CN108833609B - Local network destination IP address selection method based on historical topology measurement data - Google Patents

Local network destination IP address selection method based on historical topology measurement data Download PDF

Info

Publication number
CN108833609B
CN108833609B CN201810666495.XA CN201810666495A CN108833609B CN 108833609 B CN108833609 B CN 108833609B CN 201810666495 A CN201810666495 A CN 201810666495A CN 108833609 B CN108833609 B CN 108833609B
Authority
CN
China
Prior art keywords
local network
node
nodes
super
address
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810666495.XA
Other languages
Chinese (zh)
Other versions
CN108833609A (en
Inventor
张宇
余卓勋
张晔
张宏莉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Institute of Technology
Original Assignee
Harbin Institute of Technology
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 Harbin Institute of Technology filed Critical Harbin Institute of Technology
Priority to CN201810666495.XA priority Critical patent/CN108833609B/en
Publication of CN108833609A publication Critical patent/CN108833609A/en
Application granted granted Critical
Publication of CN108833609B publication Critical patent/CN108833609B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/50—Address allocation
    • H04L61/5007—Internet protocol [IP] addresses
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/50—Address allocation
    • H04L61/5046—Resolving address allocation conflicts; Testing of addresses

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Cardiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

A local network destination IP address selection method based on historical topology measurement data belongs to the technical field of network topology measurement. The connection relation of the local network IP addresses is analyzed by utilizing a large amount of historical topological data of the local network, the generation of a target IP address set is guided, the measuring efficiency of the local network is improved, and a key topological structure of the local network is found. The method comprises the steps of obtaining an IP address topological relation of a local network by utilizing historical topological measurement data, analyzing a connection relation between IP addresses in the local network according to relevant knowledge of graph theory, dividing the IP addresses in the local network into a large number of IP clusters, selecting a representative IP address from each IP cluster to add into a destination IP address set, executing topological measurement by using a clicking Glass server, and analyzing the capability of the local network topological measurement data in the discovery work of the local network topological data. The experimental result shows that the target IP address set generated by the invention has stronger discovery capability aiming at the measurement work of the local network.

Description

Local network destination IP address selection method based on historical topology measurement data
Technical Field
The invention relates to a local network destination IP address selection method, belonging to the technical field of network topology measurement.
Background
When a Looking Glass server is used for carrying out traceroute measurement, fine-grained measurement cannot be usually carried out on a local network of ten million levels of IP addresses in a short time, in a traditional target IP address set, IP address segments are divided into the same size by obtaining the IP address segments of the local network, and an IP address is randomly generated from each IP address segment to generate the target IP address set. The method does not consider the actual connection relation of the IP addresses in the local network, and in the actual measurement work, the topological distance between the destination IP addresses is expected to be selected to be longer. The Ark platform of CAIDA can always perform fine-grained measurement on the networks all over the world, and can utilize the measurement data to analyze the topological relation among the IP addresses of local networks, help to select a destination IP address and guide the measurement work of a logging Glass server. The traditional target IP address selection method has the defects of low topology measurement efficiency and weak capability of discovering a key topology structure of a local network.
Disclosure of Invention
The invention provides a local network destination IP address selection method based on historical topological measurement data, which analyzes the connection relation of local network IP addresses by using a large amount of historical topological data of local networks and guides the generation of a destination IP address set so as to improve the measurement efficiency aiming at the local networks and discover a key topological structure of the local networks.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a local network destination IP address selection method based on historical topology measurement data is realized by the following steps:
(1) acquiring historical topological data of a local network, and generating a local network topological graph according to the historical topological data;
(2) defining a super node (IP cluster) based on a local network topology map;
(3) dividing the local network topological graph into a plurality of super nodes according to the super nodes defined in the step (2), and generating a target IP address set by using the plurality of super nodes;
(4) and measuring the local network by using the Looking Glass as a measuring point and the generated target IP address set as a target point to obtain local network topology measuring data.
Further, the process of acquiring the historical topology data of the local network comprises the following steps:
historical topology data is downloaded from a third party platform (such as the Ark platform of CAIDA), and a traceroute path with a destination IP address located inside a local network is selected as the local network historical topology data.
Further, the process of generating the local network topology map according to the historical topology data comprises the following steps:
selecting a first IP address in each hop TTL of a traceroute path as a node of the TTL, generating a plurality of undirected edges according to the order of the TTL, and aggregating the edges of at least one node (IP address) positioned in a local network to form an undirected graph G which is recorded as a topological relation graph G of the local network.
Further, for a certain node set in the local network topology map
Figure GDA0003069313670000023
If it is not
Figure GDA0003069313670000024
Is a number v1Is a tree of root nodes, and
Figure GDA0003069313670000022
node of (3) through v1Connected with other nodes in the local network topological graph, the node set is called
Figure GDA0003069313670000025
Is a super node, called v1Is a super node
Figure GDA0003069313670000021
The other nodes are common nodes.
Further, the algorithm for partitioning the local network topology map into a plurality of super nodes according to the super nodes defined in the step (2) mainly comprises the following steps:
(a) traversing all leaf nodes (the node degree is 1), if no leaf node exists, ending the algorithm, and taking the father node of the leaf nodes as a root node;
(b) if the father node is not in the root node set of the super nodes, declaring a new super node taking the father node as the root node;
(c) if the leaf node is the root node of a certain super node, merging the super node corresponding to the leaf node into the super node corresponding to the parent node of the leaf node, and if the leaf node is not the root node of the super node, directly adding the leaf node into the super node corresponding to the parent node of the leaf node;
(d) and (c) removing all leaf nodes and corresponding edges thereof, and entering the step (a).
After settlement by the algorithm, the super node set reserved in (c) is a plurality of super nodes divided by the local network topological graph.
Further, the generation of the destination IP address set by using a plurality of super nodes means that one or a plurality of surviving IP addresses are randomly selected from each super node to form the destination IP address set, and the surviving IP addresses are IP addresses that can be reached by using a ping test.
Further, in the step (4), the Looking Glass is used as a measuring point, the generated destination IP address set is used as a target point, the two are subjected to cartesian product to form a local network topology measuring task, and the local network is measured according to the measuring task.
The invention has the beneficial effects that:
the method comprises the steps of obtaining an IP address topological relation of a local network by using historical topological measurement data, analyzing a connection relation between IP addresses in the local network according to relevant knowledge of graph theory, dividing the IP addresses in the local network into a large number of IP clusters, selecting a representative IP address from each IP cluster to be added into a destination IP address set, performing topological measurement by using a clicking Glass server, and analyzing the capability of the topological measurement in the work of discovering the local network topological data. The experimental result shows that the target IP address set generated by the invention has stronger discovery capability aiming at the measurement work of the local network, and the discovery capability is shown in the table 1.
Evaluating the capability of the selected target IP address set for finding topological data in the local network measurement work, selecting available measurement points of the Looking Glass part as a measurement server, and selecting the topological data of the local network to generate the target IP address set; the method comprises the steps of obtaining IP address sections of a local network from a geographic positioning database (IP2location), dividing all the IP address sections into the same number of IP address sections according to the number of super nodes generated by the algorithm, randomly selecting one or a plurality of survival IP addresses (ping test) from each IP address section, finally generating a target IP address set with the same size, simultaneously measuring by using machines with the same configuration, and comparing topology discovery capability.
Drawings
FIG. 1 is a schematic diagram of a super node (vertex) of the present invention, and FIG. 2 is a schematic diagram of super node generation.
Detailed Description
The first embodiment is as follows: as shown in fig. 1 to fig. 2, an implementation process of a local network destination IP address selection method based on historical topology measurement data according to this embodiment is as follows:
(1) acquiring historical topological data of a local network, and generating a local network topological graph according to the historical topological data;
(2) defining a super node (IP cluster) based on a local network topology map; i.e. the definition of super nodes (IP clusters) based on graph theory knowledge;
(3) dividing the local network topological graph into a plurality of super nodes according to the super nodes defined in the step (2), and generating a target IP address set by using the plurality of super nodes;
(4) and measuring the local network by using the Looking Glass as a measuring point and the generated target IP address set as a target point to obtain local network topology measuring data.
The step (1) mainly comprises two parts, namely two parts for analyzing data basis of the relation between IP addresses in the local network, (1.1) acquisition of historical topology data, and (1.2) generation of a local network topology graph. The detailed description of the two parts is as follows:
(1.1) obtaining historical topology data: at present, some third party platforms (such as Ark platform of CAIDA) continuously measure the networks all over the world and the data set is public, and a part of a traceroute destination IP address located inside a local network (determined by using a geographic location database IP2location) is selected from the measurement results of the third party platforms to form historical measurement data of the local network;
(1.2) generating a local network topological graph: the method comprises the following steps that a measurement result obtained from a third-party platform is a traceroute, the traceroute is converted into an edge, and the specific method comprises the following steps: the traceroute path is a path, a first IP address in each hop of TTL is selected as a node of the TTL, a plurality of undirected edges are generated according to the order of the TTL, and edges with at least one node (IP address) positioned in a local network are aggregated together to form an undirected graph G which is recorded as a topological relation graph G of the local network.
The step (2) is mainly the definition of an IP cluster, analyzes a local network topological relation graph, aggregates closely-connected IP addresses together to form the IP cluster, records the IP cluster as a super node, and analyzes the definition of the super node and the feasibility of dividing the local network topological relation graph into a plurality of super nodes from the basic graph theory knowledge, and is described in detail as follows:
definition 1 for the undirected connected graph G (V, E), the nodes with the degree of 1 are called leaf nodes, and all the leaf nodes are removed to obtain a new undirected connected graph G1(V1,E1) Repeating the above steps n times to obtain a multidirectional connection graph Gn(Vn,En) Up to GnThere is no leaf node in the tree, and all the leaf nodes are called V1(V1＝V-Vn)。
Theorem 1 if GnIf there is no leaf node in the tree, GnAll the node degrees in (1) are greater than and V1A point in (1) must pass only VnWith one node in VnAre connected.
Evidence that a) does not set v1∈V1Knowing that G is a directed-less connectivity graph, v1Must pass at least VnWith one node in VnAre connected.
b) Is not provided with v1∈V1Let v be1Through VnNode with two or more nodes in VnWherein other nodes are connected without setting two nodes as v2,v3(v2,v3∈Vn) Then v1To v2There is a path which does not pass v3，v1To v3There is a path which does not pass v2And due to GnIs a connectivity graph, then v1,v2,v3Can form a ring, obviously impossible, so v1At most by VnWith one node in VnAre connected.
Definition 2 for a set of nodes
Figure GDA0003069313670000041
If it is not
Figure GDA0003069313670000042
Is a number v1A tree of root nodes is called a node set
Figure GDA0003069313670000044
Is a super node, called v1Is a super node
Figure GDA0003069313670000043
The other nodes are common nodes.
Theorem 2 without vn∈Vn，{v2,v3,...,vj|j>＝1}∈V1，{v2,v3,...,vjAll nodes in the lattice are through vnAnd VnAre connected, then the node v can be generatednAs root node, { v2,v3,...,vjSuper node with common node
Figure GDA0003069313670000045
Attestation a) provenance node v2Through node vnAnd V2＝V1-{v2,v3,...,vjThe nodes in the method are connected: if v is2At least one does not pass through vnRoad and V2Are connected, then v2Through V2And GnAre linked and not via vnAnd GnConnected, contradict theorem 1. In the same way { v2,v3,...,vjOther nodes in the tree are also by vnAnd V2Are connected.
b) As understood from definition 1, { v2,v3,...,vj,vnThere is no ring between them, so { v }2,v3,...,vj,vnCan be constructed with vnTheorem 2 holds true for the tree of the root node.
Theorem 3 node set V1All nodes in (1) and (V)nWherein part of the nodes form a plurality of super nodes, V1The node in (1) is a normal node, VnIs the root node.
Prove to V1The nodes in (1) are divided as follows: if V1All nodes are passing through VnA certain node in and VnIf other nodes are connected, the nodes are divided into the same node set, and theorem 2 shows that theorem 3 holds.
The super node structure defined in this step is shown in fig. 1.
Step (3) is mainly a destination IP address set generation algorithm, a plurality of super nodes are generated from the local network topological graph in step (1) according to the definition in step (2), and the main steps are as follows:
(a) traversing all leaf nodes (the node degree is 1), if no leaf node exists, ending the algorithm, and taking the father node of the leaf nodes as a root node;
(b) if the parent node is not in the root node set of supernodes, declaring a new supernode having the parent node as the root node
(c) If the leaf node is the root node of a certain super node, merging the super node corresponding to the leaf node into the super node corresponding to the parent node of the leaf node, and if the leaf node is not the root node of the super node, directly adding the leaf node into the super node corresponding to the parent node of the leaf node;
(d) and (c) removing all leaf nodes and corresponding edges thereof, and entering the step (a).
The super node generation process is shown in fig. 2 (1- >2- >3- >4 in the figure), and the algorithm pseudo code of the super node is as follows:
Figure GDA0003069313670000051
Figure GDA0003069313670000061
after the algorithm is finished, the result of the V _ Super is the division of the Super nodes, the key of each element in the V _ Super is a root node, and the value is a common node. If the generated graph G is not connected due to limited historical topological data, the generated graph G can be regarded as a plurality of undirected connected graphs, and the algorithm can still be executed. And finally, selecting one or a plurality of surviving IP (ping test) addresses from each super node to be added into the destination IP address set.
In the step (4), the Looking Glass is used as a measuring point, the generated destination IP address set is used as a target point, the two are subjected to Cartesian product to form a local network topology measuring task, and the local network is measured according to the measuring task.
Examples
For a better illustration of the invention, the inventive solution will be described in detail in connection with specific measurement work: the invention mainly utilizes the historical topological data of the local network to generate the target IP address set, thereby improving the measuring efficiency of the local network and the discovering ability of the topological structure of the local network. The invention mainly comprises four steps:
(1) acquiring historical topological data and generating a local network topological graph;
(2) the definition of super nodes (IP clusters) based on graph theory knowledge;
(3) generating algorithm of destination IP address set;
(4) and measuring the local network by using the Looking Glass, and evaluating the topology discovery capability of the invention.
The step (1) is mainly to generate a topological graph of the local network, which is used for analyzing the topological relation between the IP addresses of the local network, and comprises the following two parts:
(1.1) obtaining historical topology data: downloading Ark platform of CAIDA to measure topological data of the network all over the world in 2018, 2 months, and selecting measured data of a traceroute target IP address in a local network from the measurement results as an original data set;
(1.2) generating a local network topological graph: the measurement result obtained from the Ark platform is a traceroute path, and the traceroute path is converted into an edge firstly, and the specific method is as follows: selecting a first IP address in each hop TTL in traceroute as a node of the TTL, generating a plurality of undirected edges according to the order from small to large of the TTL, and aggregating the edges of at least one node (IP address) positioned in a local network to form an undirected graph G which is recorded as a topological relation graph G of the local network.
The step (2) mainly relates to the definition of the super nodes and the division feasibility of the super nodes, and the principle is the same as the content of the invention
Step (3) is a process of selecting a destination IP address set, about 10 ten thousand nodes exist in a topological relation graph G of a local network, and a detailed partitioning process of a super node is as follows:
(a) traverse all leaf nodes (node degree 1) with their parent nodes as root nodes (about 70% of the points in the first round are leaf nodes);
(b) if the father node is not in the root node set of the super nodes, generating a new super node taking the father node as the root node (the number of the super nodes in the first round is about 1 ten thousand);
(c) if the leaf node is the root node of a certain super node, merging the super node corresponding to the leaf node into the super node corresponding to the parent node of the leaf node, and if the leaf node is not the root node of the super node, directly adding the leaf node into the super node corresponding to the parent node of the leaf node (along with the increase of the turns, the super nodes are merged ceaselessly, and the number is reduced);
(d) and (3) removing all leaf nodes and corresponding edges thereof, entering the step (a), and executing 6 rounds of algorithms to obtain 3384 super nodes.
And (4) evaluating the topological measurement capability of the target IP address set generated in the invention, designing a comparison experiment to measure the local network, wherein the measurement points used in the comparison experiment are all from Looking Glass, the measurement points of different interfaces can execute measurement tasks concurrently, the available 1000 measurement points are from 121 different interfaces, in order to improve the measurement efficiency, the different interfaces execute the measurement tasks simultaneously, and the measurement points in the same interface are recycled. The method comprises the steps that (a) a surviving IP address is randomly selected from each super node and added into an IP address set, 3384 IP addresses are totally generated, a target IP address set is generated, (b) IP address segments are divided in an equal scale, 3384 surviving target IP addresses are randomly generated, and the target IP address set is generated. The measurement task is generated by the Cartesian product of the measurement point and the target IP address set, the measurement is executed, the measurement result is shown in a table 1 (the IP addresses of the local network: all the IP addresses positioned in the local network; the router IP addresses of the local network: the router IP addresses positioned in the local network; the edges of the local network: the edges at least one node of which is positioned in the local network; the IP addresses of the local network connected to the outside: the IP addresses of the local network directly connected with other networks), and the experimental result shows that the target IP address set generated by the method has stronger discovery capability for the measurement work of the local network, which is shown in the table 1.
TABLE 1
Figure GDA0003069313670000071

Claims (3)

1. A local network destination IP address selection method based on historical topology measurement data is characterized in that the method is realized by the following steps:
(1) acquiring historical topological data of a local network, and generating a local network topological graph according to the historical topological data;
(2) defining a super node based on a local network topological graph;
(3) dividing the local network topological graph into a plurality of super nodes according to the super nodes defined in the step (2), and generating a target IP address set by using the plurality of super nodes;
(4) measuring a local network by using a Looking Glass server as a measuring point and using the generated target IP address set as a target point to obtain local network topology measuring data;
the process of acquiring the historical topology data of the local network comprises the following steps:
downloading historical topology data from a third-party platform, and selecting a traceroute with a target IP address positioned in a local network as the historical topology data of the local network;
the process of generating the local network topological graph according to the historical topological data comprises the following steps:
selecting a first IP address in each hop TTL of a traceroute as a node of the TTL, generating a plurality of undirected edges according to the order of the TTL, and aggregating the edges of at least one node positioned in a local network to form an undirected graph G which is marked as a topological relation graph G of the local network;
for a certain node set in a local network topology map
Figure FDA0003069313660000011
Figure FDA0003069313660000012
Is a number v1Is a tree of root nodes, and
Figure FDA0003069313660000013
node of (3) through v1Connected with other nodes in the local network topological graph, the node set is called
Figure FDA0003069313660000014
Is a super node, called v1Is a super node
Figure FDA0003069313660000015
The other nodes are common nodes;
the main steps of dividing the local network topological graph into a plurality of super nodes according to the super nodes defined in the step (2) are as follows:
(a) calling the nodes with the node degree of 1 as leaf nodes, traversing all the leaf nodes, if no leaf node exists, ending, and taking the father nodes of the leaf nodes as root nodes;
(b) if the father node is not in the root node set of the super nodes, declaring a new super node taking the father node as the root node;
(c) if the leaf node is the root node of a certain super node, merging the super node corresponding to the leaf node into the super node corresponding to the parent node of the leaf node, and if the leaf node is not the root node of the super node, directly adding the leaf node into the super node corresponding to the parent node of the leaf node;
(d) removing all leaf nodes and corresponding edges thereof, and entering the leaf nodes and the corresponding edges into the step (a);
after the above steps are finished, the super node set reserved in (c) is a plurality of super nodes divided by the local network topological graph.
2. The method as claimed in claim 1, wherein the step of generating the destination IP address set by using a plurality of super nodes is to randomly select one or more surviving IP addresses from each super node to form the destination IP address set, wherein the surviving IP addresses are IP addresses reachable by using ping test.
3. The method for selecting the destination IP address of the local network based on the historical topology measurement data as claimed in claim 2, wherein in step (4), the Looking Glass server is used as the measurement point, the generated destination IP address set is used as the target point, the measurement point and the target point are cartesian products to form a local network topology measurement task, and the local network is measured according to the measurement task.
CN201810666495.XA 2018-06-25 2018-06-25 Local network destination IP address selection method based on historical topology measurement data Active CN108833609B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810666495.XA CN108833609B (en) 2018-06-25 2018-06-25 Local network destination IP address selection method based on historical topology measurement data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810666495.XA CN108833609B (en) 2018-06-25 2018-06-25 Local network destination IP address selection method based on historical topology measurement data

Publications (2)

Publication Number Publication Date
CN108833609A CN108833609A (en) 2018-11-16
CN108833609B true CN108833609B (en) 2021-07-13

Family

ID=64137492

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810666495.XA Active CN108833609B (en) 2018-06-25 2018-06-25 Local network destination IP address selection method based on historical topology measurement data

Country Status (1)

Country Link
CN (1) CN108833609B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108923963B (en) * 2018-06-25 2021-05-25 哈尔滨工业大学 Method for realizing automatic topology measurement by using Looking Glass measurement points
CN111064817B (en) * 2019-12-13 2022-02-22 中国人民解放军战略支援部队信息工程大学 City-level IP positioning method based on node sorting

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101610505A (en) * 2009-07-15 2009-12-23 常熟理工学院 The realization system of full IP mobile wireless sensor network of next generation route
CN101883039A (en) * 2010-05-13 2010-11-10 北京航空航天大学 Data transmission network of large-scale clustering system and construction method thereof
CN101917310A (en) * 2010-08-27 2010-12-15 电子科技大学 Network topology identification method based on flow analysis
CN103581329A (en) * 2013-11-14 2014-02-12 中国科学院大学 Construction method for topological structure based on clustered peer-to-peer network streaming media direct broadcast system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090172192A1 (en) * 2007-12-28 2009-07-02 Christian Michael F Mapless Global Traffic Load Balancing Via Anycast

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101610505A (en) * 2009-07-15 2009-12-23 常熟理工学院 The realization system of full IP mobile wireless sensor network of next generation route
CN101883039A (en) * 2010-05-13 2010-11-10 北京航空航天大学 Data transmission network of large-scale clustering system and construction method thereof
CN101917310A (en) * 2010-08-27 2010-12-15 电子科技大学 Network topology identification method based on flow analysis
CN103581329A (en) * 2013-11-14 2014-02-12 中国科学院大学 Construction method for topological structure based on clustered peer-to-peer network streaming media direct broadcast system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
《Internet路由器级拓扑测量中目标选择方法研究》;姜誉；何松;《通信学报》;20060225;全文 *
《IP级网络拓扑测量技术的研究与实现》;薛键；李东；张宇;《智能计算机与应用》;20140201;全文 *

Also Published As

Publication number Publication date
CN108833609A (en) 2018-11-16

Similar Documents

Publication Publication Date Title
Murdock et al. Target generation for internet-wide IPv6 scanning
Green et al. A fast algorithm for streaming betweenness centrality
Han et al. Fault-tolerant relay node placement in heterogeneous wireless sensor networks
Harb et al. K-means based clustering approach for data aggregation in periodic sensor networks
Zhao et al. Efficient shortest paths on massive social graphs
Vermeulen et al. Multilevel MDA-lite Paris traceroute
Liu et al. Distributed graph summarization
Shavitt et al. Quantifying the importance of vantage points distribution in internet topology measurements
CN108228724A (en) Power grid GIS topology analyzing method and storage medium based on chart database
Aktunc et al. A dynamic modularity based community detection algorithm for large-scale networks: DSLM
CN104202211A (en) Autonomous system level network topology identification method combining active and passive measurement
CN102215136A (en) Flow topology generation method and device
Cancela et al. Polynomial-time topological reductions that preserve the diameter constrained reliability of a communication network
CN108833609B (en) Local network destination IP address selection method based on historical topology measurement data
CN107611962B (en) Power grid system branch searching method and system and electronic equipment
Shen et al. Geographic location-based network-aware qos prediction for service composition
Baraglia et al. Godel: Delaunay overlays in p2p networks via gossip
Dimitropoulos et al. Revisiting Internet AS-level topology discovery
CN106982164A (en) A kind of method for discovering network topology and equipment
Pan et al. Overlapping community detection via leader-based local expansion in social networks
CN107679107A (en) A kind of grid equipment accessibility querying method and system based on chart database
KR102175199B1 (en) Topology discovery system and method in block-chain network
Shavitt et al. Quantifying the importance of vantage point distribution in internet topology mapping (extended version)
Ma et al. GWS-Geo: A graph neural network based model for street-level IPv6 geolocation
Magoni Network topology analysis and internet modelling with nem

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB03 Change of inventor or designer information
CB03 Change of inventor or designer information

Inventor after: Zhang Yu

Inventor after: Yu Zhuoxun

Inventor after: Zhang Ye

Inventor after: Zhang Hongli

Inventor before: Zhang Yu

Inventor before: Zhang Ye

Inventor before: Zhang Hongli

GR01 Patent grant
GR01 Patent grant