CN107404411A - A kind of parallel network topology discovery method - Google Patents

A kind of parallel network topology discovery method Download PDF

Info

Publication number
CN107404411A
CN107404411A CN201710810497.7A CN201710810497A CN107404411A CN 107404411 A CN107404411 A CN 107404411A CN 201710810497 A CN201710810497 A CN 201710810497A CN 107404411 A CN107404411 A CN 107404411A
Authority
CN
China
Prior art keywords
detection
node
collection
router
ttl
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.)
Granted
Application number
CN201710810497.7A
Other languages
Chinese (zh)
Other versions
CN107404411B (en
Inventor
玄世昌
苘大鹏
王巍
朱新立
靳小鹏
郁晨
张超
张宁
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 Engineering University
Original Assignee
Harbin Engineering University
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 Engineering University filed Critical Harbin Engineering University
Priority to CN201710810497.7A priority Critical patent/CN107404411B/en
Publication of CN107404411A publication Critical patent/CN107404411A/en
Application granted granted Critical
Publication of CN107404411B publication Critical patent/CN107404411B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/12Discovery or management of network topologies

Abstract

The invention discloses a kind of parallel network topology discovery method, discovering network topology field is related generally to.Wherein mainly include concurrent topology discovery and dynamic wheel rotary-die type two parts;Concurrent topology discovery is mainly made up of pretreatment module and route exploration module.Shared data is carried out by round-robin queue between pretreatment module and route exploration module.Pretreatment module and route exploration module perform parallel.The present invention reduces the traffic between detection source by way of changing shared global stopping collection, and adds accidental validation algorithm and parallel detecting mechanism during Topology Discovery to improve Topology Discovery integrity degree and detection efficient.It is under different network topology environment test result indicates that, improved topology discovery method improves detection efficient and integrality than Doubletree method, while reduces to share between detection source and global stop the traffic caused by collection.Therefore, improved topology discovery method has higher performance on discovering network topology.

Description

A kind of parallel network topology discovery method
Technical field
The invention of this reality, which is related to, refers to a kind of discovering network topology field, and in particular to a kind of parallel network Topology Discovery Method.
Background technology
As expanding day by day for network topology scale is increasingly complicated with structure, brought greatly to efficient network management Challenge.The premise that network manager is efficiently managed network is to grasp the topological structure of network.Therefore, how to improve It is the principal concern of efficient method for discovering network topology to ensure integrality while discovering network topology efficiency, while this It is related to the availability of topology discovery method.Therefore need badly and method for discovering network topology is studied.
Traditional Traceroute methods can produce substantial amounts of redundancy detection when carrying out discovering network topology, can not be efficient Acquisition network topology structure.In order to reduce redundancy detection, Doubletree methods are this respectively by setting two to stop collection Ground stops collection and global stops collection to reduce in the sensing point during discovering network topology redundancy between redundancy and sensing point.Although Doubletree methods reduce redundancy detection, improve detection efficient, but Doubletree methods there is also some not Foot part:First, the overall situation is stopped collection by detection source is sent to master control node, master control node is distributed to each detection source again after collecting, The traffic is big between detection source;Second, detection source is when detecting each jump route of destination node, it is necessary to waits detection data to be sent The feedback information of bag, the detection of next-hop route then could be carried out, reduces detection efficient;Three be due to Network Load Balance With the multiple access problem of router some router interfaces can not be found, so as to cause the integrality of discovering network topology not Foot.
Qiao Hong et al. proposes a kind of improvement topology discovery method based on Doubletree methods, for Doubletree side Method is optimized in weak point present on the traffic, and the global one-dimension storage formula for stopping collection of Doubletree methods is changed For two-dimensional storage mode.And increase a mark whether changed for each BloomFilter, between such detection source every time When exchanging data, it is only necessary to exchange the data volume changed.This method reduces the traffic to a certain extent, but still needs Want the global communications burden for stopping collection, not being reduced master control Node station of master control Node station processing.In addition, this method does not also solve Certainly Doubletree algorithms are in network topology probe the problem of integrality deficiency.
Yang Xu proposes a kind of gradual topology probe strategy based on revolution detection, and Doubletree methods are being communicated The weak point of amount is improved, and specific practice is that detection source is divided into several packets, is handed between packet Change global stopping collection.Although it the method reduce number of communications and total traffic, it is not necessary to which the master control Node station merger overall situation is stopped Only collect, the new global collection that stops then being transmitted further to all detection sources.But each detection source is grouped in same in this method Only have a probe node machine to be detected in moment, each detection source packet is too strong according to lazyness to previous packet in addition, So that flexibility is not strong.
The content of the invention
The present invention is a kind of parallel network topology discovery method, it is therefore an objective to reduces the traffic between detection source while improves topology It was found that integrity degree and detection efficient.
A kind of parallel network topology discovery method includes concurrent topology discovery and dynamic wheel rotary-die type, wherein dynamic wheel revolving die Formula is for implementing concurrent topology discovery;Topology Discovery is wherein achieved by the steps of:
Described concurrent topology discovery is made up of pretreatment module and route exploration module.Visited in pretreatment module and route Survey between module and shared data is carried out by round-robin queue.Pretreatment module and route exploration module perform parallel.
Step 1:Concentrated in pretreatment module from detection purpose and choose a detection destination node;
Step 2:Host detection is carried out, calculates detection source Node station to total hop count of detection destination node;
Step 3:Destination node, ttl value, probe mark are saved in round-robin queue in the form of structure;And enter The route exploration module being made up of initial probe module, exchange detecting module and task scheduling modules;
Step 4:The non-probe node of minimum is oriented using improved binary search algorithm in initial probe module min;
Step 5:Parallel detecting, the result of detection are carried out to non-probe node using multithreading by task scheduling modules Interim stopping is stored in concentrate;
Step 6:The interim router node for stopping concentrating is copied to destination node, intermediate router joint form The overall situation stops concentrating, and router node of the ttl value less than or equal to h/2 is saved in intermediate router node and locally stops collection, By router node of the ttl value more than or equal to h/2 with intermediate router node, TTL is saved in local stopping collection;
Step 7:Interim stopping collection being emptied, discharges and stops collecting shared space temporarily.
Step 8:In detecting module is exchanged send TTL be equal to h/2 values probe data packet, wait-for-response message and from The middle interface mid for obtaining router, if destination node, router interface mid stops concentrating in the overall situation, then illustrates Part II It had been detected with the path of Part III, had been verified in Part II and Part III using accidental validation algorithm, had been turned To step 9;Otherwise step 10 is gone to;
Step 9:If router interface mid is locally stopping concentrating, illustrate that the path of Part I has been detected Cross, Part I is verified using accidental validation algorithm, goes to step 13;Otherwise using TTL in a manner of h/2 successively decreases Detected backward, go to step 13.
Step 10:If router interface mid is locally stopping concentrating, illustrate that the path of Part I has been detected Cross, Part I is verified using accidental validation algorithm, goes to step 4;Otherwise reduced by half in Part I using improved Lookup algorithm finds the non-probe node min of minimum, goes to step 11.
Step 11:Detection source sends the probe data packet that TTL is 3h/4 values, and wait-for-response message simultaneously therefrom obtains route The interface mid2 of device, if destination node, router interface mid2 stops concentrating in the overall situation, then has illustrated the path of Part III Through being detected, then detection is carried forward in a manner of h/2 is incremental using TTL, is calculated afterwards in Part III using accidental validation Method is verified, goes to step 13;Otherwise step 12 is gone to.
Step 12:Stop concentrating if router interface mid, TTL and router interface mid2, TTL are simultaneously local, Then verified in Part II using accidental validation algorithm, detection be carried forward in a manner of 3h/4 is incremental using TTL afterwards, Go to step 13;Otherwise detection is carried forward in a manner of 3h/4 is incremental using TTL, goes to step 13.
Step 13:Parallel detecting, the knot of detection are carried out to non-probe node using multithreading by task scheduling modules Fruit is stored in interim stopping and concentrated.Then by the interim router node for stopping concentrating with destination node, intermediate router node Form copies global stopping to and concentrated, and router node of the ttl value less than or equal to h/2 is preserved with intermediate router joint form Stop collection to local, ttl value be more than or equal to router node that h/2 is less than or equal between 3h/4 with intermediate router node, TTL forms are saved in local stopping collection.Interim stopping collection finally being emptied, discharges and stops collecting shared space temporarily.
The dynamic rotation mode embodiment flow is as follows:
Step 1:During initial probe, master control Node station sends probe instructions to all detection source Node stations, and specifies and visit Survey source SiThe purpose collection packet D that Node station should detecti, all detection source SiThe corresponding purpose collection D of the parallel detection of Node stationi Packet.As detection source Node station SiCorresponding purpose collection D is completed in detectioniAfter all purposes node in packet, detection source is to master control section Point machine, which is sent, oneself has completed purpose collection DiThe detection of packet.Master control Node station detects in each detection source Node station of guarantee Dynamic task scheduling, master control node are carried out according to the process performance of detection source Node station on the basis of all detection purpose collection packets Machine sends instruction to detection source Node station and indicates detection source Node station SiShould be purpose collection DiThe global of packet stops collection PiSend Give detection source Node station Sj, while indicate that detection source Node station should receive purpose collection DkThe global of packet stops collection Pk
Step 2:Detection source SjNode station receives purpose collection DiThe global of packet stops collection Pi, start to purpose collection DiPacket Detected according to the corresponding mode of step 1);
Step 3:After dynamic robin scheduling by g-1 step 2, each detection source Node station is completed to institute There is the detection of detection purpose collection packet.
The advantage of the invention is that:
Parallel network topology proposed by the present invention finds method, is visited by way of changing the shared overall situation and stopping collection to reduce The traffic between survey source, and addition accidental validation algorithm and parallel detecting mechanism are complete to improve Topology Discovery during Topology Discovery Whole degree and detection efficient.Under different network topology environment, improved topology discovery method and Doubletree methods are entered Gone contrast experiment, test result indicates that, improved topology discovery method than Doubletree method improve detection efficient and Integrality, while reduce the traffic caused by shared global stopping collection between detection source.As can be seen here, compare Doubletree Method, improved topology discovery method have higher performance on discovering network topology.
Brief description of the drawings
Fig. 1 is concurrent topology discovery block diagram;
Fig. 2 is initial probe module map;
Fig. 3 is to exchange detecting module figure;
Fig. 4 is dynamic rotation ideograph.
Embodiment
The present invention is described further below in conjunction with the accompanying drawings.
Initial probe module, during destination node is detected, all result of detections are all first saved in interim stopping Concentrate, finally copying the interim result for stopping concentration to the overall situation according to certain condition again stops collection or local stopping concentration. The flow of initial probe module is as shown in Fig. 2 idiographic flow is as follows:
1) the non-probe node min of minimum is oriented using improved binary search algorithm;
2) parallel detecting is carried out to non-probe node using multithreading by task scheduling modules, the result of detection is stored in It is interim to stop concentrating.
3) by the interim router node for stopping concentrating with destination node, intermediate router joint form copies the overall situation to and stopped Only concentrate, router node of the ttl value less than or equal to h/2 is saved in local stopping collection with intermediate router node, by ttl value For router node more than or equal to h/2 with intermediate router node, TTL is saved in local stopping collection.
4) interim stopping collection being emptied, discharges and stops collecting shared space temporarily.
Detecting module is exchanged, using the thought divided and ruled, detection source to the path between detection destination node according to distance Ttl value is divided into 3 parts:1 to h/2 is Part I;H/2 to 3h/4 is Part II;3h/4 to h is Part III.Visiting During surveying destination node, result of detection is all first saved in interim stopping and concentrated, finally again by the interim result for stopping concentrating Global stopping collection being copied to according to certain condition or local stopping is concentrated.Detect forward and the process that detects backward and Flow in Doubletree methods is basically identical.It is as follows to exchange detecting module substantially flow:
1) probe data packet that TTL is equal to h/2 values, wait-for-response message and the interface mid for therefrom obtaining router are sent, If destination node, router interface mid stops concentrating in the overall situation, then illustrate the path of Part II and Part III by Detected, and verified, gone to 2) using accidental validation algorithm in Part II and Part III;Otherwise go to 3).
2) if router interface mid is locally stopping concentrating, illustrate that the path of Part I had been detected, make Part I is verified with accidental validation algorithm, gone to 6);Otherwise visited backward in a manner of h/2 successively decreases using TTL Survey, go to 6).
3) if router interface mid is locally stopping concentrating, illustrate that the path of Part I had been detected, it is right Part I is verified using accidental validation algorithm, is gone to 4);Otherwise improved binary search algorithm is used in Part I The non-probe node min of minimum is found, is gone to 4).
4) detection source sends the probe data packet that TTL is 3h/4 values, and wait-for-response message simultaneously therefrom obtains connecing for router Mouth mid2, if destination node, router interface mid2 stops concentrating in the overall situation, then illustrates that the path of Part III is visited Surveyed, then detection is carried forward in a manner of h/2 is incremental using TTL, carried out afterwards in Part III using accidental validation algorithm 6) checking, goes to;Otherwise go to 5).
Stop concentrating if 5) router interface mid, TTL and router interface mid2, TTL are simultaneously local, then the Two parts are verified using accidental validation algorithm, and detection is carried forward in a manner of 3h/4 is incremental using TTL afterwards, is gone to 6);Otherwise detection is carried forward in a manner of 3h/4 is incremental using TTL, gone to 6).
6) parallel detecting is carried out to non-probe node using multithreading by task scheduling modules, the result of detection is stored in It is interim to stop concentrating.Then by the interim router node for stopping concentrating with destination node, intermediate router joint form copy Stop concentrating to the overall situation, router node of the ttl value less than or equal to h/2 is saved in local stop with intermediate router joint form Only collect, ttl value is more than or equal into h/2 is protected less than or equal to the router node between 3h/4 with intermediate router node, TTL forms It is stored to local stopping collection.Interim stopping collection finally being emptied, discharges and stops collecting shared space temporarily.Exchange the detailed of detecting module Flow is as shown in Figure 3.
Dynamic rotation mode embodiment is as shown in Figure 4:
1) during initial probe, master control Node station sends probe instructions to all detection source Node stations, and specifies detection source The purpose collection packet Di that Si Node stations should detect, corresponding purpose collection Di points of the parallel detection of all detection source Si Node stations Group.After all purposes node in corresponding purpose collection Di packet is completed in detection source Node station Si detections, detection source is to master control section Point machine sends the detection for oneself having completed purpose collection Di packets.Master control Node station detects in each detection source Node station of guarantee Dynamic task scheduling, master control node are carried out according to the process performance of detection source Node station on the basis of all detection purpose collection packets Machine sends the global stopping collection Pi hairs for indicating that detection source Node station Si should be grouped purpose collection Di to detection source Node station Detection source Node station Sj is given, while indicates that detection source Node station should receive the global of purpose collection Dk packets and stop collection Pk.
2) detection source Sj Node stations receive the global of purpose collection Di packets and stop collection Pi, start to purpose collection Di packets according to The corresponding mode of step 1) is detected.
3) after the dynamic robin scheduling by g-1 step 2, each detection source Node station is completed to all spies Survey the detection of purpose collection packet.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies Change, equivalent substitution, improvement etc., should be included in the scope of the protection.

Claims (3)

1. a kind of parallel network topology discovery method includes concurrent topology discovery and dynamic wheel rotary-die type, wherein dynamic wheel rotary-die type It is for implementing concurrent topology discovery;Topology Discovery is wherein achieved by the steps of:
Step 1:Concentrated in pretreatment module from detection purpose and choose a detection destination node;
Step 2:Host detection is carried out, calculates detection source Node station to total hop count of detection destination node;
Step 3:Destination node, ttl value, probe mark are saved in round-robin queue in the form of structure;And enter by first Beginning detecting module, the route exploration module for exchanging detecting module and task scheduling modules composition;
Step 4:The non-probe node min of minimum is oriented using improved binary search algorithm in initial probe module;
Step 5:Parallel detecting is carried out to non-probe node using multithreading by task scheduling modules, the result of detection preserves Stop concentrating interim;
Step 6:The interim router node for stopping concentrating is copied to the overall situation with destination node, intermediate router joint form Stop concentrating, router node of the ttl value less than or equal to h/2 is saved in local stopping collection with intermediate router node, by TTL For router node of the value more than or equal to h/2 with intermediate router node, TTL is saved in local stopping collection;
Step 7:Interim stopping collection being emptied, discharges and stops collecting shared space temporarily;
Step 8:The probe data packet that TTL is equal to h/2 values is sent in detecting module is exchanged, wait-for-response message simultaneously therefrom obtains The interface mid of router is taken, if destination node, router interface mid stops concentrating in the overall situation, then illustrates Part II and the The path of three parts had been detected, and was verified in Part II and Part III using accidental validation algorithm, went to step Rapid nine;Otherwise step 10 is gone to;
Step 9:If router interface mid is locally stopping concentrating, illustrate that the path of Part I had been detected, Part I is verified using accidental validation algorithm, goes to step 13;Otherwise using TTL in a manner of h/2 successively decreases to After detected, go to step 13;
Step 10:If router interface mid is locally stopping concentrating, illustrate that the path of Part I had been detected, Part I is verified using accidental validation algorithm, goes to step 4;Otherwise looked into by half using improved in Part I Look for algorithm to find the non-probe node min of minimum, go to step 11;
Step 11:Detection source sends the probe data packet that TTL is 3h/4 values, and wait-for-response message simultaneously therefrom obtains router Interface mid2, if destination node, router interface mid2 stops concentrating in the overall situation, then illustrate the path of Part III by Detected, then detection is carried forward in a manner of h/2 is incremental using TTL, entered afterwards in Part III using accidental validation algorithm Row checking, goes to step 13;Otherwise step 12 is gone to;
Step 12:Stop concentrating if router interface mid, TTL and router interface mid2, TTL are simultaneously local, then exist Part II is verified using accidental validation algorithm, and detection is carried forward in a manner of 3h/4 is incremental using TTL afterwards, is gone to Step 13;Otherwise detection is carried forward in a manner of 3h/4 is incremental using TTL, goes to step 13;
Step 13:Parallel detecting is carried out to non-probe node using multithreading by task scheduling modules, the result of detection is protected Stop concentrating in the presence of interim;Then by the interim router node for stopping concentrating with destination node, intermediate router joint form Copy global stopping to concentrating, router node of the ttl value less than or equal to h/2 is saved in this with intermediate router joint form Ground stops collection, and ttl value is more than or equal into h/2 less than or equal to the router node between 3h/4 with intermediate router node, TTL shapes Formula is saved in local stopping collection;Interim stopping collection finally being emptied, discharges and stops collecting shared space temporarily.
A kind of 2. parallel network topology discovery method according to claim 1, it is characterised in that described dynamic wheel revolving die Formula embodiment flow is as follows:
Step 1:During initial probe, master control Node station sends probe instructions to all detection source Node stations, and specifies detection source SiThe purpose collection packet D that Node station should detecti, all detection source SiThe corresponding purpose collection D of the parallel detection of Node stationiPacket; As detection source Node station SiCorresponding purpose collection D is completed in detectioniAfter all purposes node in packet, detection source is to master control Node station Send and oneself completed purpose collection DiThe detection of packet;Master control Node station is ensureing that it is all that each detection source Node station detects Detect and dynamic task scheduling is carried out according to the process performance of detection source Node station on the basis of purpose collection is grouped, master control Node station is given Detection source Node station sends instruction and indicates detection source Node station SiShould be purpose collection DiThe global of packet stops collection PiIt is sent to spy Survey source node machine Sj, while indicate that detection source Node station should receive purpose collection DkThe global of packet stops collection Pk
Step 2:Detection source SjNode station receives purpose collection DiThe global of packet stops collection Pi, start to purpose collection DiPacket according to The corresponding mode of step 1) is detected;
Step 3:After dynamic robin scheduling by g-1 step 2, each detection source Node station is completed to all spies Survey the detection of purpose collection packet.
A kind of 3. parallel network topology discovery method according to claim 1, it is characterised in that described concurrent topology hair Now it is made up of pretreatment module and route exploration module;Entered between pretreatment module and route exploration module by round-robin queue Row shared data;Pretreatment module and route exploration module perform parallel.
CN201710810497.7A 2017-09-11 2017-09-11 Parallel network topology discovery method Active CN107404411B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710810497.7A CN107404411B (en) 2017-09-11 2017-09-11 Parallel network topology discovery method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710810497.7A CN107404411B (en) 2017-09-11 2017-09-11 Parallel network topology discovery method

Publications (2)

Publication Number Publication Date
CN107404411A true CN107404411A (en) 2017-11-28
CN107404411B CN107404411B (en) 2020-12-25

Family

ID=60397157

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710810497.7A Active CN107404411B (en) 2017-09-11 2017-09-11 Parallel network topology discovery method

Country Status (1)

Country Link
CN (1) CN107404411B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112737861A (en) * 2021-01-06 2021-04-30 郑州埃文计算机科技有限公司 IP interface layer network topology detection method
CN116708273A (en) * 2023-07-28 2023-09-05 中国电子科技网络信息安全有限公司 Method for generating stopping judgment classification model, network topology detection method and device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080232262A1 (en) * 2007-03-19 2008-09-25 Alina Beygelzimer Method and apparatus for network topology discovery using closure approach
CN103516615A (en) * 2013-07-15 2014-01-15 中国人民解放军国防科学技术大学 Method and device for determining network topology

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080232262A1 (en) * 2007-03-19 2008-09-25 Alina Beygelzimer Method and apparatus for network topology discovery using closure approach
CN103516615A (en) * 2013-07-15 2014-01-15 中国人民解放军国防科学技术大学 Method and device for determining network topology

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
乔宏等: "基于改进DoubleT ree 算法的网络拓扑发现方法", 《计算机工程与科学》 *
杨旭: "《硕士论文》", 15 July 2016 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112737861A (en) * 2021-01-06 2021-04-30 郑州埃文计算机科技有限公司 IP interface layer network topology detection method
CN116708273A (en) * 2023-07-28 2023-09-05 中国电子科技网络信息安全有限公司 Method for generating stopping judgment classification model, network topology detection method and device
CN116708273B (en) * 2023-07-28 2023-10-31 中国电子科技网络信息安全有限公司 Method for generating stopping judgment classification model, network topology detection method and device

Also Published As

Publication number Publication date
CN107404411B (en) 2020-12-25

Similar Documents

Publication Publication Date Title
CN104184664B (en) Route forwarding table items generation method and device
CN103581021B (en) The method and apparatus of business detection under software defined network
CN104298541A (en) Data distribution algorithm and data distribution device for cloud storage system
CN106161610A (en) A kind of method and system of distributed storage
CN103699618B (en) The generation method and system of data sheet
CN105099922A (en) Cross-VXLAN (Virtual eXtensible Local Area Network) data message forwarding method and device
CN103139081B (en) Distributed hashtable routing table update method and node
CN109155760A (en) Path is grouped using fixed header size to record
CN103078796B (en) A kind of route computing method and equipment
CN109361596A (en) Route computing method, device and electronic equipment
CN100403699C (en) Distributing wireless network key node detecting method
CN107404411A (en) A kind of parallel network topology discovery method
CN102694732A (en) Method and system for constructing virtual network based on local virtualization
CN102546380B (en) Modified tree-based multicast routing scheme
US10157076B2 (en) Virtual machine migration device and method
CN104093182B (en) A kind of method for obtaining a plurality of reliable communication path in multilayer wireless network based on field strength
CN106453091A (en) Equivalent routing management method and device of router forwarding plane
CN104376712B (en) Lack transport information complementing device and its method
CN104954477A (en) Large-scale graph data stream partitioning method and system based on concurrency improvement
CN104683232B (en) The route selection method of long Connection Service can be provided in a kind of VANET
CN106851766A (en) A kind of wireless sense network route method of high robust low latency
CN106656656A (en) Network device package capture method and device
CN101459586A (en) Network address dispensing method and routing method for long link shape ZigBee network
CN110086819B (en) FRIT-based attack tracing method
CN103945412B (en) A kind of WSN void node detection methods based on connectedness

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
GR01 Patent grant
GR01 Patent grant