CN101729390B - Distributed Hash Table (DHT) network and construction method and nodes thereof - Google Patents
Distributed Hash Table (DHT) network and construction method and nodes thereof Download PDFInfo
- Publication number
- CN101729390B CN101729390B CN2008102246528A CN200810224652A CN101729390B CN 101729390 B CN101729390 B CN 101729390B CN 2008102246528 A CN2008102246528 A CN 2008102246528A CN 200810224652 A CN200810224652 A CN 200810224652A CN 101729390 B CN101729390 B CN 101729390B
- Authority
- CN
- China
- Prior art keywords
- network
- node
- stand
- state
- adds
- 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
Links
Images
Abstract
The invention discloses a construction method of Distributed Hash Table (DHT) network. The method comprises the steps of receiving a request for adding the nodes out of the network into the network; and adding the nodes out of the network into the network or setting in standby state according to the number of the nodes added in the network and the number of the nodes in standby state. The invention also discloses the nodes in the DHT network and the DHT network. By adopting the invention, the impact suffered in network construction can be reduced and the stability of the network can be enhanced.
Description
Technical field
The present invention relates to communication technical field, especially relate to distributed hashtable DHT network and construction method thereof, node.
Background technology
Distributed hashtable (DHT, Distributed Hash Table) is generally used for Distributed Storage.In the DHT network, each network node realizes that the storage that homogenizes, each node be responsible for that a route is among a small circle handled and the storage of partial data (data).That is to say that the DHT network is formed data link, thereby reached resource-sharing and the purpose of communicating by letter with physical link work station that each is isolated or host interconnection together.
The basic principle of DHT network is: at first; Each data resource Object table is shown as a data resource object directory entry (K; V) right; K is called keyword, is the cryptographic hash of file description information (like filename, reference number of a document etc.), and V is the descriptor (like IP address, title etc.) of the node of actual storage file.All data resource object indexing clauses and subclauses (promptly all (K; V) to) form a file index Hash table; As long as the K value of input file destination just can be found all address of node of storing this document or other descriptors from this document index Hash table.Then, this document index Hash table is divided into a lot of sections fritters, is distributed to these local Hash tables on all the participation nodes in the system, make each node be responsible for safeguarding a local Hash table wherein according to rules specific.Like this, during the querying node file, the node of storage this document that this node is searched promptly: contain in the Hash table piecemeal of being safeguarded to search (then after finding, the node that query message is routed to this storage this document gets final product for K, V) right node.
The practical implementation method of DHT algorithm has a variety of, like KAD, Chord etc., below brief account Chord algorithm.
The Chord algorithm is structural P 2 P (Point to Point, the point-to-point) agreement that is proposed by MIT (Massachu-setts Institute of Technology, the Massachusetts Institute of Technology).Each node and keyword have unique ID (sign), and wherein node ID is the mapping to node address, and keyword ID is the mapping to this keyword.
The ID spatial organization that is made up of the ID sign is an ID ring, so when ID length was m, the span of ID was [0～2m-1].Keyword k is stored in and prolongs on first node ID corresponding nodes of ID ring clockwise direction, and this node is called the descendant node (successor node) of keyword k.Fig. 1 has shown that an ID length is 3 chord ID ring and storage.In Fig. 1, can see that having ID in the system is 0,1,3 totally 3 nodes; ID is 1,2,6 totally 3 piece of data.Then according to above-described storage principle; Be about to storage in the node of node ID more than or equal to data ID; The node that node ID is minimum; Then data 1 should be stored on the node 1, and data 2 should be stored on the node 3, and data 6 should be stored in (clockwise direction is searched node successively on the ID ring) on the node 0.
Keyword is accomplished through the consistency hash function to the mapping of ID.(With high probability w.h.p.) is assigned to the keyword of similar number to each node in the system, when N node adds system, has only the keyword of O (1/N) to move simultaneously with very high probability.
Node in the Chord algorithm is only preserved the address information of part of nodes, and these address informations are called routing iinformation.Through effectively organizing routing iinformation and based on this resource location algorithm, the resource location only needs O (logN) to jump, and each node is preserved the information of other nodes of O (logN), simultaneously, and the adding of node and leave and need O (log
2N) message, wherein N is an interstitial content.
The inventor is through analyzing, and it is following not enough to find that above-mentioned prior art exists:
1, in the DHT network; Each node is responsible for the storage of the data in a part of Hash space separately, when therefore new node adds network, if add great deal of nodes simultaneously; Then data migration repeatedly between a plurality of newly added nodes probably; The routing table of other nodes also needs to bring in constant renewal in the network, and network is impacted, and influences network stabilization.
2, in the DHT network; Each node is responsible for the storage of the data in a part of Hash space separately, in case so certain node off-line, the load on this node will be transferred to descendant node; May cause the excessive off-line then of load on the descendant node, even cause the malignant event of the whole network node off-line.
Concrete, when node off-line, its descendant node is transferred in the load meeting of this node thereupon, if former data quantity stored is bigger on the off-line node, then descendant node possibly can't bear the load capacity of increase, then because of the excessive off-line of load; Even more serious is; If the node average load in the network exceeds certain load capacity threshold value; This descendant node all is transferred to other node with the load of withdrawing from node; May cause other node because of the excessive off-line then of load, this pernicious load transfer causes the whole network node off-line of domino formula probably.
Summary of the invention
The embodiment of the invention provides a kind of construction method of distributed hashtable DHT network, in order to reduce the impact to network, improves network stabilization, and this method comprises:
Receive the request that the outer node of said network adds said network;
According to interstitial content that adds said network and the interstitial content that is in stand-by state, the node that said network is outer adds said network or is set to stand-by state.
The embodiment of the invention also provides the node in a kind of DHT network, in order to reduce the impact to network, improves network stabilization, comprising:
Receiver module is used to receive the request that the outer node of said network adds said network;
Add processing module, be used for according to interstitial content that adds said network and the interstitial content that is in stand-by state, the node that said network is outer adds said network or is set to stand-by state.
The embodiment of the invention also provides the node in a kind of DHT network, in order to reduce the impact to network, improves network stabilization, comprising:
Sending module is used to initiate to add the request of said network; And, after receiving the indication that adds said network again, initiate to add the request of said network again;
Receiver module is used for after being set to stand-by state, receives the indication that adds said network again.
The embodiment of the invention also provides a kind of DHT network, in order to reduce the impact to network, improves network stabilization, comprising:
Service node comprises the sending module that is used to initiate add the request of said network;
Management node comprises being used to receive the receiver module that said service node adds the request of said network; And, be used for according to interstitial content that has added said network and the interstitial content that is in stand-by state, said service node is added said network or is set to the adding processing module of stand-by state.
In the embodiment of the invention, receive the request that the outer node of said network adds said network, according to interstitial content that adds said network and the interstitial content that is in stand-by state; The node that said network is outer adds said network or is set to stand-by state; Make that the adding of node is controlled, the disposable adding network of great deal of nodes can not occur, the phenomenon that causes data between newly added node, to move back and forth; Can reduce the impact that network receives, strengthen stability of network.
Description of drawings
Fig. 1 is that ID length is 3 chord ID loop data storage sketch map in the background technology;
Fig. 2 is the construction method flow chart of distributed hashtable DHT network in the embodiment of the invention;
Fig. 3 is the centralized resources pond formula DHT network diagram in the embodiment of the invention;
Fig. 4 is the distributed resource pond formula DHT network diagram in the embodiment of the invention;
Fig. 5, Fig. 7, Figure 10 and Figure 11 are the structural representations of the node in the DHT network in the embodiment of the invention;
Fig. 6 is the structural representation of the adding processing module in the embodiment of the invention;
Fig. 8, Fig. 9 are the structural representations of the processed offline module in the embodiment of the invention.
Embodiment
Great deal of nodes adds network simultaneously in the prior art in order to solve; Network is impacted, influence the problem of stability of network, the embodiment of the invention provides a kind of solution: make up a kind of new DHT network; For these network settings are in the node of stand-by state, also can be described as secondary node.Before making up the DHT network, the interstitial content that has added said network is zero with the interstitial content that is in stand-by state.In the process that makes up the DHT network; Request with the adding network that receives the outer node initiation of network; After receiving this request, this node is added network or is set to stand-by state, it is according to being: the interstitial content and the interstitial content that is in stand-by state that have added said network.Through aforesaid node adding method, can make up the DHT network described in the ultra embodiment of the invention gradually.
Below in conjunction with Figure of description the embodiment of the invention is elaborated.
As shown in Figure 2, in the embodiment of the invention, the construction method of distributed hashtable DHT network comprises:
In the flow process shown in Figure 2, step 202 can comprise when implementing:
The interstitial content that adds said network and the ratio value and the preset proportion value that are in the interstitial content of stand-by state are compared;
With the ratio value of interstitial content that is in stand-by state during less than the preset proportion value, the node that said network is outer adds said network at the interstitial content that adds said network;
When the interstitial content that adds said network was not less than the preset proportion value with the ratio value that is in the interstitial content of stand-by state, the outer node of said network was set to stand-by state.
The value of preset proportion described in the enforcement can for example be considered high reliability according to the performance index requirements setting, can the node ratio that be in stand-by state be heightened; Consider high-performance, can the node ratio that add network be heightened.
Lift an example; The preset proportion value of the interstitial content of supposing to add the interstitial content of said network and be in stand-by state is between 4:1 to 5:1, and after the node outside receiving network added the request of network, the interstitial content of finding to have added said network was 10:1 with the ratio value that is in the interstitial content of stand-by state; That is to say that the shared ratio of the node that is in stand-by state is on the low side; This possibly be owing to once big accident, has added the node off-line of network in a large number, and the node that causes being in stand-by state is by a large amount of requisitions; Then this moment, can be set to stand-by state by the outer node of said network; And if the interstitial content that has added said network this moment is 3:1 with the ratio value that is in the interstitial content of stand-by state; That is to say that the node that is in stand-by state is too much; This possibly be owing to there is a large amount of nodes to recover from status of fail; In this case, node that can said network is outer adds network.In the enforcement, when the node that said network is outer adds network, can be this node distribution node sign, and indicate this node to add network by the node identification that distributes; The node that is in stand-by state is not assigned with node identification, does not participate in storage, is in holding state, prepares to add network at any time.
After having made up the DHT network by flow process shown in Figure 2; Each node is responsible for the storage of a part of data separately, when possible situation occurring: when in the network node off-line being arranged, for avoid load transfer on the off-line node to descendant node, may cause the situation of descendant node because of the excessive off-line of load; And then avoid the malignant event of the whole network node off-line; In the embodiment of the invention, select to be in the node of stand-by state, promptly aforementioned secondary node; Add said network, take over the legacy data of this off-line node of off-line node storage.When confirming in the said network node off-line to be arranged, the node of selecting to be in stand-by state adds said network, the legacy data of storage off-line node.That is, make the node that is in stand-by state of selection take over the work of off-line node.
It is multiple to confirm to have in the network method of node off-line to have in the enforcement, and in the DHT network, the off-line of node can be by network aware, the descendant node of the regular Ping of each node oneself for example, perhaps, by the regular detection node survival condition of management node etc.
The selection mode that is in the node of stand-by state in the enforcement can have multiple, for example: in being in the node of stand-by state, select node to add said network at random; Can certainly select according to strategy, for example: in being in the node of stand-by state, the node of selecting to be set to stand-by state at first adds said network.In network, can store node state tabulation, can comprise node that has added network and the relevant information that is in the node of stand-by state in the tabulation, like the IP address etc.Follow-up this tabulation capable of using confirms to have added the interstitial content and the ratio value that is in the interstitial content of stand-by state of said network, thus the adding of Control Node.
Among the embodiment; After having selected to be in the node of stand-by state; Can distribute the original identification information of off-line node for the node of selecting that is in stand-by state, indicate the node that is in stand-by state of said selection to add said network by said identification information.
After distributing the original identification information of off-line node for the node of selecting that is in stand-by state, the node that is in stand-by state of said selection can obtain the legacy data and the storage of off-line node according to said identification information from the backup node of off-line node.
In the flow process shown in Figure 2, can be by node implementation step that adds the DHT network 201 and step 202; For example by a node implementation step 201, by another node implementation step 202; Can certainly among the embodiment below,, describe only by a node implementation step 201 and step 202, this node is called the management node in the network for convenient only to specify by a node implementation step 201 and step 202.
Management node in the network can be one, also can be a plurality of, is used for the adding and the off-line of network node are handled.Definite method of management node can have multiple, for example can directly be specified by network, also can elect through network node.Whole network can be regarded as a resource pool, by a plurality of node common storage resources that distribute.If the management node in the network is one, then can be referred to as centralized resources pond DHT network, its structure can be as shown in Figure 3; If the management node in the network is a plurality of, then can be referred to as distributed resource pond DHT network, its structure can be as shown in Figure 4.
Among the embodiment; Can network be divided into a plurality of zones, division methods has many kinds, for example can divide according to the prefix of node identification; With the sign prefix be 000000 node ownership to zone one, be that 000001 node ownership is second-class to the zone with the sign prefix; A management node is set in each zone, and each management node is responsible for the network struction in the one's respective area.
Can be in the enforcement according to strategy, making between a plurality of management nodes can be alternately or can not be mutual, if can be mutual, the then management node node state of delivery network stored tabulation each other regularly, thus realize management to the whole network node; Certainly if consider the simplicity of system, also can be provided with can not be mutual between a plurality of management nodes.
The example that is configured to distributed hashtable DHT network; In this example; The node that has added network distributes and data allocations according to adopting the DHT algorithm to carry out node identification, with 128 as Hash space size, use SHA-128 as basic hash function; Suppose with first six digits as regional code, then prefix 000000 corresponding region 1; Prefix 000001 corresponding region 2; Prefix 000010 corresponding region 3; Prefix 000011 corresponding region 4; Prefix 000100 corresponding region 5; By that analogy to 32 zones.
In the enforcement management node of each area configurations (like the configuration management IP addresses of nodes etc., and notice is given other node in the zone).
When the outer node application of network adds network; Report application with joining request and add the management node in the zone, if the preset interstitial content that adds network is 4:1 with the ratio value that is in the interstitial content of stand-by state, if in this moment network; The node that has added network is 7:1 with the ratio that is in the node of stand-by state; That is, the node proportion that is in stand-by state is on the low side, and then the outer node of the said network of management node is set to stand-by state.And if the node that has added network this moment is 3:1 with the ratio that is in the node of stand-by state, that is, the node that is in stand-by state is too much, and then the management node node that said network is outer adds said network.
After the outer node of said network is allowed to add network, can obtain node identification according to the method for classical DHT, log-on data is moved process.
The practical implementation method of DHT algorithm has a variety of, like KAD, Chord etc., is example with the Chord algorithm in this example, and the regular Ping's of the node in the network oneself is follow-up, when having perceived node off-line, notifies the management node in its affiliated area.Management node can select a node that is in stand-by state to add network according to the node state tabulation of storing in the network, can select to be set at first the node of stand-by state, also can select certain to be in the node of stand-by state at random.Management node distributes the original identification information of off-line node for the node of selecting that is in stand-by state, indicates selected node to add network by this identification information.
The said selecteed node that is in stand-by state is obtained the legacy data and the storage of off-line node according to identification information from the backup node of off-line node.Do not have load on the initiate node, so its this process of startup is to have no excessive pressure for physical equipment.
The construction method of the construction method of centralized resources pond DHT network and above-mentioned distributed hashtable DHT network is similar, repeats no more here.
Based on same inventive concept, the embodiment of the invention also provides the node in a kind of DHT network, and its structure is as shown in Figure 5, can comprise:
Add processing module 502, be used for according to interstitial content that adds said network and the interstitial content that is in stand-by state, the node that said network is outer adds said network or is set to stand-by state.
As shown in Figure 6, among the embodiment, adding processing module 502 can comprise:
Add processing sub 602, when being used for ratio value at the interstitial content that adds said network and the interstitial content that is in stand-by state less than the preset proportion value, the node that said network is outer adds said network;
The subsequent use submodule 603 that is provided with is used for when the interstitial content that adds said network is not less than the preset proportion value with the ratio value that is in the interstitial content of stand-by state, and the outer node of said network is set to stand-by state.
As shown in Figure 7, among the embodiment, the node in the DHT network shown in Figure 5 can also comprise:
Processed offline module 701 is used for when definite said network has node off-line, and the node of selecting to be in stand-by state adds said network, the legacy data of storage off-line node.
As shown in Figure 8, among the embodiment, processed offline module 701 can comprise:
The first chooser module 801 is used for being in the node of stand-by state, and the node of selecting to be set to stand-by state at first adds said network;
Or the second chooser module 802 is used for being in the node of stand-by state, selects node to add said network at random.
As shown in Figure 9, among the embodiment, processed offline module 701 can comprise:
Based on same inventive concept, the present invention implements the node in a kind of DHT network also is provided, and its structure is shown in figure 10, can comprise:
Sending module 1001 is used to initiate to add the request of said network; And, after receiving the indication that adds said network again, initiate to add the request of said network again;
Among the embodiment, receiver module 1002 can also receive original identification information of distributing to off-line node in the said network when receiving the indication that adds said network again.At this moment, shown in figure 11, the node in the DHT network shown in Figure 10 can also comprise:
Based on same inventive concept, the embodiment of the invention also provides a kind of DHT network, can comprise:
Service node comprises the sending module that is used to initiate add the request of said network;
Management node comprises being used to receive the receiver module that said service node adds the request of said network; And, be used for according to interstitial content that has added said network and the interstitial content that is in stand-by state, said service node is added said network or is set to the adding processing module of stand-by state.
Among the embodiment, said management node is a plurality of, is arranged at the zones of different in the said network respectively; Adding processing module in each management node adds service node the one's respective area or is set to stand-by state according to added interstitial content in the one's respective area and the interstitial content that is in stand-by state.
In the above-mentioned DHT network; Service node can be in different conditions: as be in the state that initiation adds the request of said network; After management node added processing, service node can be in state that has added said network or the stand-by state that does not add said network.Management node can be for one or more, if management node is one, then with centralized resources pond DHT network class shown in Figure 3 seemingly; If there are a plurality of management nodes, then with distributed resource pond DHT network class shown in Figure 4 seemingly.Wherein, Fig. 3, Fig. 4 show the node that adds said network and the relation of management node.
In the embodiment of the invention, receive the request that the outer node of said network adds said network, according to interstitial content that adds said network and the interstitial content that is in stand-by state; The node that said network is outer adds said network or is set to stand-by state; Make that the adding of node is controlled, the disposable adding network of great deal of nodes can not occur, the phenomenon that causes data between newly added node, to move back and forth; Can reduce the impact that network receives, strengthen stability of network.
When in network, node off-line being arranged; Can select a node that is in stand-by state to take over the off-line node; Load on the off-line node can not transferred to descendant node; Descendant node can not avoid occurring in the network possibility of the pernicious off-line of great deal of nodes because of transshipping off-line then, has further strengthened stability of network.
Obviously, those skilled in the art can carry out various changes and distortion to the present invention and not break away from the spirit and scope of the present invention.Like this, if these modifications of the present invention and distortion belong within claim of the present invention and the equivalent technologies scope thereof, then the present invention also is intended to comprise these changes and is out of shape interior.
Claims (12)
1. the construction method of a distributed hashtable DHT network is characterized in that, this method comprises:
Receive the request that the outer node of said network adds said network;
The interstitial content that adds said network and the ratio value and the preset proportion value that are in the interstitial content of stand-by state are compared;
With the ratio value of interstitial content that is in stand-by state during less than the preset proportion value, the node that said network is outer adds said network at the interstitial content that adds said network;
When the interstitial content that adds said network was not less than the preset proportion value with the ratio value that is in the interstitial content of stand-by state, the outer node of said network was set to stand-by state.
2. the method for claim 1 is characterized in that, this method further comprises:
When confirming in the said network node off-line to be arranged, the node of selecting to be in stand-by state adds said network, the legacy data of storage off-line node.
3. method as claimed in claim 2 is characterized in that, the node of selecting to be in stand-by state adds said network and comprises:
In being in the node of stand-by state, the node of selecting to be set to stand-by state at first adds said network;
Or, in being in the node of stand-by state, select node to add said network at random.
4. method as claimed in claim 2 is characterized in that, the node of selecting to be in stand-by state adds said network and comprises:
For the node of selecting that is in stand-by state distributes the original identification information of off-line node, indicate the node that is in stand-by state of said selection to add said network by said identification information.
5. method as claimed in claim 4 is characterized in that, the legacy data of said storage off-line node comprises:
The node that is in stand-by state of said selection obtains the legacy data and the storage of off-line node according to said identification information from the backup node of off-line node.
6. the node in the DHT network is characterized in that, comprising:
Receiver module is used to receive the request that the outer node of said network adds said network;
Add processing module; Be used for according to interstitial content that adds said network and the interstitial content that is in stand-by state; The node that said network is outer adds said network or is set to stand-by state; Wherein said adding processing module comprises: comparison sub-module is used for the interstitial content that adds said network and the ratio value and the preset proportion value that are in the interstitial content of stand-by state are compared; Add processing sub, when being used for ratio value at the interstitial content that adds said network and the interstitial content that is in stand-by state less than the preset proportion value, the node that said network is outer adds said network; The subsequent use submodule that is provided with is used for when the interstitial content that adds said network is not less than the preset proportion value with the ratio value that is in the interstitial content of stand-by state, and the outer node of said network is set to stand-by state.
7. node as claimed in claim 6 is characterized in that, also comprises:
The processed offline module is used for when definite said network has node off-line, and the node of selecting to be in stand-by state adds said network, the legacy data of storage off-line node.
8. node as claimed in claim 7 is characterized in that, said processed offline module comprises:
The first chooser module is used for being in the node of stand-by state, and the node of selecting to be set to stand-by state at first adds said network;
Or the second chooser module is used for being in the node of stand-by state, selects node to add said network at random.
9. node as claimed in claim 7 is characterized in that, said processed offline module comprises:
Distribution sub module, the node that is in stand-by state that is used to selection distributes the original identification information of off-line node;
The indication submodule is used to indicate the node that is in stand-by state of said selection to add said network by said identification information.
10. the node that the DHT network is outer is characterized in that, comprising:
Sending module is used to initiate to add the request of said network; And, after receiving the indication that adds said network again, initiate to add the request of said network again;
Receiver module is used for after being set to stand-by state, receives the indication that adds said network again, receives original identification information of distributing to off-line node in the said network;
Acquisition module is used for obtaining the legacy data of off-line node from the backup node of off-line node according to said identification information;
Memory module is used to store the data of obtaining;
Wherein, after the request of the said network of said adding was received by the node in the DHT network, the interstitial content that the node in the DHT network will add said network compared with the ratio value and the preset proportion value that are in the interstitial content of stand-by state; With the ratio value of interstitial content that is in stand-by state during less than the preset proportion value, the node that said network is outer adds said network at the interstitial content that adds said network; When the interstitial content that adds said network is not less than the preset proportion value with the ratio value that is in the interstitial content of stand-by state; The outer node of said network is set to stand-by state; When in confirming network, node off-line being arranged; Selection is in the node of stand-by state, for the node of selecting that is in stand-by state distributes the original identification information of off-line node, indicates the node that is in stand-by state of said selection to add said network again according to said identification information.
11. a DHT network is characterized in that, comprising:
Service node comprises the sending module that is used to initiate add the request of said network;
Management node comprises being used to receive the receiver module that said service node adds the request of said network; And; Be used for according to interstitial content that adds said network and the interstitial content that is in stand-by state; Said service node is added said network or is set to the adding processing module of stand-by state; Wherein, the said adding processing module interstitial content that will add said network and the ratio value and the preset proportion value that are in the interstitial content of stand-by state compares; During less than the preset proportion value, said service node is added said network with the ratio value of interstitial content that is in stand-by state at the interstitial content that adds said network; When the interstitial content that adds said network was not less than the preset proportion value with the ratio value that is in the interstitial content of stand-by state, said service node was set to stand-by state.
12. DHT network as claimed in claim 11 is characterized in that, said management node is a plurality of, is arranged at the zones of different in the said network respectively; Adding processing module in each management node adds said service node the one's respective area or is set to stand-by state according to added interstitial content in the one's respective area and the interstitial content that is in stand-by state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102246528A CN101729390B (en) | 2008-10-22 | 2008-10-22 | Distributed Hash Table (DHT) network and construction method and nodes thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102246528A CN101729390B (en) | 2008-10-22 | 2008-10-22 | Distributed Hash Table (DHT) network and construction method and nodes thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101729390A CN101729390A (en) | 2010-06-09 |
| CN101729390B true CN101729390B (en) | 2012-11-07 |
Family
ID=42449637
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008102246528A Active CN101729390B (en) | 2008-10-22 | 2008-10-22 | Distributed Hash Table (DHT) network and construction method and nodes thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101729390B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103209136A (en) * | 2012-01-11 | 2013-07-17 | 中兴通讯股份有限公司 | Network load control method and register server |
| CN104077423B (en) * | 2014-07-23 | 2017-05-03 | 山东大学（威海） | Consistent hash based structural data storage, inquiry and migration method |
| CN109218126B (en) * | 2017-06-30 | 2023-10-17 | 中兴通讯股份有限公司 | Method, device and system for monitoring node survival state |
| CN107734008A (en) * | 2017-09-27 | 2018-02-23 | 柏科数据技术（深圳）股份有限公司 | Method, apparatus, node device and the storage medium of a kind of troubleshooting in data-storage system |
| CN111126571B (en) * | 2019-12-25 | 2023-07-14 | 福建天晴数码有限公司 | R-CNN network optimization method based on DHT network and storage medium |
| CN111064798B (en) * | 2019-12-25 | 2022-10-21 | 深圳市网心科技有限公司 | Node network access method, device, client node and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1691619A (en) * | 2004-04-27 | 2005-11-02 | 国家数字交换系统工程技术研究中心 | Method for implementing self-organizing network |
| CN101286920A (en) * | 2008-05-15 | 2008-10-15 | 华中科技大学 | Expandable resource lookup method in peer-to-peer network with structure |
-
2008
- 2008-10-22 CN CN2008102246528A patent/CN101729390B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1691619A (en) * | 2004-04-27 | 2005-11-02 | 国家数字交换系统工程技术研究中心 | Method for implementing self-organizing network |
| CN101286920A (en) * | 2008-05-15 | 2008-10-15 | 华中科技大学 | Expandable resource lookup method in peer-to-peer network with structure |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101729390A (en) | 2010-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101729390B (en) | Distributed Hash Table (DHT) network and construction method and nodes thereof | |
| KR100537581B1 (en) | Master node selection in clustered node configurations | |
| CN102006184B (en) | Management method, device and network device of stack link | |
| CN104572727A (en) | Data querying method and device | |
| CN107547391B (en) | Message transmission method and device | |
| CN107040393B (en) | Route management method and equipment | |
| CN103326925B (en) | A kind of information push method and device | |
| CN102934412B (en) | Server cluster | |
| CN102035861A (en) | Load balancing method and system of single-jump distributing type hash table superposition network | |
| CN103441932A (en) | Host routing table entry generating method and device | |
| JPH01503269A (en) | data communication network | |
| CN102571936B (en) | Method, device and system for searching data | |
| CN103179009B (en) | A kind of dynamic self-adapting call method of distributed management system | |
| CN202870563U (en) | Distributed comprehensive monitoring system | |
| CN101354720A (en) | Distributed memory database data system and sharing method thereof | |
| US11463376B2 (en) | Resource distribution method and apparatus in Internet of Things, device, and storage medium | |
| CN107547339A (en) | A kind of gateway media access control MAC address feedback method and device | |
| CN101902388A (en) | Expandable fast discovery technology for multi-stage sequencing resources | |
| CN101963978A (en) | Distributed database management method, device and system | |
| CN105243078A (en) | File resource distribution method, system and apparatus | |
| CN107046503B (en) | Message transmission method, system and device | |
| CN1905516B (en) | Method of allocating bidirection forwarding detecting conversational authentication code | |
| CN102868775A (en) | Method for expanding capacity of address resolution protocol table, and message forwarding method and device | |
| CN112001800B (en) | Method and device for processing business in block chain system | |
| CN101420352B (en) | Method and system for key value transition in structured P2P network |
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 |