CN101854383B - Large-scale network resource searching method based on de Bruijn image - Google Patents
Large-scale network resource searching method based on de Bruijn image Download PDFInfo
- Publication number
- CN101854383B CN101854383B CN2010101583767A CN201010158376A CN101854383B CN 101854383 B CN101854383 B CN 101854383B CN 2010101583767 A CN2010101583767 A CN 2010101583767A CN 201010158376 A CN201010158376 A CN 201010158376A CN 101854383 B CN101854383 B CN 101854383B
- Authority
- CN
- China
- Prior art keywords
- node
- bruijn
- resource
- sign
- search
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a large-scale network resource searching method based on a de Bruijn image, aiming to solve the technical problem of requirements on small node degree and low searching delay of large-scale network searching. The method comprises the following steps of: firstly, naming a resource object; secondly, realizing network topology by adopting a general de Bruijn image; thirdly, processing joining and exiting events of nodes by adopting an identification splitting and merging method; and realizing the searching of the resources by adopting a bit-by-bit matching mode and judging whether the searching is finished or not by adopting a front-to-back matching rule. By adopting the method, the average degree of the network topology is 4, and the maximum searching delay is only logdN, which can greatly improve the large-scale network resource searching efficiency.
Description
Technical field
The present invention relates to the resource search method in the computer network, especially a kind of resource search method based on de Bruijn figure.
Background technology
P2P (peer-to-peer) network is a kind of network of rising in recent years.In the P2P network, each node logically is reciprocity, does not have the branch of client and server, can directly communicate between each node with mutual.At present, the P2P network all has wide application at key areas such as scientific research, ecommerce, E-Government and military applications.In order to realize that resources effective is shared and comprehensive utilization, the P2P network user need search for satisfactory resource, and resource searching is one of key technology of P2P network.
According to resource organizations's pattern, the P2P network can be divided into two kinds usually: structuring (Structured) P2P network and destructuring (Unstructured) P2P network.Structured P 2 P network has obtained extensive application at present owing to have performance more reliably on Internet.
Resource searching problem in the structured P 2 P network can abstractly be: how on large scale network, to realize the effective search of resource object, and adapt to the dynamic change of node.
At present, the resource search method in the structured P 2 P network mainly comprises Chord, CAN, Pastry, Tapestry etc.In said method, each node all has unique sign, and based on certain algorithm building network topology between node.Each node is all safeguarded one " transmitting ", preserves the information of relevant neighbours' node.Each resource object obtains the resource object sign according to its keyword through hash function.The sign of resource object and node sign belong to same or analogous name space usually, and each node all is responsible for the part of resource object identifier space.When node added or withdraws from, each junction associated need be revised and transmit, and dynamically adjusted the identifier space scope that it is responsible for, to safeguard the consistency of distributed Hash table.When resource searching, each node " is transmitted " the resource searching forwards to corresponding neighbours' node according to it, accomplishes search up to final arrival destination node.The important parameter of estimating the large-scale network resource search performance comprises the node number of degrees and search delay etc.The node number of degrees are meant the size of " routing table " safeguarded on each node; Search delay is meant the logic leapfrog number that the first resource searching request is transmitted in system.The standard of evaluating network searching method mainly comprises the node number of degrees and search delay, and searching method should have the less node number of degrees on the one hand efficiently, should have lower search delay on the other hand.But existing searching method is not all realized the compromise preferably of the node number of degrees and search delay.
Summary of the invention
Technical problem to be solved by this invention: search for the little and low demand of search delay to the node number of degrees to large scale network; A kind of large-scale network resource searching method based on de Bruijn figure is proposed; This method can either have the less node number of degrees (being less routing table), has lower search delay again.
In order to address the above problem, the technical scheme that the present invention proposes is:
The first step; Be the resource object name: to each node and resource object; Adopt document " RFC 1321:The MD5Message-DigestAlgorithm " (http://www.ietf.org/rfc/rfc1321.txt, April 1992) described MD5 (eap-message digest 5) algorithm to name.
In second step, adopt de Bruijn figure realization network topology: (d D) is a kind of directed graph to de Bruijn figure B, and wherein d is the base (span that is each character in the node sign is 0 to d-1) of each node sign, the length that D identifies for each node.Each some u=u
1u
2... u
DHave the d bar to go out the limit: to any α ∈ 0,1,2 ..., d-1}, some u have one to some v=u
2u
3... u
Dα goes out the limit.Document " A Combinatorial Problem " (Proc.of Koninklijke Nederlundse Academic van Watenschappen, vol.A49, pp.758-764,1946) proves that the maximum delay of de Bruijn figure is log
dN, the node number of degrees are 2d.
The nodal point number of traditional de Bruijn figure is d
D, can not hold the node of arbitrary number.For example, get d=2, then nodal point number can only be 2
1=2,2
2=4,2
3=8 ...Therefore, we adopt document " Factoring and Scaling Kautz Digraphs " (Research Report 94-15, LIP ENSL; 69364Lyon, France, the general de Bruijn figure definition that Apr.1994) proposes: make GB (d; N) representing base is the vague generalization de Bruijn figure of n for d and node number; ((GB (d, n)) is respectively its point set V for GB (d, n)) and limit collection E so
V(GB(d，n))＝{0，1，…，n-1}， (1)
E(GB(d，n))＝{[i，(d×i+α)modn]|0≤α≤d-1}， (2)
Each node sign has its corresponding binary representation.Because B (d, D) ≡ GB (d, d
D), so vague generalization de Bruijn figure is the superset (being that any conventional de Bruijn figure can be expressed as a vague generalization de Bruijn figure) of traditional de Bruijn figure, and vague generalization de Bruijn figure can hold the node of arbitrary number.Node in the some map network topology of vague generalization de Bruijn figure connects between the node in the map network topology of limit, can obtain corresponding network topology.For simplified design, with the sign of the point among the binary number representation vague generalization de Bruijn figure.
In the 3rd step, incident is withdrawed from the adding of adopting " identification splitting act of union " to handle node: when new node P adds fashionablely, the node sign length of at first choosing the arbitrary node V:V that satisfies following condition in the network is not more than its arbitrary neighbours' node; Node V divides then, establishes node V and is designated v
1v
2... v
k, after then node P adds, v
1v
2... v
kBe split into v
1v
2... v
k0 and v
1v
2... v
k1, the sign of node V becomes v
1v
2... v
k0, the sign of new node P is set to v
1v
2... v
k1.And then upgrade the routing table of P, V and junction associated according to vague generalization de Bruijn figure, promptly the annexation of each node satisfies formula (2).It is the inverse process that adds that node withdraws from.When detecting node Q and leave, at first choose certain a pair of sibling Y=y that satisfies following condition in the network
1y
2... y
P-10 and Y '=y
1y
2... y
P-1The sign length of 1:Y and Y ' all is not less than its neighbours separately; Then the sign of node Y is changed into the sign of Q, the sign of Y ' changes y into
1y
2... y
P-1(being that Y and Y ' merge); And upgrade the routing table of junction associated according to vague generalization de Bruijn figure.
In the 4th step, adopt " coupling by turn " mode to realize the search of resource: resource searching is accomplished through transmitting search message between node.At first definition " front and back matching value ": to any two character string u=u
1u
2... u
mAnd v=v
1v
2... v
n, the front and back matching value of u and v (be designated as QH (u is meant that v)=x) the back x position of u is identical with the preceding x position of v, for example 100
11With
11010 front and back matching value is 2.Each node calculates front and back matching value x own and the target resource object after receiving the resource object search message, in routing table, selecting front and back matching value with the target resource object then is that the neighbours of x+1 transmit.
In the 5th step, adopt " front and back coupling " rule judgment whether to accomplish search: node u=u
1u
2... u
mHave resource s=s
1s
2... s
r, and if only if: QH (u, s)=| u|, perhaps QH (u, s)=| u|-1 ∧ u
1=s
rWhen satisfying the front and back matched rule, the search of accomplishing resource is described, finish, proceed search otherwise changeed for the 4th step.
Adopt the present invention can reach following technique effect:
1. in the present invention, each node is organized neighborhood according to their sign through simulation vague generalization de Bruijn figure, by document " Factoring and Scaling Kautz Digraphs " (Research Report 94-15; LIP ENSL; 69364Lyon, France, Apr.1994); The average number of degrees that can know this network topology are 4, and maximum search postpones to be merely log
dN.
2. on the basis that forms good topological structure, matching mode was transmitted resource searching message before and after the present invention used, and can resource searching message be forwarded to from the near neighbours' node of destination node exactly, improved the resource searching efficient of large scale network.
Description of drawings
Fig. 1 is an overview flow chart of the present invention;
Fig. 2 is a de Bruijn illustrated example;
Fig. 3 is that node adds the processing procedure example.
Embodiment
The present invention mainly comprised for five steps (as shown in Figure 1): 1. be the resource object name; 2. adopt de Bruijn figure to realize network topology; 3. incident is withdrawed from the adding of adopting " identification splitting act of union " to handle node; 4. adopt " coupling by turn " mode to realize the search of resource; 5. adopt " front and back coupling " rule judgment whether to accomplish search.
In second step, the present invention adopts de Bruijn figure to realize network topology, and Fig. 2 has provided the example of two de Bruijn figure, and left side figure is B (2,2), and right figure is B (2,3).
In the 3rd step, incident is withdrawed from the adding that the present invention adopts " identification splitting act of union " to handle node, and Fig. 3 has provided a node and added the example of handling.Before new node adds, V=000 in left figure; After new node added, the sign of V became 0000, and P is designated 0001, and connection relation is shown in right figure: the limit neighbours that go out of V=0000 are 001,010, and going into the limit neighbours is 010,100; The limit neighbours that go out of P=0001 are 100,101, and going into the limit neighbours is 010,100.
Claims (2)
1. large-scale network resource searching method based on de Bruijn figure is characterized in that may further comprise the steps:
The first step is the resource object name: to each node and resource object, adopt the MD5 algorithm to name;
In second step, adopt de Bruijn figure to realize network topology: to adopt general de Bruijn figure definition: make GB (d, n) represent base is that the vague generalization de Bruijn of n schemes for d and node number, ((GB (d, n)) is respectively its point set V for GB (d, n)) and limit collection E so
V(GB(d，n))＝{0，1，…，n-1}， (1)
E(GB(d，n))＝{[i，(d×i+α)modn]|0≤α≤d-1}， (2)
Node in the some map network topology of vague generalization de Bruijn figure connects between the node in the map network topology of limit, promptly obtains corresponding network topology;
In the 3rd step, incident is withdrawed from the adding of adopting " identification splitting act of union " to handle node: when new node P adds fashionablely, the node sign length of at first choosing the arbitrary node V:V that satisfies following condition in the network is not more than its arbitrary neighbours' node; Node V divides then, establishes node V and is designated v
1v
2... v
k, after then node P adds, v
1v
2... v
kBe split into v
1v
2... v
k0 and v
1v
2..v
k1, the sign of node V becomes v
1v
2... v
k0, the sign of new node P is set to v
1v
2... v
k1; And then upgrade the routing table of P, V and junction associated according to vague generalization de Bruijn figure, promptly the annexation of each node satisfies formula (2); When detecting node Q and leave, at first choose certain a pair of sibling Y=y that satisfies following condition in the network
1y
2... y
P-10 and Y '=y
1y
2... y
P-1The sign length of 1:Y and Y ' all is not less than its neighbours separately; Then the sign of node Y is changed into the sign of Q, the sign of Y ' changes y into
1y
2... y
P-1And upgrade the routing table of junction associated according to vague generalization de Bruijn figure;
In the 4th step, adopt " coupling by turn " mode to realize the search of resource: resource searching is accomplished through transmitting search message between node; At first definition " front and back matching value ": to any two character string u=u
1u
2... u
mAnd v=v
1v
2... v
nThe front and back matching value of u and v, be designated as QH (u, v)=x; The back x position that is meant u is identical with the preceding x position of v; Each node calculates front and back matching value x own and the target resource object after receiving the resource object search message, in routing table, selecting front and back matching value with the target resource object then is that the neighbours of x+1 transmit;
In the 5th step, adopt " front and back coupling " rule judgment whether to accomplish search: node u=u
1u
2... u
mHave resource s=s
1s
2... s
r, and if only if: QH (u, s)=| u|, perhaps QH (u, s)=| u|-1 ∧ u
1=s
r, when satisfying the front and back matched rule, the search of accomplishing resource is described, finish; Otherwise changeed for the 4th step and proceed search.
2. the large-scale network resource searching method based on de Bruijn figure as claimed in claim 1, the sign that it is characterized in that the point among the vague generalization de Bruijn figure is with binary number representation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101583767A CN101854383B (en) | 2010-04-28 | 2010-04-28 | Large-scale network resource searching method based on de Bruijn image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101583767A CN101854383B (en) | 2010-04-28 | 2010-04-28 | Large-scale network resource searching method based on de Bruijn image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101854383A CN101854383A (en) | 2010-10-06 |
| CN101854383B true CN101854383B (en) | 2012-07-04 |
Family
ID=42805647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101583767A Expired - Fee Related CN101854383B (en) | 2010-04-28 | 2010-04-28 | Large-scale network resource searching method based on de Bruijn image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101854383B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103699813B (en) * | 2013-12-10 | 2017-05-10 | 深圳先进技术研究院 | Method for identifying and removing repeated bidirectional edges of bidirectional multistep De Bruijn graph |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442479A (en) * | 2007-11-22 | 2009-05-27 | 华为技术有限公司 | Method, equipment and system for updating route in P2P peer-to-peer after node failure |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070280276A1 (en) * | 2006-05-30 | 2007-12-06 | Tellabs Operations, Inc. | Method and apparatus for managing client layer communications services |
-
2010
- 2010-04-28 CN CN2010101583767A patent/CN101854383B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442479A (en) * | 2007-11-22 | 2009-05-27 | 华为技术有限公司 | Method, equipment and system for updating route in P2P peer-to-peer after node failure |
Non-Patent Citations (1)
| Title |
|---|
| 张一鸣等.一种面向大规模P2P系统的快速搜索算法.《软件学报》.2008,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101854383A (en) | 2010-10-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Triantafillou et al. | Towards a unifying framework for complex query processing over structured peer-to-peer data networks | |
| Li et al. | Fissione: A scalable constant degree and low congestion dht scheme based on kautz graphs | |
| JP2002335269A (en) | Peer-to-peer name resolution protocol(pnrp) and multilevel cache for use therewith | |
| CN101719155B (en) | Method of multidimensional attribute range inquiry for supporting distributed multi-cluster computing environment | |
| CN101854383B (en) | Large-scale network resource searching method based on de Bruijn image | |
| CN101094244A (en) | Method of high performance distributed Hash table in P2P system | |
| Tato et al. | Designing overlay networks for decentralized clouds | |
| Zöls et al. | The hybrid chord protocol: A peer-to-peer lookup service for context-aware mobile applications | |
| KR102596700B1 (en) | System and method for inter shard transaction in blockchain network | |
| Zhang et al. | A new algorithm of service discovery based on DHT for mobile application | |
| Bharathi et al. | Model based controller design for a spherical tank | |
| CN104168147B (en) | A kind of node maitaining method that one-dimensional chained list is based on for P2P network monitorings | |
| CN101040506B (en) | Method for initializing a peer-to-peer data network | |
| Ma et al. | A new P2P network routing algorithm based on ISODATA clustering topology | |
| Makikawa et al. | Balance and proximity-aware Skip Graph construction | |
| Lee et al. | P2P trust model: The resource chain model | |
| Cao et al. | A multi-level Super peer based P2P architecture | |
| Yu et al. | A semi-structured overlay for multi-attribute range queries in cloud computing | |
| Fan et al. | GA_Chord: An improvement to Chord algorithm based on group autonomy in structed P2P network | |
| Mokadem et al. | An efficient resource discovery while minimizing maintenance overhead in sdds based hierarchical dht systems | |
| Raju et al. | Reputation and trust management models in peer-to-peer | |
| Mashayekhi et al. | Efficient semantic based search in unstructured peer-to-peer networks | |
| Zhao et al. | An improved chord based on counting bloom filter and topology-aware lookup | |
| Mizutani et al. | MARIF: Multiple queries look-up architecture using range information feedback in a DHT network | |
| Yu et al. | Improving the lookup performance of Chord network by hashing landmark clusters |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120704 Termination date: 20160428 |