.2583),则将一级权重集ω与一级评判矩阵S相乘,得到的性能评判集为T= ω XS = (0.2234, 0.1812,0.3119,0.2853),可见性能评判集中的元素0.3119为最大元素,其对应的性能等级为第三等级,因此将该第三等级作为对网络设备进行测试所得到的测试结果,该测试结果即为测试的该网络设备的综合性能。 0.275,0.2333,0.2583), the set of weights [omega] and a weight matrix S is multiplied by an evaluation, performance evaluation set is obtained T = ω XS = (0.2234, 0.1812,0.3119,0.2853), set the visible performance evaluation element 0.3119 for the largest element, which corresponds to the third level performance level, and therefore the level of the third network device as the obtained result of the test, the overall performance of the network device is the test results of this test.

[0197] 另外,对于指标细分为三级、四级或者更多及时,也可以采用上述方法逐级向上确定网络设备的综合性能。 [0197] Further, the index is subdivided into three, four or more timely, said method determining step by step up overall performance of network devices may be employed.

[0198] 图3为本发明实施例提供的与图1所示的测试方法对应的测试装置结构示意图,具体包括: [0198] Fig 3 a schematic view of a test apparatus test method provided configuration shown in Fig. 1 corresponding to the embodiment of the present invention comprises:

[0199] 测试模块301,用于对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值; [0199] The test module 301, each indicator for multiple network test equipment, to obtain a plurality of corresponding test values ​​of each indicator;

[0200] 权重集确定模块302,用于根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性程度值,以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集; [0200] module 302 determines a set of weights, a plurality of test values ​​for each indicator corresponding to the obtained value to determine the relative degree of importance between each indicator, the relative degree of importance between each of the indicators to determine Fuzzy evaluation matrix element is configured, the configuration of the fuzzy evaluation matrix determined set of weights;

[0201] 评判子集确定模块303,用于针对每个指标,根据得到的该指标对应的多个测试值,以及设定的每个性能等级对应的指标测试值范围,确定测试的该指标在每个性能等级上的百分比,以测试的该指标在每个性能等级上的百分比为元素构造评判子集; [0201] Evaluation subset determination module 303 is configured for each indicator, the indicator in accordance with a plurality of corresponding test values ​​obtained, and the performance index for each test value range corresponding to a set level, the index determined in the test the percentage on each performance level to the percentage indicator test performance level on each element of a subset is configured judgment;

[0202] 测试结果确定模块304,用于根据分别针对每个指标确定的评判子集,确定评判矩阵,将确定的权重集与评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0202] The test results determination module 304, configured to judge each subset determined for each index, evaluation matrix is ​​determined, the determined set of weights and evaluation matrix multiplied by the set of the performance evaluation of the network device, The performance Evaluation of the concentration of said element corresponding to the maximum level of performance test results obtained for the determination.

[0203] 图4为本发明实施例提供的与图2所示的另一种测试方法对应的测试装置结构示意图,具体包括: [0203] Fig 4 a schematic view of a test apparatus for testing the structure of another embodiment of the method shown in FIG. 2 provided corresponding to the embodiment of the present invention comprises:

[0204] 测试模块401,用于对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值; [0204] The test module 401, two indicators for each network device performs multiple tests to obtain a plurality of test values ​​for each corresponding secondary indexes;

[0205] 二级权重集确定模块402,用于针对所述网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要程度值,以确定的各二级指标之间的相对重要程度值为元素构造该一级指标对应的二级模糊评判矩阵,根据构造的二级模糊评判矩阵,确定该一级指标对应的二级权重集; [0205] two sets of weights determining module 402, one for each of the metrics for the network device, in accordance with a plurality of test values ​​of the two indexes in an index corresponding to the obtained determining the level indicators values ​​of relative importance between the two indicators of the relative importance between the two indexes to determine the configuration of elements of the value of the index corresponding to a secondary fuzzy evaluation matrix, according to a configuration of two fuzzy evaluation matrix, it is determined that the two weights corresponding to a weight set index;

[0206] 二级评判子集确定模块403,用于针对该一级指标下的每个二级指标,根据得到的该二级指标对应的对个测试值,以及设定的每个性能等级对应的二级指标测试值范围,确定测试的该二级指标在每个性能等级上的百分比,以测试的该二级指标在每个性能等级上的百分比为元素构造该二级指标对应的二级评判子集; [0206] Evaluation two subset determination module 403, used for each of the two indexes in an index, based on test values, and set corresponding to each of the performance level corresponding to the two index obtained percentage in each of two performance level percentage range index test, a test determines the two indicators to the two test performance level indicator on each configuration index corresponds to the two elements of two judgment subset;

[0207] 一级评判集确定模块404,用于根据分别针对该一级指标下的各二级指标构造的二级评判子集,确定该一级指标对应的二级评判矩阵,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,得到该一级指标对应的一级评判集; [0207] Evaluation of a set determination module 404, a judgment for each subset of two secondary indicators each configured in one of the index, an index corresponding to the determined evaluation matrix according to two, the one two indicators corresponding to the set of weights corresponding to a two evaluation index matrix multiplication, to give an evaluation of the set corresponding to an index;

[0208] 一级评判矩阵确定模块405,用于根据分别针对所述网络设备的每个一级指标确定的一级评判集,确定一级评判矩阵; [0208] an evaluation matrix determination module 405, configured to judge each set of level indicators determined for each of said one network device, determines an evaluation matrix;

[0209] 测试结果确定模块406,用于确定所述网络设备的每个一级指标之间的相对重要程度值,以每个一级指标之间的相对重要程度值为元素构造一级模糊评判矩阵,根据构造的一级模糊评判矩阵确定一级权重集,将确定的一级权重集与一级评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 [0209] The test results determination module 406 for determining the relative importance of each of an index value between the network device, to a relative importance between each of the elements is an indicator configured to judge a Fuzzy matrix, is determined according to a fuzzy evaluation matrix configuration of a set of weights, the weight is multiplied by a set of weights determined with an evaluation matrix, to obtain performance evaluation set of the network device, the performance evaluation of the maximum concentration of elements determining a corresponding performance level for the test results obtained.

[0210] 本发明实施例提供一种测试方法及装置,该方法对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值,根据每个指标之间的相对重要性程度值构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集,并确定测试的每个指标在每个性能等级上的百分比,据此确定评判矩阵,将确定的权重集与评判矩阵相乘,得到该网络设备的性能评判集,并将其中最大的元素对应的性能等级确定为得到的测试结果。 Example embodiments provide a method and apparatus for testing [0210] the present invention, the method for each indicator test multiple network devices to obtain a plurality of corresponding test values ​​of each indicator, the relative importance of each metric according to the degree of the fuzzy evaluation matrix configuration value, determined according to weight the fuzzy set evaluation matrix structure weight, and the percentage of each indicator is determined on each test performance level, whereby evaluation matrix is ​​determined, multiplied by the determined set of weights and evaluation matrix, performance evaluation set to give the network device, and wherein the elements corresponding to the maximum level of performance test results obtained for the determination. 通过上述方法,测试装置对每个指标都进行了测试,并综合了各个指标对网络设备的综合性能的影响,因此可以对网络设备的综合性能进行准确的测试。 By the above method, the test device for each indicator were tested, and the combination of overall performance impact indicators each network device, it can accurately test the overall performance of the network device.

[0211] 显然,本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。 [0211] Obviously, those skilled in the art can make various modifications and variations of the present application without departing from the spirit and scope of the present disclosure. 这样,倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内,则本申请也意图包含这些改动和变型在内。 Thus, if these modifications and variations of the present disclosure of the present application claims and their equivalents within the scope thereof, the present application is intended to cover these modifications and variations.

Claims (13)

1.一种测试方法,其特征在于,包括: 对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值; 根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性程度值,以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集; 针对每个指标,根据得到的该指标对应的多个测试值,以及设定的每个性能等级对应的指标测试值范围,确定测试的该指标在每个性能等级上的百分比,以测试的该指标在每个性能等级上的百分比为元素构造评判子集;其中,确定测试的该指标在每个性能等级上的百分比,具体包括:针对每个性能等级,确定得到的该指标对应的多个测试值中,属于该性能等级对应的指标测试值范围内的测试值的个数,将确定的个数与得到的该指标对应的多个测试值的总数 1. A testing method comprising: for each indicator test multiple network devices to obtain a plurality of test values ​​corresponding to each index; a plurality of corresponding test values ​​obtained for each index, determined the relative value between the degree of importance of each indicator, the relative degree of importance between each metric is configured to determine the value of the fuzzy evaluation matrix elements, the fuzzy set evaluation matrix configured to determine the weight of the weight; for each metric, obtained according to a plurality of test values ​​corresponding to the index, and an index range of values ​​for each test performance level corresponding to a set percentage of the indicator in each test is determined performance level to test the performance level indicator on each the percentage of subset element configured judgment; wherein the indicator to determine the percentage of tests on each of the performance level, comprises: for each performance level, a plurality of test values ​​of the index corresponding to the determined resulting in belonging to the performance Total number of test values ​​within the range of index values ​​corresponding to the level test, the number of the determined plurality of test values ​​of the index corresponding to the obtained 比值,确定为测试的该指标在该性能等级上的百分比;以测试的该指标在每个性能等级上的百分比为元素构造评判子集,具体包括:根据该指标在每个性能等级上的百分比,构造评判子集Yi = (yn,yi2,..yiP),其中,Yi为第i个指标的评判子集,P为性能等级的个数,对于Yi中的任意元素yiq,yi(1为该第i个指标在第q个性能等级上的百分比; 根据分别针对每个指标确定的评判子集,确定评判矩阵,将确定的权重集与评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 Ratio, determined as the percentage of the test on the level of performance indicators; the percentage indicator to test performance level on each element is configured to judge subset comprises: on each percentage based on the performance level indicator configured to judge subset Yi = (yn, yi2, .. yiP), where, Yi is the i-th subset evaluation index, P is the number of performance levels, for any element of yiq Yi, yi (1 to the percentage of the i-th index in the individual energy levels of q; the Evaluation of subsets determined for each metric evaluation matrix is ​​determined, multiplied by the determined weights and weight set evaluation matrix, to obtain the performance evaluation of the network device set, and the performance evaluation of the concentration of elements corresponding to the maximum level of performance test results obtained for the determination.
2.如权利要求1所述的方法,其特征在于,对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值,具体包括: 根据设定次数,对网络设备的每个指标进行设定次数的测试,得到每个指标对应的多个测试值; 根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性程度值,具体包括: 以得到的每个指标对应的多个测试值为元素构造第一测试矩阵: 2. The method according to claim 1, wherein each indicator test multiple network devices to obtain a plurality of indicators each corresponding test values ​​comprises: according to the set number of times, the network device each test set number indicators, each indicator to obtain a plurality of values ​​corresponding to the test; in accordance with a plurality of corresponding test values ​​for each metric obtained to determine the relative value between the degree of importance of each indicator comprises: each indicator corresponding to a plurality of test elements of the value obtained in the first test configuration matrix:
Figure CN102447593BC00021
其中,η为所述设定次数,m为所述网络设备的指标的个数,Xij为对m个指标中的第j个指标进行的第i次测试所得到的测试值; 对所述第一测试矩阵进行归一化处理,得到第二测试矩阵: Wherein, η is the set number, m is the index of the number of network devices, Xij is the value of the test to test the m i in the j-th index index of the obtained; the first a test matrix normalized, to obtain a second test matrix:
Figure CN102447593BC00022
其中,Xi ij为第一测试矩阵中的Xu的归一化值; 将所述第二测试矩阵的转置矩阵与所述第二测试矩阵相乘,得到第三测试矩阵: Wherein, Xi ij is the normalized value of the first test the matrix Xu; the second transposed matrix of the test matrix and the second matrix multiplication the test, the third test matrix is ​​obtained:
Figure CN102447593BC00031
根据为每个指标设定的基准值,将每个基准值相应的添加到所述第三测试矩阵中,得到第四测试矩阵: The index set for each reference value, the reference value corresponding to each is added to the third test matrix, to obtain a fourth test matrix:
Figure CN102447593BC00032
其中,为m个指标中第i个指标的基准值; 对所述第四测试矩阵进行归一化处理,得到第五测试矩阵: Wherein, the reference value m i th indicators indicators; the fourth test matrix is ​​normalized to give a fifth test matrix:
Figure CN102447593BC00033
其中,α,u为第四测试矩阵中的^彳的归一化值^^+卩为第四测试矩阵中的βπ+Μ的归一化值; 根据所述第五测试矩阵,采用下述公式确定各指标与各指标的基准值的距离: Wherein, α, u is the fourth test matrix of left foot ^ ^ ^ + normalized value for the normalized values ​​Jie fourth test matrix of βπ + Μ; matrix according to the fifth test, using the following formula to determine the index with the index from the reference value:
Figure CN102447593BC00034
其中,i = I, 2, 3...1iudtlrt为m个指标中的第i个指标与该第i个指标的基准值的距离; 针对任意两个指标,根据下述公式确定该两个指标之间的相对重要性程度值: Where, i = I, 2, 3 ... 1iudtlrt indicators of m i-th index value from the reference index of the i-th; for any two indices, determined according to the formula of the two indicator the relative importance of the degree of value between:
Figure CN102447593BC00035
其中,ru为m个指标中的第i个指标相对于第j个指标的相对重要性程度值。 Wherein, ru indicators of m i-th index value relative degree of importance with respect to the j-th index.
3.如权利要求1或2所述的方法,其特征在于,以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵,具体包括: 构造下述矩阵作为模糊评判矩阵: 3. The method according to claim 1, characterized in that, between the relative importance degree of each indicator element is configured to determine the value of the fuzzy evaluation matrix comprises: configuration of the fuzzy evaluation matrix as the following matrix:
Figure CN102447593BC00041
其中,ru为第i个指标相对于第j个指标的相对重要性程度值,riJ与rji的和值为1,且L的值为0.5 ; 根据构造的模糊评判矩阵确定权重集,具体包括: 对构造的模糊评判矩阵进行模糊一致化处理,得到模糊一致矩阵M: Wherein, Ru is the i-th index value of the relative degree of importance of the j-th index, and rji and Rij is 1, and L is 0.5; the fuzzy set evaluation matrix configuration of the weight determining the weight, comprises: fuzzy evaluation matrix configuration consistent blurring processing, to obtain fuzzy consistent matrix M:
Figure CN102447593BC00042
其中,对于模糊一致矩阵M中的任意元素—台汰台, rik为模糊评判矩阵中第i行第k列的元素, Wherein, for any element of fuzzy consistent matrix M - elimination station units, fuzzy evaluation matrix Rik elements in row i and column k,
Figure CN102447593BC00043
为模糊评判矩阵中第j行第k列的元素; 根据模糊一致矩阵M,釆用下述公式确定每个指标对应的权重: Fuzzy evaluation matrix element in the j row of the k-th column; the fuzzy consistent matrix M, preclude the index determined by the following formula for each corresponding weight:
Figure CN102447593BC00044
其中,Oi为第i个指标对应的权重; 以确定的每个指标对应的权重为元素构成权重集:ω = (CO1, ω2, ω3...(Oni)。 Wherein Oi is the i-th weight corresponding to the weight indexes; index corresponding to each weight determined by the weight element constituting the weight set weight: ω = (CO1, ω2, ω3 ... (Oni).
4.如权利要求1所述的方法,其特征在于,根据分别针对每个指标确定的评判子集,确定评判矩阵,具体包括: 根据分别针对每个指标确定的评判子集,确定下述评判矩阵: 4. The method according to claim 1, characterized in that, based on the judging of subsets determined for each indicator, determining evaluation matrix comprises: based on the judging of subsets determined for each indicator, determined in the following evaluation matrix:
Figure CN102447593BC00045
其中,对于Y中的任意元素yiq,yiq为第i个指标在第q个性能等级上的百分比。 Wherein, for any element in yiq Y, YIQ percentage of the i-th in the first level index q can be personalized.
5.—种测试方法,其特征在于,包括: 对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值;针对所述网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要性程度值,以确定的各二级指标之间的相对重要性程度值为元素构造该一级指标对应的二级模糊评判矩阵,根据构造的二级模糊评判矩阵,确定该一级指标对应的二级权重集; 针对该一级指标下的每个二级指标,根据得到的该二级指标对应的多个测试值,以及设定的每个性能等级对应的二级指标测试值范围,确定测试的该二级指标在每个性能等级上的百分比,以测试的该二级指标在每个性能等级上的百分比为元素构造该二级指标对应的二级评判子集; 根据分别针对该一级指标下的各二级指标 5.- testing methods, characterized by comprising: for each of the two indicator test multiple network devices, to obtain a corresponding plurality of test values ​​for each two indexes; for each one of the network device index, according to a plurality of test values ​​of the two indexes in an index obtained corresponding value to determine the relative importance degree among the two indices at the level indicators to determine the index of each of two the relative importance degree between elements of the value of the index corresponding to a configuration of two fuzzy evaluation matrix, according to two fuzzy evaluation matrix configuration, determining that the two weights corresponding to a weight set index; for this level indicators each secondary index, according to a plurality of test values ​​of the two corresponding to the obtained index, and two indicator test performance level for each range of values ​​corresponding to a set, the two indicators identified in each test performance level percentage, the percentage of the secondary indicators to test the performance level on each of the two construction elements two evaluation indexes corresponding to the subset; according to various indicators in the two-level indicators for each 造的二级评判子集,确定该一级指标对应的二级评判矩阵,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,得到该一级指标对应的一级评判集;根据分别针对所述网络设备的每个一级指标确定的一级评判集,确定一级评判矩阵;确定所述网络设备的每个一级指标之间的相对重要性程度值,以每个一级指标之间的相对重要性程度值为元素构造一级模糊评判矩阵,根据构造的一级模糊评判矩阵确定一级权重集,将确定的一级权重集与一级评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 Judgment made two subsets, determining the indicator corresponding to a two evaluation matrix, the weight of a weight set index corresponding to two multiplied by the corresponding one of the two index evaluation matrix, an indicator to give the the relative importance of each of the level indicators to determine the network device; corresponding to an evaluation set; the evaluation set separately determined for each one index of said one network device, determines an evaluation matrix degree value, the fuzzy evaluation matrix relative importance degree to an index value between each element of a configuration, determining a set of weights according to a fuzzy evaluation matrix configuration, a weight set with a determined weight evaluation matrix multiplied by the performance evaluation of the set of network devices, the performance evaluation of the concentration of elements corresponding to the maximum level of performance test results obtained for the determination.
6.如权利要求5所述的方法,其特征在于,对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值,具体包括: 根据设定次数,对网络设备的每个二级指标进行设定次数的测试,得到每个二级指标对应的多个测试值; 针对所述网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要程度值,具体包括: 针对所述网络设备的每个一级指标,以得到的该一级指标下的每个二级指标对应的多个测试值为元素构造第一测试矩阵: 6. The method according to claim 5, characterized in that, for each two indexes multiple test network device to obtain a plurality of test values ​​for each corresponding secondary indicators comprises: according to the number set, secondary indicators for each network device test set times, to obtain a plurality of test values ​​for each corresponding secondary indexes; metrics for each one of the network device, according to an index in the obtained a plurality of test values ​​of the two indicators corresponding to the determined value of the relative importance between the two indicators in an indicator that comprises: an indicator for each of the network device to obtain the a each secondary indicators corresponding to the plurality of test level indicator is a first test matrix elements configured:
Figure CN102447593BC00051
其中,η为所述设定次数,m为该一级指标下的二级指标的个数,Xgu为对第g个一级指标下的第j个二级指标进行的第i次测试所得到的测试值; 对所述第一测试矩阵进行归一化处理,得到第二测试矩阵: Wherein, η is the set number, m the number of secondary indicators for an indicator in, Xgu i obtained for the first test of the j-th secondary indicators of the indicators of a g th the test value; the first test matrix normalized, to obtain a second test matrix:
Figure CN102447593BC00052
其中,Xgf ij为第一测试矩阵中的xgu的归一化值; 将所述第二测试矩阵的转置矩阵与所述第二测试矩阵相乘,得到第三测试矩阵: Wherein, Xgf ij is the normalized value of the first test the matrix xgu; the second transposed matrix of the test matrix and the second matrix multiplication the test, the third test matrix is ​​obtained:
Figure CN102447593BC00061
根据为该一级指标下的每个二级指标设定的基准值,将每个基准值相应的添加到所述第三测试矩阵中,得到第四测试矩阵: According to each of the secondary indicators for an index setting the reference value, the reference value corresponding to each is added to the third test matrix, to obtain a fourth test matrix:
Figure CN102447593BC00062
其中,β δω+Μ为该第g个一级指标下的第i个二级指标的基准值; 对所述第四测试矩阵进行归一化处理,得到第五测试矩阵: Wherein, β δω + Μ reference value for the i-th index in the two-level indicators of g; the fourth test matrix is ​​normalized to give a fifth test matrix:
Figure CN102447593BC00063
其中,ag' u为第四测试矩阵中的CIgu的归一化值,Pg'm+u为第四测试矩阵中的β gm+1;i的归一化值; 根据所述第五测试矩阵,采用下述公式确定该一级指标下的各二级指标与各二级指标的基准值的距离: Wherein, ag 'u is a normalized value of the fourth test matrix of CIgu, Pg'm + u fourth test matrix β gm + 1; i is a normalized value of; according to the fifth test matrix , using the following formula to determine each of the two indexes in the index with a reference index value of two distances:
Figure CN102447593BC00064
其中,i = 1,2,3...m,dgi,m+1为该第g个一级指标下的第i个二级指标与该第i个二级指标的基准值的距尚; 针对该一级指标下的任意两个二级指标,根据下述公式确定该两个二级指标之间的相对重要性程度值: Where, i = 1,2,3 ... m, dgi, m + 1 for the g i-th index in the two level indicators with the reference value of the i-th still from two indicators; any two of the two-level indicators of the indicators in determining the extent of the relative importance value between the two secondary indicators according to the formula:
Figure CN102447593BC00065
其中,rgij为该第g个一级指标下的第i个二级指标相对于该第g个一级指标下的第j个二级指标的相对重要性程度值。 Wherein, rgij for the i-th index in the two-level indicators of g relative degree of importance of the j-th value in the two indicators g of level indicators.
7.如权利要求5或6所述的方法,其特征在于,以确定的各二级指标之间的相对重要程度值为元素构造该一级指标对应的二级模糊评判矩阵,具体包括: 构造下述矩阵作为该一级指标对应的二级模糊评判矩阵: 7. The method of claim 5 or claim 6, characterized in that the relative importance between the two indexes to determine the configuration of elements of the value of the index corresponding to a two fuzzy evaluation matrix comprises: configuration Examples of the following matrix index corresponding to a secondary fuzzy evaluation matrix:
Figure CN102447593BC00071
其中,rgu为该第g个一级指标下的第i个二级指标相对于该第g个一级指标下的第j个二级指标的相对重要性程度值,rgiJ与rgji的和值为1,且rgn的值为0.5 ; 根据构造的二级模糊评判矩阵,确定该一级指标对应的二级权重集,具体包括: 对构造的二级模糊评判矩阵进行模糊一致化处理,得到二级模糊一致矩阵Mg: Wherein, rgu g of this first phase of the i-th secondary indicators in an indicator of the relative degree of importance value j secondary indicators in the g-th level indicators, rgiJ value and the rgji 1, and a value of 0.5 rgn; the two fuzzy evaluation matrix configuration, determining an index corresponding to the two sets of weights comprises: for two fuzzy judgment matrix configuration is consistent with the fuzzy processing, to obtain two fuzzy consistent matrix Mg:
Figure CN102447593BC00072
其中,对于二级模糊一致矩阵Mg中的任意元素Igij, Wherein, for any two elements in the Mg matrix fuzzy consistent Igij,
Figure CN102447593BC00073
rgik为二级模糊评判矩阵中第i行第k列的元素,rgjk为二级模糊评判矩阵中第j行第k列的元素; 根据二级模糊一致矩阵Mg,釆用下述公式确定该一级指标下的每个二级指标对应的权重: rgik two fuzzy evaluation matrix for the element in row i and column k, rgjk two fuzzy evaluation of matrix elements of the j-th row of the k-th column; according to the two fuzzy consistent matrix Mg, which preclude a determined by the following equation each two-level indicators in indicators weight corresponding to:
Figure CN102447593BC00074
其中,Ogi为该第g个一级指标下的第i个二级指标对应的权重; 以确定的该一级指标下的每个二级指标对应的权重为元素构成该一级指标对应的二级权重集:ω g = (COg1, Qg2, Qg3...ω gm)。 Wherein, Ogi for the i-th secondary indicators corresponding to the weight in g of level indicators; each two indexes in the index to determine a corresponding weight which constitutes an index for the element corresponding to two stage set of weights: ω g = (COg1, Qg2, Qg3 ... ω gm).
8.如权利要求5所述的方法,其特征在于,根据得到的该二级指标对应的多个测试值,以及设定的每个性能等级对应的二级指标测试值范围,确定测试的该二级指标在每个性能等级上的百分比,具体包括: 针对每个性能等级,确定得到的该二级指标对应的多个测试值中,属于该性能等级对应的指标测试值范围内的测试值的个数,将确定的个数与得到的该二级指标对应的多个测试值的总数的比值,确定为测试的该二级指标在该性能等级上的百分比; 以测试的该二级指标在每个性能等级上的百分比为元素构造该二级指标对应的二级评判子集,具体包括: 根据该二级指标在每个性能等级上的百分比,构造下述二级评判子集: Ygi = (ygn,ygi2,ygi3...ygiP),其中,Ygi为第g个一级指标下的第i个二级指标的二级评判子集,P为性能等级的个数,对于Ygi中的任意元素ygiq, 8. The method according to claim 5, wherein a plurality of test values ​​of the two corresponding to the obtained index, and two indicator test performance level for each range of values ​​corresponding to a set, determining that the test the percentage of secondary indexes on each performance level, comprises: a plurality of test values ​​of the two indicators corresponding for each of the performance levels obtained determined, the value of the index belonging to the test range of the test value of the performance level corresponding to the the ratio of the total number, the number of the determined plurality of test values ​​corresponding to the two index obtained, the test determined the percentage of the two indices of performance levels; the two indicators for testing on each performance level as a percentage of two judgment element is configured corresponding to the subset of the two indicator comprises: according to the percentage in each of the two indicators of the performance level, the following two judgment configured subsets: YGI = (ygn, ygi2, ygi3 ... ygiP), wherein, YGI g for the first secondary subset of the i-th judgment indicators in two level indicators, P is the number of performance levels for the YGI any element ygiq, ygiq为该g个一级指标下的第i个二级指标在第q个性能等级上的百分比。 ygiq percentages for the i-th index in the two-level indicators on a g q-th individual energy levels.
9.如权利要求8所述的方法,其特征在于,根据分别针对该一级指标下的各二级指标构造的二级评判子集,确定该一级指标对应的二级评判矩阵,具体包括: 根据分别针对该一级指标下的各二级指标确定的二级评判子集,确定下述二级评判矩阵: 9. The method according to claim 8, characterized in that, for each subset of two two evaluation indexes in the configuration of each of an index, an index is determined that the corresponding two evaluation matrix, comprising the specific : Evaluation the two subsets are determined for each of the two indexes in an index, determined by the following two evaluation matrix:
Figure CN102447593BC00081
其中,对于Yg中的任意元素ygiq,ygi(1为该第g个一级指标下的第i个二级指标在第q个性能等级上的百分比。 Wherein, for any element of ygiq Yg, YGI (percentage of 1 g for the i-th index in the two-level indicators on the individual energy levels of the q.
10.如权利要求9所述的方法,其特征在于,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,得到该一级指标对应的一级评判集,具体包括: 采用下述公式确定该一级指标对应的一级评判集: 10. The method according to claim 9, wherein the weight indexes corresponding to a set of two weight multiplied by the corresponding one of the two index evaluation matrix, an indicator to give an evaluation of the corresponding set, comprises: using the following formula to determine an evaluation index corresponding to one of the set of:
Figure CN102447593BC00082
其中,Sg为第g个一级指标对应的一级评判集,COg= (COg1, Qg2, Qg3...COgJ为确定的该第g个一级指标对应的二级权重集,Sg1, Sg2, Sg3...Sgp为该第g个一级指标分别对应P个性能等级的评判权重; 根据分别针对所述网络设备的每个一级指标确定的一级评判集,确定一级评判矩阵,具体包括: 根据分别针对所述网络设备的每个一级指标确定的一级评判集,确定下述一级评判矩阵: Wherein, Sg is the g-th an evaluation index corresponding to a current collector, COg = (COg1, Qg2, Qg3 ... COgJ to determine the g-level indicators corresponding to the two sets of weights, Sg1, Sg2, Sg3 ... Sgp g of this first-level indicators respectively correspond to individual energy level evaluation right weight P; evaluation matrix separately for each of a set of evaluation index of the network device determines a level determined according to a particular comprising: the evaluation set is determined separately for each one index of said one network device, an evaluation matrix is ​​determined by the following:
Figure CN102447593BC00083
其中,S为确定的一级评判矩阵,h为所述网络设备的一级指标的个数,对于一级评判矩阵S中的任意元素Sg,,Sgq为所述网络设备的h个一级指标中的第g个一级指标对应第q个性能等级的评判权重。 Wherein, S is the determination of an evaluation matrix, h is the index number of a network device, a judgment for any element in the matrix S Sg ,, Sgq h to the network device level indicators the g-th level indicators corresponding to individual can judge right q-level weights.
11.如权利要求5所述的方法,其特征在于,以每个一级指标之间的相对重要程度值为元素构造一级模糊评判矩阵,具体包括: 构造下述矩阵作为所述一级模糊评判矩阵: 11. The method according to claim 5, characterized in that the fuzzy evaluation matrix to relative importance between an index value of each element of a structure, comprises: a configuration, as the following matrix Fuzzy evaluation matrix:
Figure CN102447593BC00091
其中,h为所述网络设备的一级指标的个数,r^.为所述网络设备的h个一级指标中的第i个一级指标相对于第j个一级指标的相对重要性程度值,与的和值为1,且rii的值为0.5 ; 根据构造的一级模糊评判矩阵确定一级权重集,具体包括: 对构造的一级模糊评判矩阵进行模糊一致化处理,得到一级模糊一致矩阵M: Wherein, h is an index of the number of network devices, r ^. H level indicators to the network device in an i-th index with respect to the j-th level indicators relative importance of degree value, and the value of 1, and a value of 0.5 rii; determining a set of weights according to a fuzzy evaluation matrix configuration comprises: a pair of fuzzy blur evaluation matrix configuration consistent treatment, to obtain a level fuzzy consistent matrix M:
Figure CN102447593BC00092
其中,对于一级模糊一致矩阵M中的任意元素 Wherein a fuzzy consistent for any element of the matrix M
Figure CN102447593BC00093
rik为一级模糊评判矩阵中第i行第k列的元素,rJk为一级模糊评判矩阵中第j行第k列的元素; 根据一级模糊一致矩阵M,采用下述公式确定每个一级指标对应的权重: rik is a blur evaluation matrix element in row i and column k, rJk a fuzzy evaluation of matrix elements of the j-th row of the k-th column; according to a fuzzy consistent matrix M, using the following formula to determine each a right-level indicators corresponding weight:
Figure CN102447593BC00094
其中,COi为该第i个一级指标对应的权重; 以确定的每个一级指标对应的权重为元素构成一级权重集:ω = (CO1, ω2,O 3...0 h) ο Wherein, COi an index for the i-th corresponding weight; the weight of each one of the indicators to determine a corresponding weight of an element constituting the weight set: ω = (CO1, ω2, O 3 ... 0 h) ο
12.一种测试装置,其特征在于,包括: 测试模块,用于对网络设备的每个指标进行多次测试,得到每个指标对应的多个测试值; 权重集确定模块,用于根据得到的每个指标对应的多个测试值,确定每个指标之间的相对重要性程度值,以确定的每个指标之间的相对重要性程度值为元素构造模糊评判矩阵,根据构造的模糊评判矩阵确定权重集; 评判子集确定模块,用于针对每个指标,根据得到的该指标对应的多个测试值,以及设定的每个性能等级对应的指标测试值范围,确定测试的该指标在每个性能等级上的百分t匕,以测试的该指标在每个性能等级上的百分比为元素构造评判子集;其中,确定测试的该指标在每个性能等级上的百分比,具体包括:针对每个性能等级,确定得到的该指标对应的多个测试值中,属于该性能等级对应的指标测试值范围内的测 12. A test apparatus comprising: a test module, each indicator for multiple network test equipment, to obtain a corresponding plurality of test values ​​for each index; weight set determining module, configured to obtain each index corresponding to the plurality of test values, determining the relative values ​​between the degree of importance of each indicator, the relative degree of importance between each metric is configured to determine the value of the fuzzy evaluation matrix element, constructed in accordance with Fuzzy Evaluation matrix determining sets of weights; Evaluation subset determination module, configured to, for each index, a plurality of test values ​​corresponding to the obtained index, and an index range of values ​​for each test performance level corresponding to a set, a test to determine the index on each level of performance of the index wherein the percentage, the test is determined, comprises; percentage on each performance level percentage dagger t, to test the performance level indicator on each element of a subset is configured to judge : performance level for each of the plurality of test values ​​of the index corresponding to the determined resultant belonging index measuring the test value of the performance level corresponding to the range of 值的个数,将确定的个数与得到的该指标对应的多个测试值的总数的比值,确定为测试的该指标在该性能等级上的百分比;以测试的该指标在每个性能等级上的百分比为元素构造评判子集,具体包括:根据该指标在每个性能等级上的百分比,构造评判子集Yi = (yn,yi2» yi3...yip),其中,Yi为第i个指标的评判子集,P为性能等级的个数,对于Yi中的任意元素yiq,yiq为该第i个指标在第q个性能等级上的百分比; 测试结果确定模块,用于根据分别针对每个指标确定的评判子集,确定评判矩阵,将确定的权重集与评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果。 Number of values, the total number of the determined plurality of test values ​​of the index corresponding to the ratio obtained, determined as the percentage of the indicator in the test level of performance; indicator to the performance level in each test the percentage of the judgment element is configured subset comprises: percentage of the indicator in accordance with the level of each performance, configuration subset judgment Yi = (yn, yi2 »yi3 ... yip), where, Yi is the i th Evaluation subset index, P is the number of performance levels, for any element in yiq Yi, YIQ percentage for the i-th index in the q-th individual energy levels; test result determination module for separately for each Evaluation subset indicators determined evaluation matrix is ​​determined, the weight and the determined weight set evaluation matrix multiplied by the performance evaluation of the set of network devices, the performance evaluation of the concentration of elements corresponding to the maximum performance level is determined to be obtained Test Results.
13.—种测试装置,其特征在于,包括: 测试模块,用于对网络设备的每个二级指标进行多次测试,得到每个二级指标对应的多个测试值; 二级权重集确定模块,用于针对所述网络设备的每个一级指标,根据得到的该一级指标下的各二级指标对应的多个测试值,确定该一级指标下的各二级指标之间的相对重要程度值,以确定的各二级指标之间的相对重要程度值为元素构造该一级指标对应的二级模糊评判矩阵,根据构造的二级模糊评判矩阵,确定该一级指标对应的二级权重集; 二级评判子集确定模块,用于针对该一级指标下的每个二级指标,根据得到的该二级指标对应的对个测试值,以及设定的每个性能等级对应的二级指标测试值范围,确定测试的该二级指标在每个性能等级上的百分比,以测试的该二级指标在每个性能等级上的百分比为元素构造该二 13.- kinds of testing apparatus, characterized by comprising: a test module, for each of the two indicator test multiple network devices to obtain a plurality of test values ​​for each corresponding secondary indexes; two sets of weights determined module, a plurality of test values ​​for each one index of the network device, according to various indicators in the two resulting index corresponding to a determined between the two indexes in the index a relative importance value, the relative importance between the two indexes to determine the configuration of elements of the value of the index corresponding to a secondary fuzzy evaluation matrix, according to a configuration of two fuzzy evaluation matrix, an index corresponding to the determined two sets of weights; two judgment subset determination module, configured to, for each index in the two-level indicators, based on the two test values ​​corresponding to the obtained index, and the performance level is set for each the percentage on each performance level percentage two index values ​​corresponding to the range test to determine the two test indicator, two indicators to test the performance level on each of the two construction elements 级指标对应的二级评判子集; 一级评判集确定模块,用于根据分别针对该一级指标下的各二级指标构造的二级评判子集,确定该一级指标对应的二级评判矩阵,将该一级指标对应的二级权重集与该一级指标对应的二级评判矩阵相乘,得到该一级指标对应的一级评判集; 一级评判矩阵确定模块,用于根据分别针对所述网络设备的每个一级指标确定的一级评判集,确定一级评判矩阵; 测试结果确定模块,用于确定所述网络设备的每个一级指标之间的相对重要程度值,以每个一级指标之间的相对重要程度值为元素构造一级模糊评判矩阵,根据构造的一级模糊评判矩阵确定一级权重集,将确定的一级权重集与一级评判矩阵相乘,得到所述网络设备的性能评判集,将所述性能评判集中最大的元素对应的性能等级确定为得到的测试结果O Stage two evaluation indexes corresponding to the subset; determining a set of evaluation means for evaluation for each subset of two secondary indicators each configured at the level indicators to determine the two evaluation indexes corresponding to the one according to matrix, and the weight of a weight set index corresponding to two multiplied by the corresponding one of the two index evaluation matrix, to obtain a set of evaluation index corresponding to the one; an evaluation matrix determining module configured to respectively an evaluation set for each of the index determined by a network device, a determination evaluation matrix; test result determination means for determining the relative importance of each of an index value between the network device, Fuzzy to relative importance between an index value of each element is configured an evaluation matrix, determining a set of weights according to a fuzzy evaluation matrix configuration, the weight is multiplied by a set of weights determined with an evaluation matrix to obtain performance evaluation set of the network device, the performance evaluation of the concentration of elements corresponding to the maximum performance level is determined as a test result obtained O
CN 201110372395 2011-11-21 2011-11-21 Test method and device CN102447593B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201110372395 CN102447593B (en) 2011-11-21 2011-11-21 Test method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201110372395 CN102447593B (en) 2011-11-21 2011-11-21 Test method and device

Publications (2)

Publication Number Publication Date
CN102447593A CN102447593A (en) 2012-05-09
CN102447593B true CN102447593B (en) 2014-07-02

Family

ID=46009700

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201110372395 CN102447593B (en) 2011-11-21 2011-11-21 Test method and device

Country Status (1)

Country Link
CN (1) CN102447593B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102967798B (en) * 2012-11-15 2015-10-21 深圳大学 A method and system for fault alarm electrical equipment
CN103267834B (en) * 2013-05-19 2015-05-20 山东出入境检验检疫局检验检疫技术中心 Comprehensive detection and judgment system and method for quality of cast tin-lead solder product
CN104023353A (en) * 2014-05-28 2014-09-03 上海斐讯数据通信技术有限公司 Router performance test method and test system
CN106713070A (en) * 2016-12-23 2017-05-24 中国铁路信息技术中心 Information monitoring method

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101834679A (en) 2010-04-26 2010-09-15 福建星网锐捷网络有限公司 Port test method, device and system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7217510B2 (en) * 2001-06-26 2007-05-15 Isis Pharmaceuticals, Inc. Methods for providing bacterial bioagent characterizing information

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101834679A (en) 2010-04-26 2010-09-15 福建星网锐捷网络有限公司 Port test method, device and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
朱畅华等.网络测量及其关键技术.《西安电子科技大学学报(自然科学版)》.2002,第29卷(第6期),第813-817页.
黄志忠.基于测量的网络管理技术研究.《中国优秀硕士学位论文全文数据块》.2011,正文第39-64页.

Also Published As

Publication number Publication date
CN102447593A (en) 2012-05-09

Similar Documents

Publication Publication Date Title
Hurlin et al. Second generation panel unit root tests
Hatefi et al. A common weight MCDA–DEA approach to construct composite indicators
Zhang et al. An empirical study on using the national vulnerability database to predict software vulnerabilities
Brümmer et al. The speaker partitioning problem.
CN103582884A (en) Robust feature matching for visual search
Rönkkö et al. On the adoption of partial least squares in psychological research: Caveat emptor
Brooijmans et al. New Physics at the LHC. A Les Houches Report: Physics at TeV Colliders 2009-New Physics Working Group
SG145708A1 (en) Inspection method and apparatus, lithographic apparatus, lithographic processing cell and device manufacturing method
Guo et al. Decomposition weights and overall efficiency in two-stage additive network DEA
Barrada et al. Incorporating randomness in the Fisher information for improving item‐exposure control in CATs
Basseur et al. A preliminary study on handling uncertainty in indicator-based multiobjective optimization
CN104881706A (en) Electrical power system short-term load forecasting method based on big data technology
Liu et al. Establishing an objective system for the assessment of public acceptance of nuclear power in China
CN104036501A (en) Three-dimensional image quality objective evaluation method based on sparse representation
Otranto Identifying financial time series with similar dynamic conditional correlation
Fayle et al. Reducing over-reporting of deterministic co-occurrence patterns in biotic communities
Kaneko et al. A new process variable and dynamics selection method based on a genetic algorithm‐based wavelength selection method
Ciriaci et al. The role of knowledge‐based supply specialisation for competitiveness: A spatial econometric approach
CN101933085B (en) Objective measurement of audio quality
CN103247008B (en) A method for power quality assessment of statistical indicators
Bornmann et al. How to improve the prediction based on citation impact percentiles for years shortly after the publication date?
CN103106279B (en) Species while clustering node attributes and relationships based on the similarity of structure
CN104964950B (en) Recognition Method cuttings spectrum based on laser-induced breakdown peaks multielement
Peng et al. Developing and evaluating tree height-diameter models at three geographic scales for black spruce in Ontario
Mayoral Testing for fractional integration versus short memory with structural breaks

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted