CN101399765B - Method and system for reducing hot node load in P2P network - Google Patents
Method and system for reducing hot node load in P2P network Download PDFInfo
- Publication number
- CN101399765B CN101399765B CN2007101752941A CN200710175294A CN101399765B CN 101399765 B CN101399765 B CN 101399765B CN 2007101752941 A CN2007101752941 A CN 2007101752941A CN 200710175294 A CN200710175294 A CN 200710175294A CN 101399765 B CN101399765 B CN 101399765B
- Authority
- CN
- China
- Prior art keywords
- node
- key assignments
- hot
- peer
- memory space
- 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
- Computer And Data Communications (AREA)
Abstract
The invention discloses a method for reducing the hot node load in the P2P network and a system thereof, which belongs to the P2P network field. The method comprises that: the ID of a node and the memory space of key assignments are expanded; a free node is selected as a backup node for the hot node in the P2P network, and an ID value is distributed to the free node in the expanded ID space to be added in the P2P network; the key assignments of the stored resources in the hot node and the key assignments index information are backed up on the backup node; the key assignments are converted to new key assignments in the expanded memory space when being searched; the node of the new key assignments is searched and maintained; the obtained node converts the new key assignments to the corresponding key assignments in the expanded memory space; the converted key assignments and the key assignments index information are returned. The system comprises management equipment and searching nodes. The invention reduces the hot node load in the P2P network and fully utilizes the free resources in the network; the routing overhead and the searched nodes are not increased.
Description
Technical field
The present invention relates to the peer-to-peer network field, particularly a kind of method and system that reduces hot node load in the peer-to-peer network.
Background technology
Can there be some popular data in the network, as pop music, popular film, hotline etc.When these resources when issuing in structural P 2 P (Peer to Peer, the equity) network, form one (key, value) right, leave in and have on the ID value node more close with this key value.Wherein, the key value is the mark value of resource, and value is the index information of this key assignments.For popular resource, can cause a lot of nodes clicks in the network or search this resource, thereby make it become hot point resource, correspondingly, deposit hot point resource (key, value) right node need be handled too much search request and work overloadingly, and becomes hot node.For example, referring to Fig. 1, in structured P 2 P network, key1 is the key assignments of certain hot point resource, and its value is 2, and be kept on the node 2, a lot of nodes in the network are all searched key1 as node 1,4,5,7 to node 2, cause node 2 to handle query manipulation and the excess load running, become hot node, may cause node 2 paralysis when situation is serious.
Available technology adopting Cache (backup) method solves the problem of hot node excess load, hot node is handled query requests and during overload operation, (the key of hot point resource, value) on the node that backups on other nodes, backup to usually most possible process when searching this index information and doing route.For example, in the Chord algorithm, on the node that continues before this information is placed on; Earlier on the routed path of searching near the node of hot node on the index information of a hot point resource of backup, if can't solve hot issue, just look for a node to back up this index information forward again, till solving this hot issue along the opposite direction of this routed path; Like this, just might in the cache that a lot of nodes are arranged on the routed path shows, all preserve this index information,,, then return index information if in the table resource index information is arranged so when node is searched this hot point resource,, all detect the cache table one time in each jumping; If do not have in the table, then continue to search according to routing table.Referring to Fig. 2, in the Chord algorithm, when node 3 was searched key1 and is key assignments 2, the node of searching process may be node 7 and node 1, finds key assignments 2 at last on node 2.According to searching specifically, then node 2 thinks that node 3, node 7 and node 1 all are the nodes of route process when searching, when key assignments key1 is hot point resource, node 2 just backups to (the key value) of this key assignments on node 1, node 7 and the node 3 priority, on these nodes, form cache table separately, up to solving hot issue.When in network, finding key value key1, when (or node 7 or node 3), search the cache table on the node 1 through node 1, therefrom find this key assignments, then directly return this key assignments (key value) to information, thereby has alleviated the burden of node 2.
In realizing process of the present invention, the inventor finds that above-mentioned prior art has following shortcoming at least:
1. to each node of route path process, before searching routing table, all need to search earlier the cache table, changed DHT (Distributed Hash Table, distributed hashtable) and searched algorithm, but also can increase certain expense.
The hot point resource information back-up to the node of route process, because when searching this resource information, the starting point difference, the node of corresponding route process might be different, can cause solving the hot issue that in the whole network searching of this hot point resource information is caused fully; If when the node processing ability that backs up is not high, backup node may can become new hot node again, can introduce new hot issue again.
3. because the node that initiation is searched and the node of route process do not know all whether the resource that will search is hot point resource, all to search the cache table, when the resource information of searching is not hot point resource, bring delay searching so each is jumped.
Summary of the invention
In order to reduce the load of hot node in the P2P network, the embodiment of the invention provides a kind of method and system that reduces hot node load in the peer-to-peer network.Described technical scheme is as follows:
On the one hand, a kind of method that reduces hot node load in the peer-to-peer network, described method comprises:
An expansion figure place is set; When searching algorithm and be prefix matching or approximate match mode, increase described expansion figure place respectively at the low portion of the ID of node memory space and key assignments memory space; When searching algorithm and be the suffix match mode, increase described expansion figure place respectively at the high-order portion of the ID of node memory space and key assignments memory space;
With the ID value transform of all nodes in the peer-to-peer network is ID value in the ID memory space after the described expansion, and all key assignments are transformed to key assignments in the key assignments memory space after the described expansion;
Select the backup node of idle node as the hot node in the peer-to-peer network, to be in the ID value between the ID value after the conversion of neighbor node of ID value after the conversion of described hot node and described hot node, distribute to described backup node, and described backup node joined in the described peer-to-peer network, the key assignments and the key assignments index information of described hot node stored resource backuped on the described backup node;
When searching the key assignments of described hot node stored resource, the key assignments of described hot node stored resource is transformed to an interval in the key assignments memory space after being in described expansion, and in the described interval optional key assignments as new key assignments; According to described new key assignments, search the node of safeguarding this new key assignments;
The described node that finds is transformed to described new key assignments key assignments corresponding in the key assignments memory space of key assignments after described expansion of described hot node stored resource, and returns key assignments and key assignments index information after the described conversion.
On the other hand, a kind of system that reduces hot node load in the peer-to-peer network, described system comprise management equipment and search node;
Described management equipment comprises:
Expansion module is used to be provided with an expansion figure place, when searching algorithm and be prefix matching or approximate match mode, increases described expansion figure place respectively at the low portion of the ID of node memory space and key assignments memory space; When searching algorithm and be the suffix match mode, increase described expansion figure place respectively at the high-order portion of the ID of node memory space and key assignments memory space;
Conversion module is used for ID value transform with all nodes of peer-to-peer network and is the ID value in the ID memory space after the described expansion module expansion, and all key assignments are transformed to the interior key assignments of key assignments memory space after the described expansion module expansion;
Backup module, be used for selecting the backup node of idle node as the hot node of peer-to-peer network, to be in the ID value between the ID value after the conversion of neighbor node of ID value after the conversion of described hot node and described hot node, distribute to described backup node, and described backup node joined in the described peer-to-peer network, the key assignments and the key assignments index information of described hot node stored resource backuped on the described backup node;
The described node of searching comprises:
The initial seek module, be used for when described when searching node and searching the key assignments of described hot node stored resource as start node, the key assignments of described hot node stored resource is transformed to an interval in the key assignments memory space after being in described expansion, and in described interval, choose a key assignments wantonly, and the information of searching to described new key assignments is transmitted to next node according to routing table as new key assignments;
Module is searched in the centre, is used for when described when searching node as intermediate node, receive that a node sends to the information of searching of new key assignments to be found and be transmitted to next node;
Finish to search module, be used for when described when searching node as the finish node searched, receive the information of searching that a node is sent to new key assignments to be found, described new key assignments is transformed to key assignments corresponding in the key assignments memory space of key assignments after described expansion of described hot node stored resource, and judge the key assignments that whether has in the described key assignments of searching node storage after the described conversion, if have, then return key assignments and key assignments index information after the described conversion; Otherwise, return the information of searching failure.
Technique scheme is by enlarging ID memory space and key assignments space, choose idle node as backup node, backup hot node stored resource information, and by the key assignments conversion, can be implemented in and find this resource information on backup node or the hot node, thereby reduced the load of hot node in the P2P network, played the effect of load sharing, and made full use of the idling-resource in the network; Owing to do not change the DHT algorithm, in the process of searching, except that routing table, need not to search other list item, do not increase routing cost, search the node of process can not increase yet.
Description of drawings
Fig. 1 is the work schematic diagram of hot node in the P2P network in the prior art;
Fig. 2 is the schematic diagram that backup solves hot node excess load problem in the prior art;
Fig. 3 is the method flow diagram of hot node load in the reduction peer-to-peer network that provides of the embodiment of the invention;
Fig. 4 is the schematic diagram of the embodiment of the invention at low level expanding node ID memory space;
Fig. 5 is the schematic diagram of the embodiment of the invention at high-order expanding node ID memory space;
Fig. 6 is node ID memory space expansion back ID value and a key value variation schematic diagram in the embodiment of the invention;
Fig. 7 is a schematic diagram of introducing backup node among Fig. 6 behind node ID value and the key value transform;
Fig. 8 is the routing table schematic diagram of interdependent node before backup node adds among Fig. 7;
Fig. 9 is the routing table schematic diagram that backup node adds the back interdependent node among Fig. 7;
Figure 10 is the system construction drawing of hot node load in the reduction peer-to-peer network that provides of the embodiment of the invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, embodiment of the present invention is described further in detail below in conjunction with accompanying drawing.
The embodiment of the invention is by choosing the backup node of idle node as the hot node in the P2P network, for it distributes new ID value, and backup hot node stored resource information, and by the key assignments conversion, can be implemented in and find hot point resource information on backup node or the hot node, thereby, solved the hot issue in the structured P 2 P network for hot node has played the effect of load sharing.
Referring to Fig. 3, the embodiment of the invention provides a kind of method that reduces hot node load in the peer-to-peer network, specifically comprises the steps:
Step 101: the ID memory space and the key assignments memory space of node in the expansion P2P network;
Wherein, during expansion the ID memory space is done identical expansion with the key assignments memory space, can specifically comprise:
An expansion figure place is set,, when searching algorithm and be prefix matching or approximate match mode, increases the m position respectively at the low portion of the ID of node memory space and key assignments memory space as the m position; When searching algorithm and be the suffix match mode, increase the m position respectively at the high-order portion of the ID of node memory space and key assignments memory space.
For example,, search in the prefix matching or approximate match mode of algorithm,, promptly the m position that increases newly is placed on the low portion of former ID memory space as suffix in the low portion increase m position of ID memory space at DHT referring to Fig. 4; Referring to Fig. 5, search in the suffix match mode of algorithm at DHT, in the high-order portion increase m position of ID memory space, promptly the m position that increases newly is placed on the high-order portion in former ID space as prefix.The ID memory space of supposing node is 160 bits (binary systems), as can be seen from Figure 4 and Figure 5, and total 160+m the bit of the ID memory space of expansion posterior nodal point.Wherein the m position of Zeng Jiaing can be provided with as required, as long as the ID value scope after the expansion that guarantees is not conflicted with former ID value scope, for example, can all be set to 0 to extension bits, or all be set to 1, or partly be set to 0 part and be set to 1 or the like.Expansion treatment process for the key assignments memory space is identical with the expansion treatment process of ID memory space, does not repeat them here.
Step 102:, and all key assignments in the P2P network are transformed to key assignments in the key assignments memory space after the expansion with the ID value in the ID value transform of all nodes in the P2P network ID memory space after for expansion;
Former ID value (or key assignments) can be converted to binary representation during conversion, and according to low level or the high-order increase new bit of the figure place of expanding in the step 101 in binary system ID value (or key assignments), the value of the bit that setting increases newly, binary system ID value (or key assignments) after will expanding then is converted to decimal number, the node ID value (or key assignments) after being expanded.
With the Chord algorithm is that example describes, because the Chord algorithm is to belong to the prefix matching mode, so after low level increased the m position, the node ID value after the conversion was 2 of a former ID value
mDoubly, promptly the ID value is to left m position; Also enlarge 2 when correspondingly, key assignments is issued
mDoubly.Referring to Fig. 6,8 nodes are arranged in the P2P network, be respectively node 0, node 1, node 2 ..., node 7, suppose m=3, promptly increase by 3, and 3 of expansion are set are 0 at the low level of ID memory space, then the variation of node ID value is as shown in table 1 before and after the conversion.Correspondingly, all key assignments in the P2P network are also done same conversion, as the key assignments key=2 of node 2, are key=16 after the conversion.
Table 1
Step 103: select idle node, as the backup node of the hot node in the P2P network;
Wherein, idle node can be existing idle node in the P2P network, also can be the outer idle node of P2P network.
Step 104: be backup node distribution ID value in the ID memory space after expansion, and backup node is joined in the P2P network;
Wherein, when distributing ID value,, the ID value that is between the ID value of neighbor node of the ID value of hot node and hot node can be distributed to backup node owing to the ID value of two adjacent after conversion nodes is discontinuous for backup node.
For example, referring to Fig. 7, be the P2P network structure behind node ID value among Fig. 6 and the key value transform, suppose that hot node is a node 16, the node that continues before it is a node 8, and have following ID value can utilize between node 8 and the node 16 this moment: 9,10,11,12,13,14 and 15; Suppose that existing two idle node are node 40 and node 48 in the P2P network, choose their backup nodes as hot node 16, and from above-mentioned utilizable ID value scope, appoint and get two ID values and distribute to them, as distributing the ID value for node 48 is 9, distributing the ID value for node 40 is 10, and then newly producing two backup nodes is node 9 and node 10.
Adding in the process of backup node, can change the information in the routing table on the node relevant, as the successor value (beginning the node ID number of first existence from interval starting point) among the change finger table with backup node to the P2P network.
Be example with the employing Chord algorithm network among Fig. 7 still, suppose that ID value 24 and 32 also do not distribute to node and be registered in the network as ID, promptly the descendant node of node 16 is a node 40.Node 9 and node 10 added in the P2P networks, correspondingly, can cause hot node before the continue routing table of node change, be 16 as the descendant node of original node 8, the descendant node of node 8 is nodes 9 now.Referring to Fig. 8, for before backup node node 9 and node 10 add the P2P networks, the schematic diagram of the finger table of node 8 and node 16 (routing table a kind of).Referring to Fig. 9, after node 9 and node 10 add the P2P network, the finger table of node 8 changes, and the finger table of node 16 does not change, wherein, and in the list item of node 8, ID value scope [9,10) Dui Ying successor is changed to 9, and scope [10,12) corresponding successor is changed to 10.
Step 105: with the key assignments and the key assignments index information of stored resource on the hot node in the P2P network, promptly (key, value) right, backup on the backup node;
For example, referring to Fig. 9, with on the hot node 16 (key is value) to being kept on backup node 9 and the node 10.
Step 106: when the some nodes in the P2P network are searched above-mentioned key assignments, this key assignments is transformed to a new key assignments in the key assignments memory space after being in expansion;
Wherein, the step of conversion can be specific as follows:
This key assignments is transformed to an interval in the key assignments memory space after being in expansion, and in this interval an optional key assignments as new key assignments.
The mode of the key assignments of hot point resource being carried out conversion has multiple, can calculate according to formula of expansion figure place definition, for example, referring to Fig. 9, node 40 is searched the key assignments of key=2, if ID memory space and key assignments memory space are m in the figure place of low level expansion, then are defined as follows formula and carry out conversion: ((key-1) * 2
m, key * 2
m], being about to the key value transform is an interval shown in the formula, the interval after the conversion is (8,16], this interval is in the key assignments memory space after the expansion, comprises 9,10,11,12,13,14,15 and 16, in this interval, choose a key assignments wantonly, as 12, as new key assignments.
Step 107:, search the node of safeguarding this new key assignments according to the new key assignments that conversion obtains;
The step of wherein, searching can be specific as follows:
From the node of initiating to search, search the new key assignments after the conversion, search according to routing table, up to finding the node of safeguarding this key assignments; This node can be a backup node, also can be hot node.
For example, referring to Fig. 9, suppose that the new key assignments of selecting is 9, in the Chord algorithm, begin to search from start node, when finding node 8, in the finger of node 8 table table, find the key assignments interval [9 at new key assignments 9 places, 10), the succ of this interval correspondence (being the successor value) is 9, and promptly this interval is in charge of and is safeguarded to node 9, therefore node 8 will mail to this inquiry node 9 processing, promptly finds and safeguards this interval node 9.
Step 108: on the node that finds above-mentioned new key assignments is transformed to the key assignments of correspondence in the key assignments memory space of key assignments after expansion original to be found, and is returning key assignments and key assignments index information after the conversion, promptly (key, value) right, finish search procedure.
For example, suppose the figure place that m increases for the low level at former ID memory space, the new key assignments of selecting is key=9, on node 9, calculates (key/2 according to formula of m definition
m+ 1) * 2
m, wherein, key/2
mExpression key divides exactly 2
mIf m is 3, then calculating the result according to above-mentioned formula is 16, be about to new key assignments 9 and be transformed to key assignments 16 corresponding in the key assignments memory space of key assignments 2 after expansion original to be found, and return key assignments be 16 (key is value) to information.
In order to improve the accuracy of the information of searching, further, in said method, can also increase the information of backup node backup is carried out updating steps:
When the key assignments of hot node stored resource and key assignments index information change, key assignments and key assignments index information after changing are updated on the backup node, thereby (key, value) the right consistency that can keep the preservation on hot node and the backup node.
Further, the step below in said method, can also increasing:
Judge hot node overwork whether still, promptly the focus phenomenon is not also eliminated, if, then continue in the P2P network, to increase new backup node, back up the key assignments and the key assignments index information of stored resource on the hot node, eliminate the hot node operate as normal up to hot spot phenomenon.
Wherein, the operation of judging hot node can regularly be initiated according to the default time.
In order to improve the efficient of searching and to save idling-resource in the network, further, the step below in said method, can also increasing:
When hot node recovers operate as normal, when promptly the key assignments deposited of focus node no longer is the focus key assignments, the backup node of the above-mentioned adding of deletion in the P2P network; Further, can also regain the ID value of distributing into this backup node.
Present embodiment is by enlarging ID memory space and key assignments space, choose idle node as backup node, backup hot node stored resource information, and by the key assignments conversion, can be implemented in and find this resource information on backup node or the hot node, thereby reduced the load of hot node in the P2P network, played the effect of load sharing, and made full use of the idling-resource in the network; Owing to do not change the DHT algorithm, in the process of searching, except that routing table, need not to search other list item, do not increase routing cost, search the node of process can not increase yet.By upgrading key assignments and the key assignments index information on the backup node, can improve the accuracy of the information of searching.By continuous increase backup node, hot spot phenomenon is eliminated, make hot node recover operate as normal.ID value by deletion backup node when hot node recovers operate as normal and cancellation distribute can reduce backup node
The burden of (being original idle node), and reduce the scale of network, and improve the efficient of searching.
Referring to Figure 10, the embodiment of the invention also provides a kind of system that reduces hot node load in the peer-to-peer network, and system comprises management equipment and searches node;
Management equipment comprises:
(1) expansion module is used for the ID memory space and the key assignments memory space of expanding node;
(2) backup module, be used for selecting the backup node of idle node as the hot node of peer-to-peer network, distribute the ID value for backup node in the ID memory space after expansion, and backup node joined in the peer-to-peer network, the key assignments and the key assignments index information of hot node stored resource backuped on the backup node;
Searching node comprises:
(1) initial seek module is used for when searching node and find key value as start node, and key assignments is transformed to a new key assignments in the key assignments memory space after being in expansion, and according to routing table the information of searching to new key assignments is transmitted to next node;
(2) search module in the middle of, be used for when searching node as intermediate node, receive that a node sends to the information of searching of new key assignments to be found and be transmitted to next node;
(3) finish to search module, be used for when searching node as the finish node searched, receive the information of searching that a node is sent to new key assignments to be found, this new key assignments is transformed to key assignments corresponding in the key assignments to be found key assignments memory space after expansion, and judge the key assignments that whether has in the key assignments search the node storage after the conversion, if have, then return key assignments and key assignments index information after the conversion; Otherwise, return the information of searching failure.
Wherein, expansion module can specifically comprise:
1) expanding element is used to be provided with an expansion figure place, when searching algorithm and be prefix matching or approximate match mode, increases the expansion figure place respectively at the low portion of the ID of node memory space and key assignments memory space; When searching algorithm and be the suffix match mode, increase the expansion figure place respectively at the high-order portion of the ID of node memory space and key assignments memory space;
2) converter unit is used for ID value transform with all nodes of peer-to-peer network and is the ID value in the ID memory space after the expanding element expansion, and all key assignments are transformed to the interior key assignments of key assignments memory space after the expanding element expansion.
Further, said system can also comprise:
Conversion module is used for ID value transform with all nodes of peer-to-peer network and is the ID value in the ID memory space after the expansion module expansion, and all key assignments are transformed to the interior key assignments of key assignments memory space after the expansion module expansion.
Wherein, backup module can specifically comprise:
1) initialization unit, be used for selecting the backup node of idle node as the hot node of peer-to-peer network, to be in the ID value between the ID value after the conversion of neighbor node of ID value after the conversion of hot node and hot node, distribute to backup node, and backup node is joined in the peer-to-peer network;
2) backup units is used for the key assignments and the key assignments index information of hot node stored resource are backuped to the backup node that initialization unit is selected.
Wherein, the initial seek module can specifically comprise:
1) converter unit is used for when searching node and find key value as start node, and key assignments is transformed to an interval in the key assignments memory space after being in expansion;
2) selected cell is used for the new key assignments of interval interior optional key assignments conduct that obtains from converter unit;
3) search the unit, be used for the information of searching of the new key assignments that selected cell is selected being transmitted to next node according to routing table.
In order to improve the accuracy of the information of searching, further, management equipment can also comprise:
Update module is used for when the key assignments of hot node stored resource and key assignments index information change key assignments and key assignments index information after changing being updated on the backup node.
Further, management equipment can also comprise:
Whether still monitoring module is used to judge hot node overwork, if then notify backup module to begin to increase new backup node, up to the hot node operate as normal.
Wherein, monitoring module can regularly be initiated monitoring according to the default time, judges whether hot node works overloadingly.
In order to improve efficient of searching and the idling-resource of saving network, further, management equipment can also comprise:
Backup cancellation module is used for when hot node recovers operate as normal, the backup node that the deletion backup module is selected in peer-to-peer network.
Present embodiment enlarges ID memory space and key assignments space by expansion module, and backup module is chosen idle node as backup node, backup hot node stored resource information, and search module and end by initial seek module, centre and search twice pair of key assignments of module and carry out conversion, can be implemented in and find this resource information on backup node or the hot node, thereby reduced the load of hot node in the P2P network, play the effect of load sharing, and made full use of the idling-resource in the network; Owing to do not change the DHT algorithm, in the process of searching, except that routing table, need not to search other list item, do not increase routing cost, search the node of process can not increase yet.Key assignments and key assignments index information by on the update module renewal backup node can improve the accuracy of searching.Constantly increase backup node by monitoring module control backup module, hot spot phenomenon is eliminated, make hot node recover operate as normal.When hot node recovers operate as normal, delete backup node and regain the ID value of distributing by backup cancellation module, can reduce the burden of backup node (being original idle node), and reduce the scale of network, improve the efficient of searching.
The embodiment of the invention all is suitable in structuring and non-structural P 2 P system.
The embodiment of the invention can utilize software to realize, corresponding software programs can be stored in the storage medium that can read, in the hard disk and buffer memory as router.
The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being done, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1. a method that reduces hot node load in the peer-to-peer network is characterized in that, described method comprises:
An expansion figure place is set; When searching algorithm and be prefix matching or approximate match mode, increase described expansion figure place respectively at the low portion of the ID of node memory space and key assignments memory space; When searching algorithm and be the suffix match mode, increase described expansion figure place respectively at the high-order portion of the ID of node memory space and key assignments memory space;
With the ID value transform of all nodes in the peer-to-peer network is ID value in the ID memory space after the described expansion, and all key assignments are transformed to key assignments in the key assignments memory space after the described expansion;
Select the backup node of idle node as the hot node in the peer-to-peer network, to be in the ID value between the ID value after the conversion of neighbor node of ID value after the conversion of described hot node and described hot node, distribute to described backup node, and described backup node joined in the described peer-to-peer network, the key assignments and the key assignments index information of described hot node stored resource backuped on the described backup node;
When searching the key assignments of described hot node stored resource, the key assignments of described hot node stored resource is transformed to an interval in the key assignments memory space after being in described expansion, and in the described interval optional key assignments as new key assignments; According to described new key assignments, search the node of safeguarding this new key assignments;
The described node that finds is transformed to described new key assignments key assignments corresponding in the key assignments memory space of key assignments after described expansion of described hot node stored resource, and returns key assignments and key assignments index information after the described conversion.
2. the method for hot node load is characterized in that in the reduction peer-to-peer network according to claim 1, and described method also comprises:
When the key assignments of the described resource of described hot node storage and key assignments index information change, key assignments after the described variation and key assignments index information are updated on the described backup node.
3. the method for hot node load is characterized in that in the reduction peer-to-peer network according to claim 1, and described method also comprises:
Judge the whether still overwork of described hot node, if then increase new backup node and back up described key assignments and key assignments index information, up to described hot node operate as normal.
4. the method for hot node load is characterized in that in the reduction peer-to-peer network according to claim 3, and described method also comprises:
When described hot node recovers operate as normal, the described backup node of deletion in described peer-to-peer network.
5. a system that reduces hot node load in the peer-to-peer network is characterized in that, described system comprises management equipment and searches node;
Described management equipment comprises:
Expansion module is used to be provided with an expansion figure place, when searching algorithm and be prefix matching or approximate match mode, increases described expansion figure place respectively at the low portion of the ID of node memory space and key assignments memory space; When searching algorithm and be the suffix match mode, increase described expansion figure place respectively at the high-order portion of the ID of node memory space and key assignments memory space;
Conversion module is used for ID value transform with all nodes of peer-to-peer network and is the ID value in the ID memory space after the described expansion module expansion, and all key assignments are transformed to the interior key assignments of key assignments memory space after the described expansion module expansion;
Backup module, be used for selecting the backup node of idle node as the hot node of peer-to-peer network, to be in the ID value between the ID value after the conversion of neighbor node of ID value after the conversion of described hot node and described hot node, distribute to described backup node, and described backup node joined in the described peer-to-peer network, the key assignments and the key assignments index information of described hot node stored resource backuped on the described backup node;
The described node of searching comprises:
The initial seek module, be used for when described when searching node and searching the key assignments of described hot node stored resource as start node, the key assignments of described hot node stored resource is transformed to an interval in the key assignments memory space after being in described expansion, and in described interval, choose a key assignments wantonly, and the information of searching to described new key assignments is transmitted to next node according to routing table as new key assignments;
Module is searched in the centre, is used for when described when searching node as intermediate node, receive that a node sends to the information of searching of new key assignments to be found and be transmitted to next node;
Finish to search module, be used for when described when searching node as the finish node searched, receive the information of searching that a node is sent to new key assignments to be found, described new key assignments is transformed to key assignments corresponding in the key assignments memory space of key assignments after described expansion of described hot node stored resource, and judge the key assignments that whether has in the described key assignments of searching node storage after the described conversion, if have, then return key assignments and key assignments index information after the described conversion; Otherwise, return the information of searching failure.
6. the system of hot node load is characterized in that in the reduction peer-to-peer network according to claim 5, and described management equipment also comprises:
Update module is used for when the key assignments of the described resource of described hot node storage and key assignments index information change key assignments after the described variation and key assignments index information being updated on the described backup node.
7. the system of hot node load is characterized in that in the reduction peer-to-peer network according to claim 5, and described management equipment also comprises:
Whether still monitoring module is used to judge the overwork of described hot node, if then notify described backup module to begin to increase new backup node, up to described hot node operate as normal.
8. the system of hot node load is characterized in that in the reduction peer-to-peer network according to claim 5, and described management equipment also comprises:
Backup cancellation module is used for when described hot node recovers operate as normal, the backup node that the described backup module of deletion is selected in described peer-to-peer network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2007101752941A CN101399765B (en) | 2007-09-28 | 2007-09-28 | Method and system for reducing hot node load in P2P network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2007101752941A CN101399765B (en) | 2007-09-28 | 2007-09-28 | Method and system for reducing hot node load in P2P network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101399765A CN101399765A (en) | 2009-04-01 |
| CN101399765B true CN101399765B (en) | 2011-04-13 |
Family
ID=40518033
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2007101752941A Expired - Fee Related CN101399765B (en) | 2007-09-28 | 2007-09-28 | Method and system for reducing hot node load in P2P network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101399765B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101710901B (en) * | 2009-10-22 | 2012-12-05 | 乐视网信息技术（北京）股份有限公司 | Distributed type storage system having p2p function and method thereof |
| CN102244670B (en) * | 2010-05-13 | 2013-11-06 | 北京大学 | Idle node assistance method for P2P (peer-to-peer) file transmission |
| CN102739762B (en) * | 2011-04-08 | 2017-03-29 | 中兴通讯股份有限公司 | The method and system of Data Migration during a kind of peer network node overload |
| CN102739711B (en) * | 2011-04-08 | 2017-10-17 | 中兴通讯股份有限公司 | The method and system of overload are controlled in a kind of peer-to-peer network |
| CN102891872B (en) * | 2011-07-20 | 2018-02-09 | 中兴通讯股份有限公司 | The method and system of data storage and query in a kind of peer-to-peer network |
| CN103841027B (en) * | 2012-11-26 | 2018-08-14 | 中兴通讯股份有限公司 | The map information of Hash mapping plane is registered and lookup method, equipment and system |
| CN104683422B (en) * | 2013-12-03 | 2019-01-29 | 腾讯科技（深圳）有限公司 | Data transmission method and device |
| CN106165355A (en) * | 2014-01-31 | 2016-11-23 | 交互数字专利控股公司 | For the methods, devices and systems by realizing network association based on the peerings of hash route and/or summary route |
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 |
| CN1972289A (en) * | 2006-12-01 | 2007-05-30 | 华中科技大学 | Method for suppressing free rider of nodes in peer-to-peer network |
-
2007
- 2007-09-28 CN CN2007101752941A patent/CN101399765B/en not_active Expired - Fee Related
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 |
| CN1972289A (en) * | 2006-12-01 | 2007-05-30 | 华中科技大学 | Method for suppressing free rider of nodes in peer-to-peer network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101399765A (en) | 2009-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101399765B (en) | Method and system for reducing hot node load in P2P network | |
| CN106599308B (en) | distributed metadata management method and system | |
| CN100496004C (en) | Structured peer-to-peer network system and its load query, transfer and resource seeking method | |
| CN101697526A (en) | Method and system for load balancing of metadata management in distributed file system | |
| JP2014044677A (en) | Transmission control program, communication node, and transmission control method | |
| CN104077423A (en) | Consistent hash based structural data storage, inquiry and migration method | |
| CN106775446A (en) | Based on the distributed file system small documents access method that solid state hard disc accelerates | |
| CN110417883B (en) | Design method of point-to-point network structure applied to block chain | |
| Goodrich et al. | The rainbow skip graph: a fault-tolerant constant-degree distributed data structure | |
| CN102664914A (en) | IS/DFS-Image distributed file storage query system | |
| CN103902735A (en) | Application perception data routing method oriented to large-scale cluster deduplication and system | |
| CN102045392A (en) | Interest-based adaptive topology optimization method for unstructured P2P (peer-to-peer) network | |
| CN104111924A (en) | Database system | |
| CN104184812A (en) | Multi-point data transmission method based on private cloud | |
| CN101645831A (en) | Node organization method in P2P system | |
| Xu et al. | Adaptive and scalable load balancing for metadata server cluster in cloud-scale file systems | |
| WO2011000209A1 (en) | Server, method and system for providing node information for p2p network | |
| Yu et al. | Granary: A sharing oriented distributed storage system | |
| CN107547657A (en) | A kind of method, apparatus and storage medium numbered based on one point data in cloud storage system | |
| Guirat et al. | An efficient data replication approach for structured peer-to-peer systems | |
| Marx et al. | A decentralized architecture for SPARQL query processing and RDF sharing: a position paper | |
| Zhuang et al. | Design of a more scalable database system | |
| CN103701846A (en) | Peer-to-peer network resource search optimization method based on node proximity estimation | |
| Li et al. | Efficient R-tree based indexing for cloud storage system with dual-port servers | |
| Sathya et al. | Replication strategies for data grids |
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: 20110413 Termination date: 20210928 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |