(19)
(11)EP 3 182 656 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
26.06.2019 Bulletin 2019/26

(21)Application number: 15831718.0

(22)Date of filing:  05.08.2015
(51)International Patent Classification (IPC): 
H04L 12/751(2013.01)
H04L 12/26(2006.01)
H04L 12/24(2006.01)
(86)International application number:
PCT/CN2015/086155
(87)International publication number:
WO 2016/023438 (18.02.2016 Gazette  2016/07)

(54)

METHOD AND DEVICE FOR DISCOVERING A NETWORK TOPOLOGY

VERFAHREN UND VORRICHTUNG ZUR ENTDECKUNG EINER NETZWERKTOPOLOGIE

PROCÉDÉ ET DISPOSITIF POUR DÉCOUVRIR UNE TOPOLOGIE DE RÉSEAU


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 12.08.2014 CN 201410395692

(43)Date of publication of application:
21.06.2017 Bulletin 2017/25

(73)Proprietor: Huawei Technologies Co., Ltd.
Longgang District Shenzhen, Guangdong 518129 (CN)

(72)Inventors:
  • YUAN, Yulin
    Shenzhen Guangdong 518129 (CN)
  • FAN, Xiaoji
    Shenzhen Guangdong 518129 (CN)
  • YE, Zhiming
    Shenzhen Guangdong 518129 (CN)

(74)Representative: Isarpatent 
Patent- und Rechtsanwälte Behnisch Barth Charles Hassa Peckmann & Partner mbB Friedrichstrasse 31
80801 München
80801 München (DE)


(56)References cited: : 
WO-A1-2009/118050
CN-A- 103 828 310
US-A1- 2002 165 957
US-A1- 2009 116 404
CN-A- 101 874 388
CN-A- 104 125 154
US-A1- 2008 170 513
US-A1- 2012 131 211
  
  • Denys Knertster ET AL: "Network Device Discovery", Degree project in Communication Systems, 10 June 2013 (2013-06-10), pages 1-97, XP055481084, Stockholm, Sweden ISBN: 978-3-8114-4549-9 Retrieved from the Internet: URL:https://people.kth.se/~maguire/DEGREE- PROJECT-REPORTS/130610-Denys_Knertser_and_ Victor_Tsarinenko-with-cover.pdf [retrieved on 2018-06-05]
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

TECHNICAL FIELD



[0001] Embodiments of the present invention relate to the field of network connection detection, and in particular, to a network topology discovery method and device.

BACKGROUND



[0002] In a stage of network evaluation and optimization, an operator needs to perform evaluation and optimization on a network and analyze a network element in the network and a service status. For example, the operator needs to collect configuration information of a network element, collect traffic information of a port, evaluate a capacity of the port, discover a port with overloaded traffic, and perform capacity expansion or adjust a flow path for the port. When an optimization and analysis tool is used to perform the network evaluation and optimization, generating a network topology needs to depend on the optimization and analysis tool. Network traffic evaluation, service evaluation, and emulation can be performed only based on the network topology and a result of network evaluation and analysis is displayed based on the network topology.

[0003] A network topology discovery method in the prior art is collecting network characteristic data of a network element in a to-be-analyzed network, obtaining, by means of calculation according to the collected network characteristic data and a corresponding network topology discovery algorithm, a link set corresponding to the algorithm, and further obtaining a network topology. For example, a network topology based on a port Internet Protocol (Internet Protocol, IP) address characteristic is obtained by means of calculation according to a port IP address and an IP address matching algorithm, or a network topology based on a port alias characteristic is obtained by means of calculation according to a port alias and a port alias matching algorithm, or a network topology of a network established based on Cisco devices (only a network established based on Cisco devices is supported) may be obtained according to the Cisco Discovery Protocol (Cisco Discovery Protocol, CDP), or the like.

[0004] In the prior art, when network topology discovery is performed based on a single type of network characteristic data, in a case in which network topology discovery is performed when a to-be-analyzed network does not support this type of network characteristic or a corresponding algorithm, accuracy of network topology discovery is low; or in a supporting case, because data is not completely collected or data is inconsistent, uncertainty exists in a link relationship in the network, and accuracy of network topology discovery is also relatively low. In addition, when network topology discovery is performed by using multiple types of network characteristic data, no comprehensive analysis for improving accuracy of network topology discovery is performed on multiple obtained network topology discovery results.

[0005] US 2012/0131211 A1 discloses a computing device that retrieves a set of configuration files of devices associated with a network to be analyzed and extracts from the set of configuration files Internet protocol (IP) address information to form an IP address database. The computing device also extracts from the set of configuration files border gateway protocol (BGP) routing protocols to form a BGP routing database. The computing device analyzes the IP address database and the BGP routing database to generate a database of calculated network connection information for the devices associated with the network. The computing device conducts live network discovery, on the network and based on the IP address database, to generate a database of discovered connection information, and merges the database of calculated network connection information and the database of discovered connection information to form a master database from which output files may be generated.

SUMMARY



[0006] The present invention is defined in the independent claims. The dependent claims define particular embodiments of the invention.

[0007] Aspects of the present invention provide a network topology discovery method and device, which can perform a comprehensive analysis on results obtained after network topology discovery is performed by using multiple types of network characteristic data, and improve accuracy of network topology discovery.

[0008] According to a first aspect, an embodiment of the present invention provides a network topology discovery method, where the method includes:

collecting network characteristic data of all network elements in a to-be-analyzed network;

obtaining at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gathering all links in the at least two link subsets into one set to obtain a first link set, where a confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to the link subset, confidence values of different topology discovery algorithms are different, and the link is used to connect two ports of different network elements;

obtaining a second link set by performing an operation on the first link set, where the operation includes:

combining same links in the first link set, and calculating, according to multiple confidence values of the same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining, wherein the same links are at least two links in which two ports in one link are the same as those in any other link; and

for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link; and

obtaining a network topology of the to-be-analyzed network according to each link in the second link set.



[0009] With reference to the first aspect, in a first possible implementation manner, the operation further includes: after the retaining a link having a largest confidence value in the at least two links and deleting a remaining link,
comparing the confidence values of the links in the first link set with a preset threshold, and selecting a link whose confidence value is greater than the preset threshold.

[0010] With reference to the first aspect, in a second possible implementation manner, the uncertainty reasoning algorithm includes the following:

where CFi (H) is a confidence value among multiple confidence values of the same links, CFj (H) is another confidence value among the multiple confidence values of the same links, and CFi,j (H) is a new confidence value of the same links that is calculated according to CFi (H) and CFj (H).

[0011] With reference to any possible implementation manner of the first aspect to the second possible implementation manner of the first aspect, in a third possible implementation manner,
the network characteristic data and the corresponding topology discovery algorithm include at least two types of the following combinations: a port Internet Protocol (IP) address and an Internet Protocol (IP) address matching algorithm, a port alias and a port alias matching algorithm, or port Link Layer Discovery Protocol (LLDP) neighbor information and a port Link Layer Discovery Protocol (LLDP) link algorithm.

[0012] According to a second aspect, an embodiment of the present invention provides a network topology discovery device, where the device includes:

a collection unit, configured to collect network characteristic data of all network elements in a to-be-analyzed network;

a link obtaining unit, configured to obtain at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gather all links in the at least two link subsets into one set to obtain a first link set, where a confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to the link subset, confidence values of different topology discovery algorithms are different, and the link is used to connect two ports of different network elements;

a link processing unit, configured to obtain a second link set by performing an operation on the first link set, where the operation includes:

combining same links in the first link set, calculating, according to multiple confidence values of the same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining, wherein the same links are at least two links in which two ports comprised in one link are the same as those in any other link; and

for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link; and

a topology obtaining unit, configured to obtain a network topology of the to-be-analyzed network according to each link in the second link set.



[0013] With reference to the second aspect, in a first possible implementation manner, the link processing unit is further configured to:
after the retaining a link having a largest confidence value in the at least two links and deleting a remaining link, comparing the confidence values of the links in the first link set with a preset threshold, and selecting a link whose confidence value is greater than the preset threshold.

[0014] With reference to the second aspect, in a second possible implementation manner, the uncertainty reasoning algorithm includes the following:

where CFi (H) is a confidence value among multiple confidence values of the same links, CFj (H) is another confidence value among the multiple confidence values of the same links, and CFi,j (H) is a new confidence value of the same links that is calculated according to CFi (H) and CFj (H).

[0015] With reference to any possible implementation manner of the second aspect to the second possible implementation manner of the second aspect, in a third possible implementation manner,
the network characteristic data and the corresponding topology discovery algorithm include at least two types of the following combinations: a port Internet Protocol (IP) address and an Internet Protocol (IP) address matching algorithm, a port alias and a port alias matching algorithm, or port Link Layer Discovery Protocol (LLDP) neighbor information and a port Link Layer Discovery Protocol (LLDP) link algorithm.

[0016] According to the network topology discovery method and device provided in the embodiments of the present invention, first, network characteristic data of all network elements in a to-be-analyzed network is collected; next, at least two corresponding link subsets are obtained respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and all links in the at least two link subsets are gathered into one set to obtain a first link set; then, a second link set is obtained by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link; and last, a network topology of the to-be-analyzed network is obtained according to each link in the second link set. In this way, a comprehensive analysis may be performed on results obtained after network topology discovery is performed by using multiple types of network characteristic data, so as to improve accuracy of network topology discovery.

BRIEF DESCRIPTION OF DRAWINGS



[0017] To describe the technical solutions in the embodiments of the present invention or in the prior art more clearly, the following briefly describes the accompanying drawings required for describing the embodiments or the prior art. Apparently, the accompanying drawings in the following description show some embodiments of the present invention, and persons of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.

FIG. 1 is a schematic flowchart 1 of a network topology discovery method according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of effects of a hypothetical actual network topology and physical network topologies obtained by using three types of topology discovery algorithms according to an embodiment of the present invention;

FIG. 3 is a schematic flowchart 2 of a network topology discovery method according to an embodiment of the present invention;

FIG. 4 is a schematic structural diagram 1 of a network topology discovery device according to an embodiment of the present invention; and

FIG. 5 is a schematic structural diagram 2 of a network topology discovery device according to an embodiment of the present invention.


DESCRIPTION OF EMBODIMENTS



[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following clearly and completely describes the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] An embodiment of the present invention provides a network topology discovery method. As shown in FIG. 1, the method includes:

[0020] Step 101. Collect network characteristic data of all network elements in a to-be-analyzed network.

[0021] Step 102. Obtain at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gather all links in the at least two link subsets into one set to obtain a first link set.

[0022] A confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to each link subset, confidence values of different topology discovery algorithms are different, and the link is a link that consists of two ports of different network elements.

[0023] Step 103. Obtain a second link set by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link.

[0024] Step 104. Obtain a network topology of the to-be-analyzed network according to each link in the second link set.

[0025] According to the network topology discovery method provided in this embodiment of the present invention, first, network characteristic data of all network elements in a to-be-analyzed network is collected; next, at least two corresponding link subsets are obtained respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and all links in the at least two link subsets are gathered into one set to obtain a first link set; and a second link set is obtained by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link; and last, a network topology of the to-be-analyzed network is obtained according to each link in the second link set. In this way, a comprehensive analysis may be performed on results obtained after network topology discovery is performed by using multiple types of network characteristic data, so as to improve accuracy of network topology discovery.

[0026] To make persons skilled in the art better understand the technical solution provided in this embodiment of the present invention, the following describes in detail the network topology discovery method provided in this embodiment of the present invention with reference to a specific embodiment.

[0027] Before the technical solution provided in this embodiment is described, some basic content in the technical solution is briefly described as follows:

[0028] In the technical solution provided in this embodiment, a network element is a network unit or node in a network system, and the unit is a device that can independently complete one or more functions. For example, in a GSM network system, a base station is a network element; an entity that can independently complete a function may become a network element, and therefore, a switch, a router or the like is also a network element; a link may be a physical link or a logical link.

[0029] A set that includes all network elements in a to-be-analyzed network is defined as N={N1, N2, ...Nn}, and a set that includes all ports of all network elements is P, where the jth port of the ith network element Ni is denoted as Pij, a range of i is [1, n], a range of j is [1, m], m is a quantity of ports corresponding to the network element Ni, and different network elements may have different values of m.

[0030] Network characteristic data of all network elements in the to-be-analyzed network may be collected by using a collection tool, and the network characteristic data is used as data input of a topology discovery algorithm. It is definite that the foregoing collecting network characteristic data of network elements by using a collection tool can be implemented by all persons of ordinary skills in the art. The network characteristic data includes at least two of the following: a port IP address, a port alias, port LLDP neighbor information, a network element name, port traffic, a Media Access Control (Media Access Control, MAC) forwarding table, an Address Resolution Protocol (Address Resolution Protocol, ARP) forwarding table, a routing and forwarding table, or virtual local area network (virtual local area network, VLAN) configuration information.

[0031] To facilitate description of the technical solution provided in this embodiment of the present invention, in the following embodiment, it is assumed that a quantity of network elements in the to-be-analyzed network is 5 (that is, n=5), a set that includes all network elements is N={N1, N2, N3, N4, N5}, a quantity of ports of each network element is 4 (that is, m=4), and a set that includes all physical ports of all the network elements is P={P11, P12, P13, P14, P21, P22, ..., P51, P52, P53, P54}; network characteristic data to be selected for collection during collection is a port IP address, a port alias, and port LLDP neighbor information; and correspondingly, topology discovery algorithms to be used are an IP address matching algorithm, a port alias algorithm, and a port LLDP link algorithm. It should be noted that such a selection manner is only exemplary, and is intended only to help describe the technical solution in this embodiment, and in actual application, persons skilled in the art may collect, according to an actual requirement, network characteristic data and select a corresponding topology algorithm.

[0032] It is assumed that an actual network topology and network topologies obtained by means of calculation respectively by using the IP address matching algorithm, the port alias algorithm and the port LLDP link algorithm are shown in FIG. 2.

[0033] FIG. 3 shows a network topology discovery method that is provided in this embodiment of the present invention and that is based on the foregoing content. The method includes:
Step 201. Collect network characteristic data of all network elements in a to-be-analyzed network.

[0034] Exemplarily, a port IP address, a port alias, and port LLDP neighbor information that are of each of 20 ports of 5 network elements in the to-be-analyzed network are collected.

[0035] Step 202. Obtain at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gather all links in the at least two link subsets into one set to obtain a first link set.

[0036] A confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to each link subset, confidence values of different topology discovery algorithms are different, and the link is a link that consists of two ports of different network elements.

[0037] Exemplarily, the port IP addresses of the 20 ports are used as input of the IP address matching algorithm, a link subset L1 corresponding to the algorithm is obtained by means of calculation, and sequentially, a corresponding link subset L2 is obtained by means of calculation according to the port aliases and the port alias matching algorithm and a corresponding link subset L3 is obtained by means of calculation according to the port LLDP neighbor information and the port LLDP link algorithm. Links in the link subsets L1, L2, and L3 are gathered into one set to obtain a first link set G, that is, G=L1+L2+L3. There may be two cases for the links in the first link set G. Case 1: Two or more same links exist, where two same links mean that two ports included in one link are the same as those in the other link. Case 2: Two or more links having only one same port exist.

[0038] The link subset includes several links, for example, Li=Li1 ULi2 U... ULik, where Lik is the kth link obtained by means of calculation according to a corresponding algorithm i, k∈[1, K], and K is a quantity of links obtained by means of calculation according to the corresponding algorithm; the link Lik consists of two ports, Lik=(port 1, port 2), the port 1 and the port 2 are different ports, both the port 1 and the port 2 belong to the set P, and there is no sequence between the port 1 and the port 2 in one link Lik.

[0039] Different confidence values CFs are set according to different topology discovery algorithms (or corresponding network characteristic data), and a range of the confidence value CF may be set to [-1, 1]. In addition, confidence values of all links in a link set obtained by means of calculation according to one topology discovery algorithm are the same as a confidence value of the topology discovery algorithm (or corresponding network characteristic data).

[0040] To make persons skilled in the art better understand the foregoing content, refer to FIG. 2 and the following tables.
Table 1
AlgorithmAlgorithm nameCorresponding network characteristic dataAlgorithm confidence value
1 IP address matching algorithm Port IP address 0.8
2 Port alias matching algorithm Port alias 0.5
3 LLDP link algorithm Port LLDP neighbor information 1.0
Table 2
AlgorithmAlgorithm nameCorresponding network characteristic dataAlgorithm confidenceCorresponding link subsets and confidence values corresponding to links
1 IP address matching algorithm Port IP address 0.8 L11(P11, P21)=0.8
L12(P22, P32)=0.8
L13(P33, P43)=0.8
2 Port alias matching algorithm Port alias 0.5 L21(P11, P21)=0.5
L22(P22, P32)=0.5
L23(P34, P44)=0.5
L24(P41, P51)=0.5
L25(P54, P14)=0.5
3 Port LLDP link algorithm Port LLDP neighbor information 1.0 L31(P22, P32)=1.0
L32(P33, P43)=1.0
L33(P44, P54)=1.0
Table 3
Link setLinks and confidence values corresponding to the links
G L11(P11, P21)=0.8
  L12(P22, P32)=0.8
  L13(P33, P43)=0.8
  L21(P11, P21)=0.5
  L22(P22, P32)=0.5
  L23(P34, P44)=0.5
  L24(P41, P51)=0.5
  L25(P54, P14)=0.5
  L31(P22, P32)=1.0
  L32(P33, P43)=1.0
  L33(P44, P54)=1.0


[0041] Step 203. Combine same links in the first link set, and calculate, according to respective confidence values of the multiple same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining.

[0042] Exemplarily, for at least two same links existing in the first link set G, the multiple same links are combined, that is, only one same link in the first link set G is retained, and the confidence value of the retained link after combining is calculated according to respective confidence values of the multiple same links and the uncertainty reasoning algorithm. Refer to the following Table 4.
Table 4
Link setLinks and confidence values corresponding to the linksRetained link after combining and a confidence value corresponding to the link
G L11(P11, P21)=0.8 Retaining L11(P11, P21)=0.9 after combining
L21(P11, P21)=0.5
L12(P22, P32)=0.8 Retaining L12(P22, P32)=1.0 after combining
L22(P22, P32)=0.5
L31(P22, P32)=1.0
L13(P33, P43)=0.8 Retaining L13(P33, P43)=1.0 after combining
L32(P33, P43)=1.0
L23(P34, P44)=0.5 L23(P34, P44)=0.5
L24(P41, P51)=0.5 L24(P41, P51)=0.5
L25(P54, P14)=0.5 L25(P54, P14)=0.5
L33(P44, P54)=1.0 L33(P44, P54)=1.0


[0043] The same links are at least two links in which two ports included in one link are the same as those in any other link, for example, L12, L22 and L31 in the foregoing Table 4. By using an example in which links L12, L22 and L31 are combined, the following describes how a confidence value of the retained L12 after combining is calculated according to the uncertainty reasoning algorithm.

[0044] The uncertainty reasoning algorithm in this embodiment of the present invention is based on a confidence value, and the uncertainty reasoning algorithm includes the following:

where CFi(H) is a confidence value among multiple confidence values of the same links, CFj(H) is another confidence value among the multiple confidence values of the same links, and CFi,j(H) is a new confidence value of the same links that is calculated according to CFi(H) and CFj(H).

[0045] The following describes, in detail, specific meaning of the foregoing formula:
when CFi(H)≥0 and CFj(H)≥0, the following formula is used:

when CFi(H)<0 and CFj(H)<0, the following formula is used:

when CFi(H) and CFj(H) have opposite signs, that is, CFi(H)≥0 and CFj(H)<0, or CFj(H)≥0 and CFi(H)<0, the following formula is used:



[0046] For same links that have two or more confidence values, when the confidence value of the retained link after combining is calculated, two confidence values are first selected as input of the foregoing formulas to obtain a new confidence value, and then the new confidence value and a confidence value that does not participate in the calculation are used as input of the foregoing formulas to obtain an updated confidence value. The foregoing steps are repeated until all confidence values of the links participate in the calculation of the formulas once, so as to finally obtain the confidence value of the retained link after combining.

[0047] For example, the confidence value of the retained link L12 after combining is calculated.

[0048] First, confidence values 0.8 and 0.5 of L12 and L22 are selected as input of the foregoing formulas, that is, 0.8+0.5-0.8×0.5=0.9; then, 0.9 and a confidence value 1.0 of L31 are used as input of the foregoing formulas, that is, 0.9+1.0-0.9×1.0=1.0; and finally, 1.0 is used as the confidence value of the retained link L12 after combining.

[0049] It should be noted that, a method for calculating a confidence value of a retained link after combining is not limited to the foregoing calculation formulas, and persons skilled in the art may further perform calculation by using another calculation method. For example, a weighted sum algorithm may be used. By using L12(P22, P32), L22(P22, P32), and L31(P22, P32) as an example, weighting coefficients of the confidence value of the IP address matching algorithm, of the confidence value of the port alias matching algorithm, and of the confidence value of the port LLDP link algorithm are respectively set to 0.2, 0.3, and 0.6, and the confidence value of the retained link L12 after combining is calculated: 0.8×0.2+0.5×0.3+1.0×0.6=0.91. A method for calculating a comprehensive confidence value of a link according to multiple confidence values of the same links is not limited in this embodiment of the present invention, and persons of ordinary skill in the art may use the calculation method provided in this embodiment of the present invention, or may use another calculation method.

[0050] Step 204. For at least two links having only one same port in the first link set, retain a link having a largest confidence value in the at least two links and delete a remaining link.

[0051] Exemplarily, refer to the following Table 5.
Table 5
Link setLinks and confidence values corresponding to the linksLinks and confidence values corresponding to the links
G L11(P11, P21)=0.9 L11(P11, P21)=0.9
  L12(P22, P32)=1.0 L12(P22, P32)=1.0
  L13(P33, P43)=1.0 L13(P33, P43)=1.0
  L23(P34, P44)=0.5 L24(P41, P51)=0.5
  L24(P41, P51)=0.5 L33(P44, P54)=1.0
  L25(P54, P14)=0.5  
  L33(P44, P54)=1.0  


[0052] Referring to the foregoing Table 5 in which L23 and L33 have only one same port P44 and L25 and L33 have only one same port P54, first, L33 is retained and L23 is deleted from L23 and L33, and then L33 is retained and L25 is deleted from L25 and L33; or first, L33 is retained and L25 is deleted from L25 and L33, and then L33 is retained and L23 is deleted from L23 and L33, where a sequence is not limited herein provided that selection is performed according to a principle of retaining a link having a largest confidence value among multiple links that have only one same port.

[0053] Step 205. Compare the confidence values of the links in the first link set with a preset threshold, and select a link whose confidence value is greater than the preset threshold, so as to obtain a second link set.

[0054] Exemplarily, the confidence values, of all links in the first link set, obtained according to step 203 and step 204 are compared with a preset threshold 0.7, and a link whose confidence value is greater than 0.7 is selected, so as to obtain a second link set G'. Refer to the following Table 6.
Table 6
Link setLinks and confidence values corresponding to the links
G' L11(P11, P21)=0.9
  L12(P22, P32)=1.0
  L13(P33, P43)=1.0
  L33(P44, P54)=1.0


[0055] Step 206. Obtain a network topology of the to-be-analyzed network according to each link in the second link set.

[0056] Exemplarily, the network topology of the to-be-analyzed network can be obtained according to all links L11(P11, P21), L12(P22, P32), L13(P33, P43), and L33(P44, P54) in the second link set G': the port P11 of the network element N1 is connected to the port 21 of the network element N2, the port P22 of the network element N2 is connected to the port P32 of the network element N3, the port P33 of the network element N3 is connected to the port P43 of the network element N4, and the port P44 of the network element N4 is connected to the port P54 of the network element N5. It can be seen that the network topology of the to-be-analyzed network that is obtained according to the technical solution provided in the foregoing embodiment is consistent with the network topology of the to-be-analyzed network shown in FIG. 2.

[0057] According to the network topology discovery method provided in this embodiment of the present invention, first, network characteristic data of all network elements in a to-be-analyzed network is collected; next, at least two corresponding link subsets are obtained respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and all links in the at least two link subsets are gathered into one set to obtain a first link set; then, a second link set is obtained by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link; and last, a network topology of the to-be-analyzed network is obtained according to each link in the second link set. In this way, a comprehensive analysis may be performed on results obtained after network topology discovery is performed by using multiple types of network characteristic data, so as to improve accuracy of network topology discovery.

[0058] An embodiment of the present invention provides a network topology discovery device 00. As shown in FIG. 4, the device 00 includes:

a collection unit 10, configured to collect network characteristic data of all network elements in a to-be-analyzed network;

a link obtaining unit 20, configured to obtain at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gather all links in the at least two link subsets into one set to obtain a first link set, where a confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to each link subset, confidence values of different topology discovery algorithms are different, and the link is a link that consists of two ports of different network elements;

a link processing unit 30, configured to obtain a second link set by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link; and

a topology obtaining unit 40, configured to obtain a network topology of the to-be-analyzed network according to each link in the second link set.



[0059] Optionally, the link processing unit 30 is specifically configured to:
obtain the second link set by performing an operation on the first link set, where the operation includes: combining same links in the first link set, calculating, according to multiple confidence values of the same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining, and for the at least two links having only one same port, retaining the link having the largest confidence value in the at least two links and deleting the remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link.

[0060] Optionally, the link processing unit 30 may be further specifically configured to:
obtain the second link set by performing an operation on the first link set, where the operation includes: combining same links in the first link set, calculating, according to multiple confidence values of the same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining, and for the at least two links having only one same port, retaining the link having the largest confidence value in the at least two links, deleting the remaining link, comparing the confidence values of the links in the first link set with a preset threshold, and selecting a link whose confidence value is greater than the preset threshold, where the same links are at least two links in which two ports included in one link are the same as those in any other link.

[0061] Optionally, the uncertainty reasoning algorithm includes the following:

where CFi (H) is a confidence value among multiple confidence values of the same links, CFj (H) is another confidence value among the multiple confidence values of the same links, and CFi,j (H) is a new confidence value of the same links that is calculated according to CFi (H) and CFj (H).

[0062] Optionally, the network characteristic data and the corresponding topology discovery algorithm include at least two types of the following combinations: a port Internet Protocol IP address and an Internet Protocol IP address matching algorithm, a port alias and a port alias matching algorithm, or port Link Layer Discovery Protocol LLDP neighbor information and a port Link Layer Discovery Protocol LLDP link algorithm.

[0063] This embodiment is used to implement the foregoing method embodiments; for working procedures and working principles of all units in this embodiment, reference may be made to descriptions in the foregoing method embodiments and details are not described herein.

[0064] According to the network topology discovery device provided in this embodiment of the present invention, first, network characteristic data of all network elements in a to-be-analyzed network is collected; next, at least two corresponding link subsets are obtained respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and all links in the at least two link subsets are gathered into one set to obtain a first link set; then, a second link set is obtained by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link; and last, a network topology of the to-be-analyzed network is obtained according to each link in the second link set. In this way, a comprehensive analysis may be performed on results obtained after network topology discovery is performed by using multiple types of network characteristic data, so as to improve accuracy of network topology discovery.

[0065] An embodiment of the present invention further provides a network topology discovery device 90. As shown in FIG. 5, the device 90 includes: a bus 94 and a processor 91, a memory 92, and an interface 93 that are connected to the bus 94, where the interface 93 is configured to communicate, the memory 92 is configured to store an instruction, and the processor 91 is configured to execute the instruction to:

collect network characteristic data of all network elements in a to-be-analyzed network;

obtain at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gather all links in the at least two link subsets into one set to obtain a first link set, where a confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to the link subset, confidence values of different topology discovery algorithms are different, and the link is a link that consists of two ports of different network elements;

obtain a second link set by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link; and

obtain a network topology of the to-be-analyzed network according to each link in the second link set.



[0066] Optionally, that the processor 91 executes the instruction to combine same links may specifically include:
combining same links in the first link set, and calculating, according to multiple confidence values of the same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining.

[0067] Optionally, the processor 91 executes the instruction to obtain the second link set by performing the operation on the first link set, and the operation further includes: after retaining a link having a largest confidence value in the at least two links and deleting a remaining link,
comparing the confidence values of the links in the first link set with a preset threshold, and selecting a link whose confidence value is greater than the preset threshold.

[0068] Optionally, the uncertainty reasoning algorithm includes the following:

where CFi (H) is a confidence value among multiple confidence values of the same links, CFj (H) is another confidence value among the multiple confidence values of the same links, and CFi,j (H) is a new confidence value of the same links that is calculated according to CFi (H) and CFj (H).

[0069] Optionally, the network characteristic data and the corresponding topology discovery algorithm include at least two types of the following combinations: a port Internet Protocol (IP) address and an Internet Protocol (IP) address matching algorithm, a port alias and a port alias matching algorithm, or port Link Layer Discovery Protocol (LLDP) neighbor information and a port Link Layer Discovery Protocol (LLDP) link algorithm.

[0070] This embodiment is used to implement the foregoing method embodiments; for working procedures and working principles of all units in this embodiment, reference may be made to descriptions in the foregoing method embodiments and details are not described herein.

[0071] According to the network topology discovery device provided in this embodiment of the present invention, first, network characteristic data of all network elements in a to-be-analyzed network is collected; next, at least two corresponding link subsets are obtained respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and all links in the at least two link subsets are gathered into one set to obtain a first link set; then, a second link set is obtained by performing an operation on the first link set, where the operation includes: combining same links, and for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link, where the same links are at least two links in which two ports included in one link are the same as those in any other link; and last, a network topology of the to-be-analyzed network is obtained according to each link in the second link set. In this way, a comprehensive analysis may be performed on results obtained after network topology discovery is performed by using multiple types of network characteristic data, so as to improve accuracy of network topology discovery.

[0072] Persons of ordinary skill in the art may understand that all or some of the steps of the method embodiments may be implemented by a program instructing relevant hardware. The program may be stored in a computer-readable storage medium. When the program runs, the steps of the method embodiments are performed. The foregoing storage medium includes: any medium that can store program code, such as a ROM, a RAM, a magnetic disk, or an optical disc.

[0073] Finally, it should be noted that the foregoing embodiments are merely intended for describing the technical solutions of the present invention, but not for limiting the present invention. Although the present invention is described in detail with reference to the foregoing embodiments, persons of ordinary skill in the art should understand that they may still make modifications to the technical solutions described in the foregoing embodiments or make equivalent replacements to some or all technical features thereof, without departing from the scope of the technical solutions of the embodiments of the present invention.


Claims

1. A network topology discovery method, comprising:

collecting (101, 201) network characteristic data of all network elements in a to-be-analyzed network;

obtaining (102, 202) at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gathering (102) all links in the at least two link subsets into one set to obtain a first link set, wherein a confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to the link subset, confidence values of different topology discovery algorithms are different, and the link is used to connect two ports of different network elements;

obtaining (103) a second link set by performing an operation on the first link set, wherein the operation comprises:

combining (203) same links in the first link set, and calculating, according to multiple confidence values of the same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining, wherein the same links are at least two links in which two ports in one link are the same as those in any other link; and

for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link; and

obtaining (104, 206) a network topology of the to-be-analyzed network according to each link in the second link set.


 
2. The method according to claim 1, wherein the operation further comprises: after the retaining a link having a largest confidence value in the at least two links and deleting (204) a remaining link,
comparing (205) the confidence values of the links in the first link set with a preset threshold, and selecting a link whose confidence value is greater than the preset threshold.
 
3. The method according to claim 1, wherein the uncertainty reasoning algorithm comprises the following:

wherein CFi (H) is a confidence value among multiple confidence values of the same links, CFj (H) is another confidence value among the multiple confidence values of the same links, and CFi,j (H) is a new confidence value of the same links that is calculated according to CFi (H) and CFj (H).
 
4. The method according to any one of claims 1 to 3, wherein the network characteristic data and the corresponding topology discovery algorithm comprise at least two types of the following combinations: a port Internet Protocol, IP, address and an Internet Protocol, IP, address matching algorithm, a port alias and a port alias matching algorithm, or port Link Layer Discovery Protocol, LLDP, neighbor information and a port Link Layer Discovery Protocol, LLDP, link algorithm.
 
5. A network topology discovery device (00), comprising:

a collection unit (10), configured to collect network characteristic data of all network elements in a to-be-analyzed network;

a link obtaining unit (20), configured to obtain at least two corresponding link subsets respectively by using at least two types of topology discovery algorithms and according to the network characteristic data, and gather all links in the at least two link subsets into one set to obtain a first link set, wherein a confidence value of a link in each link subset is equal to a confidence value of a topology discovery algorithm corresponding to the link subset, confidence values of different topology discovery algorithms are different, and the link is used to connect two ports of different network elements;

a link processing unit (30), configured to obtain a second link set by performing an operation on the first link set, wherein the operation comprises:

combining same links in the first link set, calculating, according to multiple confidence values of the same links and an uncertainty reasoning algorithm, a confidence value of the retained link after combining, wherein the same links are at least two links in which two ports comprised in one link are the same as those in any other link; and

for at least two links having only one same port, retaining a link having a largest confidence value in the at least two links and deleting a remaining link; and

a topology obtaining unit (40), configured to obtain a network topology of the to-be-analyzed network according to each link in the second link set.


 
6. The device (00) according to claim 5, wherein the link processing unit (20) is further configured to:
after the retaining a link having a largest confidence value in the at least two links and deleting a remaining link, comparing the confidence values of the links in the first link set with a preset threshold, and selecting a link whose confidence value is greater than the preset threshold.
 
7. The device (00) according to claim 5, wherein the uncertainty reasoning algorithm comprises the following:

wherein CFi (H) is a confidence value among multiple confidence values of the same links, CFj (H) is another confidence value among the multiple confidence values of the same links, and CFi,j (H) is a new confidence value of the same links that is calculated according to CFi (H) and CFj (H).
 
8. The device (00) according to any one of claims 5 to 7, wherein the network characteristic data and the corresponding topology discovery algorithm comprise at least two types of the following combinations: a port Internet Protocol, IP, address and an Internet Protocol, IP, address matching algorithm, a port alias and a port alias matching algorithm, or port Link Layer Discovery Protocol, LLDP, neighbor information and a port Link Layer Discovery Protocol, LLDP, link algorithm.
 


Ansprüche

1. Verfahren zur Entdeckung einer Netzwerktopologie, umfassend:

Erfassen (101, 201) von netzwerkcharakteristischen Daten aller Netzwerkelemente in einem zu analysierenden Netzwerk;

Erhalten (102, 202) mindestens zweier entsprechender Verbindungsuntersätze jeweils mithilfe mindestens zweier Typen von Topologieentdeckungsalgorythmen und entsprechend den netzwerkcharakteristischen Daten, und Erfassen (102) aller Verbindungen in den mindestens zwei Verbindungsuntersätzen in einem Satz, um einen ersten Verbindungssatz zu erhalten, wobei ein Konfidenzwert einer Verbindung in jedem Verbindungsuntersatz gleich einem Konfidenzwert eines Topologieentdeckungsalgorithmus ist, der dem Verbindungsuntersatz entspricht, Konfidenzwerte unterschiedlicher Topologieentdeckungsalgorythmen unterschiedlich sind und die Verbindung genutzt wird, um zwei Ports unterschiedlicher Netzwerkelemente zu verbinden;

Erhalten (103) eines zweiten Verbindungssatzes durch Vornehmen einer Operation an dem ersten Verbindungssatz, wobei die Operation Folgendes umfasst:

Kombinieren (203) gleicher Verbindungen in dem ersten Verbindungssatz und Berechnen, entsprechend mehreren Konfidenzwerten der gleichen Verbindungen und einem Unsicherheitsbegründungsalgorithmus, eines Konfidenzwerts der beibehaltenen Verbindung nach dem Kombinieren, wobei die gleichen Verbindungen mindestens zwei Verbindungen sind, bei denen zwei Ports in einer Verbindung die gleichen wie jene in einer beliebigen anderen Verbindung sind; und

für mindestens zwei Verbindungen, die nur einen gleichen Port aufweisen, Beibehalten einer Verbindung mit einem größten Konfidenzwert in den mindestens zwei Verbindungen und Löschen einer übrigen Verbindung; und

Erhalten (104, 206) einer Netzwerktopologie des zu analysierenden Netzwerks entsprechend jeder Verbindung in dem zweiten Verbindungssatz.


 
2. Verfahren nach Anspruch 1, wobei die Operation ferner Folgendes umfasst: nach dem Beibehalten einer Verbindung mit einem größten Konfidenzwert in den mindestens zwei Verbindungen und Löschen (204) einer übrigen Verbindung, Vergleichen (205) der Konfidenzwerte der Verbindungen in dem ersten Verbindungssatz mit einer voreingestellten Schwelle und Auswählen einer Verbindung, deren Konfidenzwert größer als die voreingestellte Schwelle ist.
 
3. Verfahren nach Anspruch 1, wobei der Unsicherheitsbegründungsalgorithmus Folgendes umfasst:

wobei CFi (H) ein Konfidenzwert von mehreren Konfidenzwerten der gleichen Verbindungen ist, CFj (H) ein weiterer Konfidenzwert von den mehreren Konfidenzwerten der gleichen Verbindungen ist, und CFi,j (H) ein neuer Konfidenzwert der gleichen Verbindungen ist, der entsprechend CFi (H) und CFj (H) berechnet wird.
 
4. Verfahren nach einem der Ansprüche 1 bis 3, wobei die netzwerkcharakteristischen Daten und der entsprechende Topologieentdeckungsalgorithmus mindestens zwei Typen der folgenden Kombinationen umfassen: eine Port-Internetprotokoll-, IP-, Adresse und einen Internetprotokoll-, IP-, Adressen-Abgleichungsalgorithmus, einen Port-Alias und einen Port-Alias-Abgleichungsalgorithmus oder Port-"Link-Layer-Discovery-Protocol-", LLDP-, Nachbarinformationen und einen Port-"Link-Layer-Discovery-Protocol-", LLDP-, Verbindungsalgorithmus.
 
5. Vorrichtung (00) zur Entdeckung einer Netzwerktopologie, umfassend:

eine Erfassungseinheit (10), die ausgelegt ist, netzwerkcharakteristische Daten aller Netzwerkelemente in einem zu analysierenden Netzwerk zu erfassen;

eine Einheit (20) zum Erhalten einer Verbindung, die ausgelegt ist, mindestens zwei entsprechende Verbindungsuntersätze jeweils mithilfe mindestens zweier Typen von Topologieentdeckungsalgorythmen und entsprechend den netzwerkcharakteristischen Daten zu erhalten, und alle Verbindungen in den mindestens zwei Verbindungsuntersätzen in einem Satz zu erfassen, um einen ersten Verbindungssatz zu erhalten, wobei ein Konfidenzwert einer Verbindung in jedem Verbindungsuntersatz gleich einem Konfidenzwert eines Topologieentdeckungsalgorithmus ist, der dem Verbindungsuntersatz entspricht, Konfidenzwerte unterschiedlicher Topologieentdeckungsalgorythmen unterschiedlich sind und die Verbindung genutzt wird, um zwei Ports unterschiedlicher Netzwerkelemente zu verbinden;

eine Verbindungs-Verarbeitungseinheit (30), die ausgelegt ist, einen zweiten Verbindungssatz durch Vornehmen einer Operation an dem ersten Verbindungssatz zu erhalten, wobei die Operation Folgendes umfasst:

Kombinieren gleicher Verbindungen in dem ersten Verbindungssatz, Berechnen, entsprechend mehreren Konfidenzwerten der gleichen Verbindungen und einem Unsicherheitsbegründungsalgorithmus, eines Konfidenzwerts der beibehaltenen Verbindung nach dem Kombinieren, wobei die gleichen Verbindungen mindestens zwei Verbindungen sind, bei denen zwei Ports, die in einer Verbindung enthalten sind, die gleichen wie jene in einer beliebigen anderen Verbindung sind; und

für mindestens zwei Verbindungen, die nur einen gleichen Anschluss aufweisen, Beibehalten einer Verbindung mit einem größten Konfidenzwert in den mindestens zwei Verbindungen und Löschen einer übrigen Verbindung; und

eine Topologieerhaltungseinheit (40), die ausgelegt ist, eine Netzwerktopologie des zu analysierenden Netzwerks entsprechend jeder Verbindung in dem zweiten Verbindungssatz zu erhalten.


 
6. Vorrichtung (00) nach Anspruch 5, wobei die Verbindungs-Verarbeitungseinheit (20) ferner konfiguriert ist:
nach dem Beibehalten einer Verbindung mit einem größten Konfidenzwert in den mindestens zwei Verbindungen und dem Löschen einer übrigen Verbindung, Vergleichen der Konfidenzwerte der Verbindungen in dem ersten Verbindungssatz mit einer voreingestellten Schwelle und Auswählen einer Verbindung, deren Konfidenzwert größer als die voreingestellte Schwelle ist.
 
7. Vorrichtung (00) nach Anspruch 5, wobei der Unsicherheitsbegründungsalgorithmus Folgendes umfasst:

wobei CFi (H) ein Konfidenzwert von mehreren Konfidenzwerten der gleichen Verbindungen ist, CFj (H) ein weiterer Konfidenzwert von den mehreren Konfidenzwerten der gleichen Verbindungen ist, und CFi,j (H) ein neuer Konfidenzwert der gleichen Verbindungen ist, der entsprechend CFi (H) und CFj (H) berechnet wird.
 
8. Vorrichtung (00) nach einem der Ansprüche 5 bis 7, wobei die netzwerkcharakteristischen Daten und der entsprechende Topologieentdeckungsalgorithmus mindestens zwei Typen der folgenden Kombinationen umfassen: eine Port-Internetprotokoll-, IP-, Adresse und einen Internetprotokoll-, IP-, Adressen-Abgleichungsalgorithmus, einen Port-Alias und einen Port-Alias-Abgleichungsalgorithmus oder Port-"Link-Layer-Discovery-Protocol-", LLDP-, Nachbarinformationen und einen Port-"Link-Layer-Discovery-Protocol-", LLDP-, Verbindungsalgorithmus.
 


Revendications

1. Procédé de découverte de la topologie de réseau, comprenant les étapes consistant à :

recueillir (101, 201) des données caractérisque du réseau pour tous les éléments de réseau dans un réseau à analyser ;

obtenir (102, 202) respectivement au moins deux sous-ensembles de liaisons correspondants en utilisant au moins deux types d'algorithmes de découverte de topologie et en fonction des données caractérisque de réseau, et rassembler (102) toutes les liaisons dans les au moins deux sous-ensembles de liaisons en un ensemble pour obtenir un premier ensemble de liaisons, dans lequel une valeur de confiance d'une liaison dans chaque sous-ensemble de liaisons est égale à une valeur de confiance d'un algorithme de découverte de topologie correspondant au sous-ensemble de liaisons, les valeurs de confiance de différents algorithmes de découverte de topologie sont différentes, et la liaison est utilisée pour connecter deux ports d'éléments de réseau différents;

obtenir (103) un second ensemble de liaisons en effectuant une opération sur le premier ensemble de liaisons, dans lequel l'opération comprend les étapes consistant à :

combiner (203) les mêmes liaisons dans le premier ensemble de liaisons et calculer, en fonction de multiples valeurs de confiance des mêmes liaisons et d'un algorithme de raisonnement sur l'incertitude, une valeur de confiance de la liaison conservée après la combinaison, dans lequel les mêmes liaisons sont au moins deux liaisons dans lesquelles deux ports dans une liaison sont identiques à ceux dans toute autre liaison ; et

pour au moins deux liaisons n'ayant qu'un seul et même port, conserver une liaison ayant la plus grande valeur de confiance dans les au moins deux liaisons et supprimer une liaison restante ; et

obtenir (104, 206) une topologie de réseau du réseau à analyser en fonction de chaque liaison du second ensemble de liaisons.


 
2. Procédé selon la revendication 1, dans lequel l'opération comprend en outre les étapes consistant à : après avoir conservé une liaison ayant la plus grande valeur de confiance parmi les au moins deux liaisons et supprimé (204) une liaison restante, comparer (205) les valeurs de confiance des liaisons dans le premier ensemble de liaisons à un seuil prédéfini et sélectionner une liaison dont la valeur de confiance est supérieure au seuil prédéfini.
 
3. Procédé selon la revendication 1, dans lequel l'algorithme de raisonnement sur l'incertitude comprend les éléments suivants :

dans lequel CFi (H) est une valeur de confiance parmi plusieurs valeurs de confiance des mêmes liaisons, CFj (H) est une autre valeur de confiance parmi les multiples valeurs de confiance des mêmes liaisons, et CFi, j (H) est une nouvelle valeur de confiance des mêmes liaisons qui est calculée en fonction de CFi (H) et CFj (H).
 
4. Procédé selon l'une quelconque des revendications 1 à 3, dans lequel les données caractérisque de réseau et l'algorithme de découverte de topologie correspondant comprennent au moins deux types des combinaisons suivantes : une adresse de protocole Internet, IP, de port et un algorithme de correspondance d'adresse de un protocole Internet, IP, un alias de port et un algorithme de correspondance d'alias de port, ou des informations de voisinage de protocole de découverte de couche de liaison, LLDP (Link Layer Discovery Protocol), de port et un algorithme de liaison de protocole de découverte de couche de liaison, LLDP, de port.
 
5. Dispositif de découverte de topologie de réseau (00), comprenant :

une unité de collecte (10), configurée pour recueillir des données caractéristiques de réseau de tous les éléments de réseau dans un réseau à analyser ;

une unité d'obtention de liaisons (20), configurée pour obtenir respectivement au moins deux sous-ensembles de liaisons correspondants en utilisant au moins deux types d'algorithmes de découverte de topologie et en fonction des données caractérisque de réseau, et rassembler toutes les liaisons dans les au moins deux sous-ensembles de liaisons en un seul ensemble pour obtenir un premier ensemble de liaisons, dans lequel une valeur de confiance d'une liaison dans chaque sous-ensemble de liaisons est égale à une valeur de confiance d'un algorithme de découverte de topologie correspondant au sous-ensemble de liaisons, les valeurs de confiance de différents algorithmes de découverte de topologie sont différentes, et la liaison est utilisée pour connecter deux ports d'éléments de réseau différents;

une unité de traitement de liaison (30), configurée pour obtenir un second ensemble de liaisons en effectuant une opération sur le premier ensemble de liaisons, dans lequel l'opération comprend les étapes consistant à :

combiner les mêmes liaisons dans le premier ensemble de liaisons, calculer, en fonction de plusieurs valeurs de confiance des mêmes liaisons et d'un algorithme de raisonnement sur l'incertitude, une valeur de confiance de la liaison conservée après la combinaison, dans lequel les mêmes liaisons sont au moins deux liaisons dans lesquelles deux ports compris dans une liaison sont identiques à ceux dans toute autre liaison ; et

pour au moins deux liaisons n'ayant qu'un seul et même port, conserver une liaison ayant la plus grande valeur de confiance dans les au moins deux liaisons et supprimer une liaison restante ; et

une unité d'obtention de topologie (40), configurée pour obtenir une topologie de réseau du réseau à analyser, en fonction de chaque liaison dans le second ensemble de liaisons.


 
6. Dispositif (00) selon la revendication 5, dans lequel l'unité de traitement de liaison (20) est en outre configurée pour :
après avoir conservé une liaison ayant la plus grande valeur de confiance parmi les au moins deux liaisons et supprimé une liaison restante, comparer les valeurs de confiance des liaisons dans le premier ensemble de liaisons à un seuil prédéfini et sélectionner une liaison dont la valeur de confiance est supérieure à un seuil prédéfini.
 
7. Dispositif (00) selon la revendication 5, dans lequel l'algorithme de raisonnement sur l'incertitude comprend les éléments suivants :

dans lequel CFi (H) est une valeur de confiance parmi plusieurs valeurs de confiance des mêmes liaisons, CFj (H) est une autre valeur de confiance parmi les multiples valeurs de confiance des mêmes liaisons, et CFi, j (H) est une nouvelle valeur de confiance des mêmes liaisons qui est calculée en fonction de CFi (H) et CFj (H).
 
8. Dispositif (00) selon l'une quelconque des revendications 5 à 7, dans lequel les données caractérisque de réseau et l'algorithme de découverte de topologie correspondant comprennent au moins deux types des combinaisons suivantes : une adresse de protocole Internet, IP, de port et un algorithme de correspondance d'adresse de protocole Internet, IP, un alias de port et un algorithme de correspondance d'alias de port, ou des informations de voisinage de protocole de découverte de couche de liaison, LLDP, de port et un algorithme de liaison de protocole de découverte de couche de liaison, LLDP de port.
 




Drawing

















Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description