JP2005210461A - Method for dynamically controlling information distribution channel by independently distributedly controlled node - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a dynamic control method of an information distribution path capable of optimizing the distribution channel of contents in a virtual network comprising an independently distributedly controlled node. <P>SOLUTION: In the dynamic control method of the information distribution channel in the independently distributedly controlled node for maintaining connection to a certain number of nodes, on the basis of an adjacent node list for indicating a connected node and a candidate node list for indicating a node connected to the adjacent node, reciprocation time required when data for confirming connection reciprocate are measured for each adjacent connected node in a measurement step. In a sorting step, the adjacent node list is sorted, on the basis of the reciprocation time obtained by measurement for each adjacent node. In a cutting step, the connection with a certain number of adjacent nodes is selectively cut from longer corresponding reciprocation time, based on the adjacent node list obtained as the sorted result. In a switching step, a certain number of candidate nodes are selected from the candidate node list for connection instead of the node in which the connection is cut. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

The present invention relates to an information network for forming an appropriate information distribution topology by route control independent of routes in a hierarchy below the IP network in an overlay network composed of autonomous distributed control nodes located in an upper layer of the IP network. The present invention relates to a route dynamic control method.
For example, when the content requesting node and the node that owns the content belong to networks provided by different Internet service providers, the request is made across networks provided by a plurality of Internet service providers. It is necessary to set an information distribution route that connects the original node and the content owning node.

In a virtual network composed of autonomous distributed control nodes as a mechanism for setting an information distribution path that connects a requesting node and a content-owning node across networks provided by a plurality of Internet service providers, A method of dynamically forming a distribution channel has been proposed (see Non-Patent Document 1).
FIG. 16 shows a configuration example of a virtual network composed of autonomous distributed control nodes.

  Communication processing between the autonomous distributed control node (node) 410 and other nodes shown in FIG. 16 is realized by the function of the communication processing unit 412, and a communication path to the communication processing unit 412 is set. A list of adjacent nodes is held in the adjacent node list 413. Further, information regarding the nodes connected to each of the above-described adjacent nodes is held in the candidate node list 414, and the adjacent node management unit 415 is held in the candidate node list 414 and the above-described adjacent node list 413. Based on the information, the node 410 is managed so as to always maintain a connection with a certain number of other nodes.

FIG. 17 is a flowchart showing the adjacent node management process.
The adjacent node management unit 415 shown in FIG. 16 reads, for example, a list of adjacent nodes held in the adjacent node list 413 at regular intervals (step 501), and in the order listed in this list, A connection confirmation message for requesting a list of other nodes connected to the node is transmitted using the function of the communication processing unit 412 (step 502).

  When the connection response message is received within a certain period of time after transmitting the above-described connection confirmation message (Yes in step 503), the adjacent node management unit 415 includes the connection response received via the communication processing unit 412. Each node indicated in the node list is added to the candidate node list 414 as a connectable node candidate (step 504). Thereafter, the adjacent node management unit 415 determines whether or not the confirmation processing for all adjacent nodes has been completed (step 505). If the determination in step 505 is negative, the process returns to step 502 to return to the next adjacent node. Select.

  On the other hand, if the connection response message is not received even after a predetermined time has elapsed since the connection confirmation message was transmitted (No determination in step 503), the adjacent node management unit 415 transmits the connection confirmation message in step 502. It is determined that the connection with the adjacent node has been disconnected, and information regarding this node is deleted from the adjacent node list 413 (step 506). Next, in order to establish a connection with a new node, the adjacent node management unit 415 first refers to the candidate node list 414 and reads, for example, the node information held at the head of the candidate node list 414 ( Step 507), based on this information, a connection with the corresponding node is established using the function of the communication processing unit 412 (Step 508). After that, the adjacent node management unit 415 deletes the information of the candidate node read out in step 507 from the candidate node list 414 (step 509), and instead stores information related to the node in the adjacent node list 413 (step 510). ), Returning to step 502, the connection confirmation process for the new adjacent node is performed.

  In this way, the connection confirmation process for all adjacent nodes held in the adjacent node list 413 is completed, and the above process is repeated until an affirmative determination is made in step 505, whereby all the connections connected to the node 410 are performed. Information on nodes adjacent to each other from the adjacent nodes is collected in the candidate node list 414. When the communication path to the adjacent nodes is disconnected, new information is collected based on the information collected in the candidate node list 414. It is possible to control the communication path to be connected to a node.

  On the other hand, when the connection confirmation message is received via the communication processing unit 412, the adjacent node management unit 415 creates a connection response message including a list of adjacent nodes held in the adjacent node list 413, and this connection The response message is transferred to the communication processing unit 412 and sent to the node that has transmitted the connection confirmation message. In addition, when receiving a request for establishing a communication path from another node via the communication processing unit 412, the adjacent node management unit 415 adds information on the requesting node to the adjacent node list 413.

In addition, the content acquisition control unit 416 illustrated in FIG. 16 executes processing for acquiring the requested content from other nodes belonging to the virtual network in response to a content acquisition request from the user.
FIG. 18 is a flowchart showing the content acquisition operation.
When receiving the content acquisition request from the user (step 521), the content acquisition control unit 416 shown in FIG. 16 first reads a list of adjacent nodes held in the adjacent node list 414 (step 522). Next, the content acquisition control unit 416 uses the function of the communication processing unit 412 to make all adjacent nodes included in the read list unique search IDs in the network (for example, a combination of a node ID and the number of searches). At the same time, search information for requesting the search for the content requested by the content acquisition request is transmitted (step 523). Thereafter, when the communication processing unit 412 receives the content search result together with the above-described search ID (step 524), the content acquisition control unit 416 determines that the search result for the content requested by the communication processing unit 412 has arrived. The incoming content is received from the communication processing unit 412 and provided to the user (step 525), and the process ends.

  On the other hand, the content list 417 shown in FIG. 16 holds a list of contents owned by the node 410, and the search ID list 418 transmits the search ID received together with the search information relayed by the node 410. It is stored corresponding to the node ID of the original adjacent node. Then, the content relay control unit 419 refers to the content list 417 and the search ID list 418 described above in accordance with the search information input via the adjacent node, and searches the search information and the content based on the information held therein. Provides a function to relay

FIG. 19 is a flowchart showing a part of the operation of relaying search information and contents. FIG. 20 is a flowchart showing another part of the operation of relaying search information and content.
The content relay control unit 419 shown in FIG. 16 first refers to the content list 417 (step 532) when the communication processing unit 412 receives search information from an adjacent node (affirmative determination in step 531). It is determined whether or not the content specified by the search information is owned (step 533).

  When the designated content is owned (Yes in step 533), the content relay control unit 419 includes the content search result indicating that the content is owned together with the search ID received in step 531. Using the function of the communication processing unit 412, the search information is sent to the adjacent node that is the transmission source (step 534), and then the corresponding content is sent to the adjacent node that is the search information transmission source together with the search ID described above. (Step 535), the content transmission process is terminated.

  On the other hand, when the specified content is not owned (No determination at step 533), the content relay control unit 419 first records the search information and the search ID in association with the node ID indicating the transmission source node. Is added to the search ID list 418 (step 537). Next, a list of adjacent nodes is read with reference to the adjacent node list 414 (step 538), and using the function of the communication processing unit 412, all adjacent nodes other than the node that is the search information transmission source are set as destinations. The search information and search ID received in step 531 are sent out (step 539), and the search information relay process is terminated.

On the other hand, if the determination in step 531 is negative, the content relay control unit 419 executes each step shown in FIG.
When the communication processing unit 412 receives a content search result together with a search ID including a node ID other than the node ID indicating itself from an adjacent node (affirmative determination in step 540), the content relay control unit 419 is based on the search ID. The node ID of the adjacent node that is the transmission source of the corresponding search information is acquired from the search ID list 418 as a destination to relay the content search result described above (step 541), and the function of the communication processing unit 412 is used. The search ID and content search result described above are sent to the adjacent node indicated by the node ID acquired in step 538 (step 542), and the relay process is terminated.

Further, when the communication processing unit 412 receives the content from the adjacent node together with the search ID including the node ID other than the node ID indicating itself (affirmative determination in step 543), the content relay control unit 419 is based on the search ID. Then, the node ID of the adjacent node that is the transmission source of the corresponding search information is acquired from the search ID list 418 as a destination to which the above-mentioned content is to be relayed (step 544), and the function of the communication processing unit 412 is used. The above-described search ID and content are sent to the adjacent node indicated by the node ID acquired at 544 (step 545). Thereafter, the content relay control unit 419 deletes the record held in the search ID list 418 regarding the search ID corresponding to the content relayed as described above (step 546), and ends the content relay processing. In the case of negative determination in step 543, appropriate error processing is performed (step 547).
Kamei Kaoru, Mori Tatsuya, Kimura Takumi “P2P Traffic Separation Method and Its Evaluation”, “2003 IEICE Communication Society Conference”, page 200, B-7-19 Tokutaka Hiroshi, Arakawa Shinichi, Murata Masayuki, "Evaluation of Distributed Routing Control Methods in Optical Path Networks", [online], February 2003, [January 20, 2004], Internet <URL: http: // www.anarg.jp/~t-toku/files/soturon.pdf>

  In the virtual network composed of the above-described conventional autonomous distributed control nodes, each autonomous distributed control node further collects information on adjacent nodes connected to the respective adjacent nodes in a candidate node list, and When a connection is required, a connection destination is selected from the candidate node list, thereby maintaining a connection with a certain number of nodes and forming an information distribution route in the virtual network.

  However, in the above-described virtual network composed of autonomous distributed control nodes, the connection with the adjacent node is maintained as long as the connection is physically possible regardless of whether the connection has an advantageous condition in the distribution of contents. In addition, when a new connection destination is selected from the candidate node list, it is not considered whether the connection with the candidate node satisfies an advantageous condition in the distribution of contents.

  On the other hand, as shown in FIGS. 18, 19 and 20, in various processes related to content acquisition, content search information, content search results, and the content itself are relayed from each node to its adjacent nodes. Depending on what characteristics each node constituting the virtual network is connected to, the performance exhibited by the virtual network in the distribution of content varies greatly.

In other words, in a virtual network where each node does not have an appropriate adjacent node, it may take a long time to transfer the content, or the number of nodes through which the content is acquired may increase. Conceivable.
An object of the present invention is to provide a dynamic control method of an information distribution route that can optimize a content distribution route in a virtual network including autonomous distributed control nodes.

A first information distribution route dynamic control method according to the present invention includes a measurement step, a rearrangement step, a cutting step, and a switching step.
The principle of the first dynamic control method of the information distribution route according to the present invention is as follows.
Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution path in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination, the measurement step is performed for each adjacent node being connected. When confirming, the round trip time required for the connection confirmation data to reciprocate with each adjacent node is measured. The rearrangement step rearranges the adjacent node list based on the round trip time obtained by the measurement for each adjacent node. The disconnecting step selectively disconnects the connection with the predetermined number of adjacent nodes from the longer corresponding round trip time based on the adjacent node list obtained as a result of the rearrangement. The switching step selects a predetermined number of candidate nodes from the candidate node list and connects them instead of the disconnected nodes.

The operation of the dynamic control method for the first information distribution path configured as described above is as follows.
For example, when the connection confirmation with the adjacent node is performed at regular time intervals, the adjacent node list is rearranged in the rearrangement step based on the round trip time measured in the measurement step. For example, the adjacent node list is rearranged in the order of short round trip time. In this way, the connection between a predetermined number of nodes located at the end of the rearranged adjacent node list is selectively disconnected, and instead, the transmission path between the same number of nodes listed in the candidate node list After a certain time, the adjacent node list is rearranged again based on the measurement result obtained in the measurement step, and the node with a long round trip time is switched to the candidate node.

By repeating such an operation, it is possible to disconnect a node having an unfavorable condition in forming an information distribution route that it takes a long time to reciprocate connection confirmation information, and to connect a node having an advantageous condition. Can be maintained.
The second information distribution route dynamic control method according to the present invention includes a counting step, a rearranging step, a cutting step, and a switching step.

The principle of the second method for dynamically controlling the information distribution route according to the present invention is as follows.
Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination, the counting step is performed for each adjacent node being connected, respectively. The number of times content is acquired from the adjacent node is counted. The rearrangement step rearranges the adjacent node list based on the number of acquisition times obtained for each adjacent node. In the disconnecting step, based on the adjacent node list obtained as a result of the rearrangement, the connection with a predetermined number of adjacent nodes is selectively disconnected from the one having a smaller number of acquisitions. The switching step selects a predetermined number of candidate nodes from the candidate node list and connects them instead of the disconnected nodes.

The operation of the dynamic control method for the second information distribution route configured as described above is as follows.
For example, when the content is acquired via the adjacent node, the counting step adds the number of acquisitions corresponding to the adjacent node that is the transmission source of the content by a numerical value “1”, thereby obtaining the number of acquisitions corresponding to each adjacent node. Counting is performed, and the counting result is held in the adjacent node list corresponding to the identification information of the adjacent node. In this way, the rearrangement step rearranges the adjacent node list that is updated each time content is acquired, for example, in the order of the number of acquisitions in the course of the connection confirmation process performed at regular time intervals. In the disconnection step, the connection with the predetermined number of nodes located at the end of the rearranged adjacent node list is selectively disconnected in the disconnection step. Instead, the same number of candidates listed in the candidate node list in the switching step. Set up a transmission path between nodes. In addition, after a certain time, the adjacent node list is rearranged based on the number of acquisitions corresponding to each adjacent node updated every time the content is acquired by the counting step, and the node with the small number of acquisitions is switched to the candidate node. .

By repeating such operations, the node having a disadvantageous condition in forming an information distribution route is separated, that is, the number of times of content acquisition is small, and conversely, the advantageous condition that the number of times of content acquisition is large is provided. Connection with the node can be maintained.
The third information distribution route dynamic control method according to the present invention includes a counting step, a measurement step, an evaluation step, a rearranging step, a cutting step, and a switching step.

The principle of the third method for dynamically controlling the information distribution route according to the present invention is as follows.
Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination, the counting step is performed for each adjacent node being connected, respectively. The number of times content is acquired from the adjacent node is counted. In the measurement step, when the connection state is confirmed for each of the adjacent nodes being connected, the round trip time required for the connection confirmation data to reciprocate with each adjacent node is measured. The evaluation step evaluates the usefulness of the connection with each adjacent node based on the number of acquisitions and the round trip time obtained for each adjacent node. The rearrangement step rearranges the adjacent node list based on the usefulness obtained for each adjacent node. The disconnecting step selectively disconnects the connection with a predetermined number of adjacent nodes because the corresponding usefulness is low based on the adjacent node list obtained as a result of the rearrangement. The switching step selects a predetermined number of candidate nodes from the candidate node list and connects them instead of the disconnected nodes.

The operation of the third information distribution route dynamic control method configured as described above is as follows.
For example, when the content is acquired via the adjacent node, the counting step adds the number of acquisitions corresponding to the adjacent node that is the transmission source of the content by a numerical value “1”, thereby obtaining the number of acquisitions corresponding to each adjacent node. Counting is performed, and the counting result is held in the adjacent node list corresponding to the identification information of the adjacent node. In this way, each time content is acquired, the number of acquisitions corresponding to each adjacent node included in the adjacent node list is updated, and further, for example, in the process of connection confirmation processing performed every predetermined time, connection confirmation is performed by a measurement step. Measure the time it takes for information to travel back and forth. Then, in the connection confirmation process, the adjacent node list is rearranged in descending order of usefulness in the rearrangement step based on the usefulness obtained for each adjacent node in the evaluation step. In the disconnection step, the connection with the predetermined number of nodes located at the end of the rearranged adjacent node list is selectively disconnected in the disconnection step. Instead, the same number of candidates listed in the candidate node list in the switching step. Set up a transmission path between nodes. In addition, after a certain period of time, the connection with each adjacent node is again performed based on the number of acquisitions corresponding to each adjacent node updated every time content is acquired by the counting step and the round trip time obtained in the measurement step. The usefulness is evaluated, and the adjacent node list is rearranged based on the obtained usefulness, and a node having a low usefulness is switched to a candidate node.

By repeating such operations, nodes with low usefulness obtained by comprehensively evaluating the number of content acquisitions and round trip times can be disconnected, and connections with nodes with high usefulness can be maintained. .
A fourth information distribution route dynamic control method according to the present invention includes a counting step, a measuring step, a ranking step, a rearranging step, a cutting step, and a switching step.

The principle of the fourth dynamic control method of the information distribution route according to the present invention is as follows.
Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination, the counting step is performed for each adjacent node being connected, respectively. The number of times content is acquired from the adjacent node is counted. In the measurement step, when the connection state is confirmed for each of the adjacent nodes being connected, the round trip time required for the connection confirmation data to reciprocate with each adjacent node is measured. In the ranking step, a rank related to the number of acquisitions and a rank related to the round-trip time are assigned to each adjacent node included in the adjacent node list based on the number of times of acquisition and round-trip time obtained for each adjacent node. The disconnecting step selectively disconnects a connection between an adjacent node whose score based on the ranking given regarding the number of acquisitions is equal to or less than a predetermined threshold and an adjacent node whose score based on the rank related to round trip time is equal to or less than the predetermined threshold. In the switching step, the number of candidate nodes that are disconnected in the disconnecting step is selected from the candidate node list, and the connection is made instead of the disconnected node.

The operation of the fourth information distribution route dynamic control method configured as described above is as follows.
For example, when the content is acquired via the adjacent node, the counting step adds the number of acquisitions corresponding to the adjacent node that is the transmission source of the content by a numerical value “1”, thereby obtaining the number of acquisitions corresponding to each adjacent node. Counting is performed, and the counting result is held in the adjacent node list corresponding to the identification information of the adjacent node. In this way, each time content is acquired, the number of acquisitions corresponding to each adjacent node included in the adjacent node list is updated, and further, for example, in the process of connection confirmation processing performed every predetermined time, connection confirmation is performed by a measurement step. Measure the time it takes for information to travel back and forth. Then, in the process of the connection confirmation process, the ranking step assigns ranks for the number of times of content acquisition and round trip time for each adjacent node, and the score corresponding to each rank is less than a predetermined threshold value for each. Are selectively disconnected in the disconnection step, and instead, in the switching step, transmission paths are set between the same number of nodes listed in the candidate node list. In addition, after a certain period of time, each adjacent node is again ranked based on the number of acquisitions corresponding to each adjacent node updated every time content is acquired in the counting step and the round trip time obtained in the measurement step. Then, the node that is determined to have a lower rank regarding the number of times of content acquisition or a round trip time is switched to a candidate node.

By repeating such operations, nodes that have a disadvantageous condition in forming an information distribution route that the number of times of content acquisition is low or the round-trip time is long are separated, and conversely, the number of times of content acquisition is high and the round-trip time is high The connection with the node having the advantageous condition that the time is short can be maintained.
A fifth information distribution route dynamic control method according to the present invention includes an update step, a first rearrangement step, a second rearrangement step, a disconnection step, and a switching step.

The principle of the fifth dynamic control method of the information distribution route according to the present invention is as follows.
Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination, the update step acquires each content from each adjacent node. In addition, the number of hits corresponding to the candidate node collected in the candidate node list as being connected to the adjacent node is updated, and the number of acquisitions of the content held corresponding to the adjacent node in the adjacent node list is updated. To do. In the first sorting step, the candidate node list is sorted based on the updated number of hits. The second rearrangement step rearranges the adjacent node list based on the number of acquisitions obtained for each adjacent node. In the disconnecting step, based on the adjacent node list obtained as a result of the rearrangement, the connection with a predetermined number of adjacent nodes is selectively disconnected from the one having a smaller number of acquisitions. In the switching step, a predetermined number of candidate nodes are selected from the candidate node list in descending order of the number of hits and connected instead of the disconnected node.

The operation of the fifth information distribution route dynamic control method configured as described above is as follows.
For example, each time content is acquired via an adjacent node, in the update step, the number of acquisitions corresponding to the adjacent node of the content transmission source in the adjacent node list is added by a numerical value “1”, and the content transmission source described above is added. The number of hits related to the candidate nodes collected in the candidate node list corresponding to the adjacent node is added by the numerical value “1”, and the adjacent node list and the candidate node list are updated. As described above, the number of acquisitions corresponding to each adjacent node included in the adjacent node list and the number of hits corresponding to the related candidate node are updated every time content is acquired, and each candidate is further updated in the first sorting step. The candidate node list is rearranged based on the number of hits corresponding to the node. And, for example, in the second rearrangement step in the process of the connection confirmation process performed every fixed time, the adjacent node list is sorted based on the number of times of acquisition of the content obtained as described above, In the disconnection step, the connection with the predetermined number of nodes located at the end of the sorted adjacent node list is selectively disconnected in the disconnection step, and instead, in the switching step, the connection is listed above the candidate node list. Set up transmission paths with the same number of nodes.

  By repeating such an operation, nodes that have a disadvantageous condition in forming an information distribution route that the number of times of content acquisition is low are separated, and instead collected for neighboring nodes that have a high number of content acquisition times. Since it is possible to switch to the candidate node, it is possible to quickly realize a state in which each autonomous distributed control node maintains a connection with an adjacent node having a high content acquisition count.

A sixth information distribution route dynamic control method according to the present invention includes an update step, a first rearrangement step, a measurement step, an evaluation step, a second rearrangement step, a disconnection step, and a switching step. It consists of steps.
The principle of the sixth method for dynamically controlling information distribution routes according to the present invention is as follows.
Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination, the update step acquires each content from each adjacent node. In addition, the number of hits corresponding to the candidate node collected in the candidate node list as being connected to the adjacent node is updated, and the number of acquisitions of the content held corresponding to the adjacent node in the adjacent node list is updated. To do. In the first sorting step, the candidate node list is sorted based on the updated number of hits. In the measurement step, when the connection state is confirmed for each of the adjacent nodes being connected, the round trip time required for the connection confirmation data to reciprocate with each adjacent node is measured. The evaluation step evaluates the usefulness of the connection with each adjacent node based on the number of acquisitions and the round trip time obtained for each adjacent node. The second sorting step sorts the neighboring node list based on the usefulness obtained for each neighboring node. The disconnecting step selectively disconnects the connection with a predetermined number of adjacent nodes from the one with the lower usefulness based on the adjacent node list obtained as a result of the rearrangement. In the switching step, a predetermined number of candidate nodes are selected from the candidate node list in descending order of the number of hits, and connected instead of the disconnected node.

The operation of the sixth information distribution route dynamic control method configured as described above is as follows.
Similar to the fifth information distribution route dynamic control method described above, in the update step, the number of acquisitions corresponding to the corresponding adjacent node in the adjacent node list and the relationship with the adjacent node in the candidate node list for each content acquisition The number of hits corresponding to the candidate node to be updated is updated, and the candidate node list is rearranged based on the number of hits corresponding to each candidate node in the first rearrangement step. In this way, by rearranging the candidate node list, it is possible to arrange nodes that are highly likely to be related to content acquisition at the top of the candidate node list. And, for example, in the measurement step performed in the process of connection confirmation processing performed every fixed time, the round trip time corresponding to each adjacent node is measured, and based on this round trip time and the number of acquisitions held in the adjacent node list, In the evaluation step, the usefulness of each adjacent node is evaluated, and in the disconnection step, the adjacent node whose connection is to be disconnected is selected. In the switching step, a node to which a communication path is newly connected is selected using the candidate node list rearranged based on the number of hits updated each time content is acquired as described above. In other words, based on the result of comprehensive evaluation of the number of content acquisitions and the round-trip time required for communication, the connection with the adjacent nodes with low usefulness may be selectively disconnected and related to content acquisition The content distribution route in the virtual network is optimized by preferentially selecting a candidate node having a high value as a new connection destination.

A seventh information distribution route dynamic control method according to the present invention includes an update step, a rearrangement step, a measurement step, a ranking step, a disconnection step, and a switching step.
The principle of the seventh dynamic control method of the information distribution route according to the present invention is as follows.
Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination, the update step acquires each content from each adjacent node. In addition, the number of hits corresponding to the candidate node collected in the candidate node list as being connected to the adjacent node is updated, and the number of acquisitions of the content held corresponding to the adjacent node in the adjacent node list is updated. To do. The rearrangement step rearranges the candidate node list based on the updated number of hits. In the measurement step, when the connection state is confirmed for each of the adjacent nodes being connected, the round trip time required for the connection confirmation data to reciprocate with each adjacent node is measured. In the ranking step, a rank related to the number of acquisitions and a rank related to the round-trip time are assigned to each adjacent node included in the adjacent node list based on the number of times of acquisition and round-trip time obtained for each adjacent node. The disconnecting step selectively disconnects a connection between an adjacent node whose score based on the ranking given regarding the number of acquisitions is equal to or less than a predetermined threshold and an adjacent node whose score based on the rank related to round trip time is equal to or less than the predetermined threshold. In the switching step, the same number of candidate nodes as the adjacent nodes whose connections were disconnected in the disconnection step are selected from the candidate node list in the descending order of the number of hits, and connected instead of the disconnected nodes.

The operation of the seventh information distribution route dynamic control method configured as described above is as follows.
Similar to the fifth information distribution route dynamic control method described above, in the update step, the number of acquisitions corresponding to the corresponding adjacent node in the adjacent node list and the relationship with the adjacent node in the candidate node list for each content acquisition The number of hits corresponding to the candidate node to be updated is updated, and the candidate node list is rearranged based on the number of hits corresponding to each candidate node in the first rearrangement step. In this way, by rearranging the candidate node list, it is possible to arrange nodes that are highly likely to be related to content acquisition at the top of the candidate node list. Then, for example, in the measurement step performed in the process of the connection confirmation process performed at regular intervals, the round trip time corresponding to each adjacent node is measured, and in the ranking step, the round trip time is determined for each adjacent node included in the adjacent node list. And a rank regarding the number of times of acquisition are used as indexes for selecting an adjacent node to be disconnected in the disconnection step. Then, instead of the separated adjacent nodes, a necessary number of candidate nodes are selected from the candidate node list rearranged as described above in the switching step, and these candidate nodes are connected as new adjacent nodes. The In other words, the connection with an adjacent node with a small number of times of content acquisition or a long round-trip time required for communication is selectively disconnected, and candidate nodes that are likely to be related to content acquisition are preferentially newly connected. First, optimize the distribution channel of content in the virtual network.

  According to the dynamic control method of the first information distribution route according to the present invention described above, in the virtual network composed of autonomous distributed control nodes based on the measurement result of the round-trip time related to information transmitted / received at the time of connection confirmation By optimizing the information distribution route, it is possible to reduce the time required to transfer information and form a network capable of distributing contents at high speed.

  Further, according to the first information distribution route dynamic control method, the content distribution route is optimized by optimizing the information distribution route in the virtual network composed of autonomous distributed control nodes based on the number of times the content is acquired from the adjacent node. Thus, it is possible to form a network that can reduce the number of nodes that pass through when acquiring the content, and reduce the load on each node involved in relaying content.

  Furthermore, according to the third information distribution route dynamic control method, the result of comprehensive evaluation of the measurement result of the round trip time related to the information transmitted and received at the time of connection confirmation and the number of times the content is acquired from the adjacent node Based on the above, by optimizing the information distribution path in the virtual network consisting of autonomous distributed control nodes, simultaneously reducing the time required to transfer information and reducing the number of nodes through which content is acquired, It is possible to form a network capable of distributing content at high speed while reducing the load on each node involved in content relay.

Further, according to the fourth information distribution route dynamic control method, the connection is made based on each of the measurement result of the round trip time related to the information transmitted and received at the time of connection confirmation and the number of times the content is acquired from the adjacent node. By selecting nodes to be maintained, it is possible to form a virtual network composed of autonomous distributed control nodes that achieves two advantageous conditions.
Furthermore, the selection of the switching destination node based on the candidate node list rearranged in consideration of the possibility related to content acquisition is applied to the second, third and fourth information distribution route dynamic control methods, By optimizing the distribution route by each autonomous control node earlier, a virtual network capable of providing a distribution route close to the ideal can be realized quickly.

  As described above, it can be used in a network that can distribute content more efficiently, that is, a network in which the time required to transfer content is short and the number of nodes involved in content distribution is reduced. Since it is possible to quickly provide desired content to the user, serviceability to the user can be improved.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[First Embodiment]
FIG. 1 shows a first embodiment of an autonomous distributed control node according to the present invention.
Note that among the components shown in FIG. 1, the same components as those shown in FIG. 16 are denoted by the symbols shown in FIG. 16, and description thereof is omitted.

  In the autonomous distributed control node 210 shown in FIG. 1 (hereinafter simply referred to as the node 210), the adjacent node list 211 corresponds to the node IDs of all adjacent nodes for which communication paths are set up with the node 210. Thus, a round-trip time required for transmitting / receiving predetermined information to / from each adjacent node is held. Further, in the adjacent node management unit 212 shown in FIG. 1, the connection control unit 213 connects or has already set a communication path with the designated node in accordance with an instruction from the management processing unit 214. Processing necessary for disconnecting the communication path is executed using the function of the communication processing unit 412. Further, in the adjacent node management unit 212 shown in FIG. 1, the round trip time measurement unit 215 monitors the operation of the communication processing unit 412 in accordance with an instruction from the management processing unit 214, and the communication processing unit 412 Measures the time required to send and receive connection confirmation information for confirming the connection with the node.

Further, the list management unit 216 illustrated in FIG. 1 updates the corresponding round trip time held in the adjacent node list 211 based on the measurement result received from the round trip time measurement unit 215, and the connection control unit 213 executes the update. The adjacent node list 211 and the candidate node list 414 are updated according to the connection or disconnection processing of the broken communication path.
FIG. 2 is a flowchart showing operations for managing adjacent nodes.

Of the steps shown in FIG. 2, those equivalent to the steps shown in FIG. 17 are denoted by the same reference numerals as those shown in FIG.
The management processing unit 214 shown in FIG. 1 reads the node ID of each adjacent node and the corresponding round trip time from the adjacent node list 211 (step 301), and rearranges each adjacent node in ascending order of the round trip time (step 301). 302).

Next, the management processing unit 214 selects, for example, k adjacent nodes located at the end of the adjacent node list 211 (step 303), and instructs the connection control unit 213 to switch these nodes to candidate nodes. As a result, the connection control unit 213 performs switching similar to the steps 506 to 510 shown in FIG. 17 (step 304).
Thereafter, from the top of the list of adjacent nodes including the node that has become a new adjacent node, the management processing unit 214 sequentially selects one of the adjacent nodes as a node to be checked and confirms the connection with the connection control unit 213. In response to this, the connection control unit 213 sends a connection confirmation message to the designated adjacent node as a destination (step 502). At this time, the management processing unit 214 instructs the round trip time measurement unit 215 to measure the round trip time, and in response to this, the round trip time measurement unit 215 waits for the connection response message to be returned after the connection confirmation message is sent. Is measured (step 305).

  When a connection response message is returned before a predetermined time has elapsed since the connection confirmation message was sent (affirmative determination in step 503), the round trip time measurement unit 215 corresponds the measurement result to the node to be confirmed described above. In response to this, the list management unit 216 updates the round-trip time stored in the adjacent node list 211 corresponding to the node ID of the corresponding adjacent node (step 306). At this time, the connection control unit 213 passes the node ID included in the connection response message received via the communication processing unit 412 to the list management unit 216 as a list of nodes adjacent to the confirmation target node. In response, the list management unit 216 updates the candidate node list 414 by adding these node IDs to the candidate node list 414 (step 504).

  On the other hand, if the connection response message is not returned before the predetermined time elapses, the node to be confirmed is removed from the adjacent nodes as described in steps 506 to 510 in FIG. A process of switching to one of the candidate nodes listed in the candidate node list 414 is performed (step 307), and the process returns to step 502, and a node that becomes a new adjacent node by this switching process is set as a node to be confirmed. Perform connection confirmation processing.

  In this way, in the process of the connection confirmation process S1 (indicated by a chain line in FIG. 2), the round-trip time required for transmission / reception of a message necessary for connection confirmation is measured by the round-trip time measuring unit 215. Prior to the connection confirmation processing S1, the management processing unit 214 sorts the adjacent node list 211 based on the round trip time, and the connection control unit 213 switches a predetermined number of adjacent nodes to candidate nodes in order of long round trip time. Thus, by applying the first information distribution route dynamic control method according to the present invention, an autonomous distributed control node that manages each adjacent node can be realized.

  In such an autonomous distributed control node, in the connection confirmation process, in addition to the adjacent node in which the connection is detected, the adjacent node that takes a long time from sending the connection confirmation message to returning the connection response message Can be selectively switched to candidate nodes. That is, while maintaining a connection with an adjacent node that can be expected to have a short time required to transfer information, the adjacent node having a disadvantageous condition that the information transfer time is expected to be long due to the length of the round trip time Can be switched to a candidate node that may possibly take less time to transfer information than the adjacent node.

In this way, each autonomous distributed control node selects adjacent nodes based on the round trip time, so that based on the distance between nodes estimated from the length of the round trip time, By optimizing, it is possible to reduce the time required for content distribution as a whole.
[Second Embodiment]
FIG. 3 shows a second embodiment of the autonomous distributed control node according to the present invention.

3 that are the same as those shown in FIG. 16 are denoted by the same reference numerals as those shown in FIG. 16, and description thereof is omitted.
In the node 210 illustrated in FIG. 3, the adjacent node list 221 indicates the number of times content has been acquired from each adjacent node corresponding to the node IDs of all adjacent nodes for which communication paths are established with the node 210. keeping. Further, each time the content acquisition control unit 416 acquires content, the hit notification unit 222 shown in FIG. 3 receives information on the adjacent node that is the transmission source of the content from the content acquisition control unit 416, and manages this information in a list management. To part 216. Further, in the adjacent node management unit 212 illustrated in FIG. 3, the management processing unit 223 controls the connection confirmation operation with each adjacent node via the connection control unit 213.

FIG. 4 is a flowchart showing the content acquisition operation.
Of the steps shown in FIG. 4, those equivalent to the steps shown in FIG. 18 are denoted by the reference numerals shown in FIG. 18, and description thereof is omitted.
When the content acquisition control unit 416 receives the content search result via the communication processing unit 412 (after the end of step 524), the hit notification unit 222 illustrated in FIG. 3 receives the content search result from the content acquisition control unit 416. The node ID indicating the transmission source adjacent node is received, and the node ID is notified to the list management unit 216 (step 311). In response to this, the list management unit 216 increments and updates the number of acquisitions of the content stored corresponding to the node ID notified to the adjacent node list 221 (step 312), and then proceeds to step 525. The content is received.

  In this way, each time content is acquired, the number of acquisitions corresponding to the adjacent nodes involved in the transfer of the content can be accumulated. The content acquisition count obtained in this way indicates the number of times each adjacent node is involved in acquiring the content requested by the user. In other words, the content acquisition count of each adjacent node held in the adjacent node list 221 is an index representing the degree of usefulness of each adjacent node in acquiring content.

Next, a method for managing adjacent nodes based on the number of times of content acquisition will be described.
FIG. 5 is a flowchart showing an operation for managing adjacent nodes.
Of the steps shown in FIG. 5, the steps equivalent to those shown in FIGS. 17 and 2 are denoted by the reference numerals shown in FIG. 17, and the description thereof is omitted.

  As shown in FIG. 5, the management processing unit 223 reads the node ID and the number of acquisition times of each adjacent node from the adjacent node list 221 (step 313), and then, for example, the content acquisition number is large in the read list of adjacent nodes. They are rearranged in order (step 313). Next, the management processing unit 223 selects k adjacent nodes located at the end of the rearranged adjacent node list in this way (step 303), and then sets these adjacent nodes in the candidate node list 414. The connection destination is switched to the same number of retained candidate nodes (step 304).

Thereafter, the connection confirmation processing S0 equivalent to the connection confirmation processing S0 surrounded by the chain line in FIG. 17 confirms the connection with all adjacent nodes including the newly connected node in step 304, and the processing is terminated. .
As described above, in the process of acquiring the content, the management processing unit 223 selects the node from which the connection is released based on the number of times of acquisition of the content collected by the hit notification unit 222 and the list management unit 216. An autonomous distributed control node to which the second information distribution route dynamic control method according to the invention is applied can be realized.

  In such an autonomous decentralized control node, in the connection confirmation process, in addition to the adjacent node in which the connection is detected to be disconnected, an adjacent node with a small number of times of content acquisition is selectively switched to a candidate node. be able to. In other words, it is disadvantageous that the probability of content acquisition is expected to be low because the number of times the content search result has been transmitted is small while maintaining connection with an adjacent node having a high probability of content acquisition. A connection with an adjacent node having a condition can be selectively disconnected, and the candidate node can be switched to a candidate node with a higher probability of content acquisition than the adjacent node.

In this way, each autonomous distributed control node selects adjacent nodes based on the number of times of content acquisition, thereby increasing the usefulness of the adjacent nodes related to content acquisition estimated from the number of times of content acquisition. Based on this, it is possible to optimize the information distribution route. This creates a virtual network that connects the nodes involved in the acquisition of content with high probability, so when each node acquires the content, the content that passes from the node that owns the content to the requesting node It is possible to reduce the number of nodes to be transferred and to reduce the total processing load on each node constituting the virtual network in the content distribution process as a whole.
[Third Embodiment]
Furthermore, it is possible to evaluate the usefulness of the adjacent node based on both the round trip time and the number of times of content acquisition, and manage the switching of the adjacent node based on the evaluation result.

FIG. 6 shows a third embodiment of the autonomous distributed control node according to the present invention.
6, the same components as those shown in FIGS. 16, 1, and 3 are denoted by the same reference numerals as those shown in FIGS. 16, 1, and 3. Is omitted.
In the node 210 illustrated in FIG. 6, the adjacent node list 224 includes the number of times content is acquired from each adjacent node corresponding to the node IDs of all adjacent nodes for which a communication path is set with the node 210. It holds a round trip time required to transmit / receive information related to connection confirmation with the node and a usefulness index indicating the degree of usefulness of the connection with the node. 6, every time the content acquisition control unit 416 acquires content, the hit notification unit 222 receives information on the acquisition destination adjacent node from the content acquisition control unit 416, and this information is received by the list management unit 216. To pass. Further, in the adjacent node management unit 212 shown in FIG. 6, the management processing unit 225 performs connection confirmation with each adjacent node via the connection control unit 213, and in the course of this connection confirmation processing, the round trip time The measurement unit 215 measures the round trip time required from the transmission of the connection confirmation message to the reception of the connection response message. In addition, the usefulness calculation unit 226 shown in FIG. 6 determines the number of acquisitions and the round trip time held corresponding to the adjacent node specified in the adjacent node list 224 in response to an instruction from the management processing unit 225. Based on this, a usefulness index that comprehensively reflects these characteristic values is calculated, and the calculated usefulness index is held in the adjacent node list 224.

In the process of acquiring content, the operation of collecting the number of acquisitions related to each adjacent node is equivalent to the description in the second embodiment described above, and is therefore omitted here.
Next, a method for managing adjacent nodes based on both the number of content acquisitions and the round trip time will be described.
FIG. 7 is a flowchart showing operations for managing adjacent nodes.

Of the steps shown in FIG. 7, those equivalent to the steps shown in FIG. 17, FIG. 2, and FIG. 5 are denoted by the reference numerals shown in FIG. 17, FIG. 2, and FIG. Is omitted.
As shown in FIG. 7, after reading the node ID and the usefulness index of each adjacent node from the adjacent node list 224 (step 315), the management processing unit 225 displays a list of the read adjacent nodes with a high usefulness index, for example. They are rearranged in order (step 316). Next, the management processing unit 225 selects k adjacent nodes positioned at the end of the rearranged adjacent node list in this way (step 303), and then sets these adjacent nodes in the candidate node list 414. The connection destination is switched to the same number of retained candidate nodes (step 304).

  Thereafter, the management processing unit 225 proceeds with connection confirmation processing for all adjacent nodes including the node newly connected in step 304 in the same manner as the connection confirmation processing S1 surrounded by a chain line in FIG. . In this process, after updating the round trip time in step 306, the management processing unit 225 instructs the usefulness calculation unit 226 to calculate the usefulness index, and in response to this, the usefulness calculation unit 226, for example, obtains The usefulness index u is calculated by the expression (1) expressed using the number of times s, the round trip time t, and predetermined constants Cs and Ct, respectively, and the obtained usefulness index u corresponds to the corresponding adjacent node. And stored in the adjacent node list 224 (step 317).

u = Cs × s + Ct / t (1)
After the update of the adjacent node list 224 is completed in this way, the process proceeds to the processes after step 504, and when the connection confirmation process for all the adjacent nodes is completed, the management process for the adjacent nodes is ended, as described above. The updated adjacent node list 224 is used for the next adjacent node management process.

  As described above, both the number of content acquisitions collected by the hit notification unit 222 and the list management unit 216 during the content acquisition process and the round trip time measured by the round trip time measurement unit 215 during the connection confirmation process are obtained. The usefulness index that is reflected is calculated by the usefulness calculation unit 226 in the process of the connection confirmation process, and based on this, the management processing unit 225 selects a node that releases the connection, thereby providing a third related to the present invention. An autonomous distributed control node to which a dynamic control method of information distribution route is applied can be realized.

  In such an autonomous distributed control node, in the connection confirmation process, in addition to the adjacent node in which the connection is detected, the usefulness index that reflects the number of times of content acquisition and the time required for the connection confirmation An adjacent node with a low can be selectively switched to a candidate node. In other words, while maintaining a connection with an adjacent node with a high overall evaluation of the probability of content acquisition and the length of round-trip time, a connection with an adjacent node with a low overall evaluation is selectively disconnected, and this adjacent It is possible to switch to a candidate node that may be involved in the acquisition of content with a higher probability than a node or that may have a round trip time shorter than that of an adjacent node. It is also possible to provide an appropriate threshold for the above-described usefulness index, and select an adjacent node whose usefulness index is lower than this threshold as a switching target.

In this way, each autonomous distributed control node selects adjacent nodes based on the above-described usefulness index, so that the probability of content acquisition and the inter-node distance estimated from the round trip time are integrated. Based on the result of the evaluation, it is possible to optimize the information distribution route. This creates a virtual network that connects nodes that are highly useful in the distribution of information, so when each node acquires content, it forwards the content from the node that owns the content to the requesting node. It is possible to reduce the number of nodes to be transferred and shorten the distance to which information is transferred, and as a whole, it is possible to reduce the total processing load and transfer time applied to each node constituting the virtual network in the content distribution process. .
[Fourth Embodiment]
FIG. 8 shows a fourth embodiment of the autonomous distributed control node according to the present invention.

8 that are equivalent to the components shown in FIGS. 16, 1, and 3 are denoted by the same reference numerals as those shown in FIGS. 16, 1, and 3. Is omitted.
In the node 210 illustrated in FIG. 8, the adjacent node list 227 includes the number of times content is acquired from each adjacent node corresponding to the node IDs of all adjacent nodes for which communication paths are established with the node 210. It holds the round trip time required to transmit / receive information related to connection confirmation with the node. Also, in the adjacent node management unit 212 shown in FIG. 8, the management processing unit 228 performs connection confirmation with each adjacent node via the connection control unit 213, and in the course of this connection confirmation processing, the round trip time The measurement unit 215 measures the round trip time required from the transmission of the connection confirmation message to the reception of the connection response message. Further, the list sort unit 229 shown in FIG. 8 is held in the adjacent node list 227 based on the number of acquisitions or the round trip time held in the adjacent node list 227 in accordance with an instruction from the management processing unit 228. The order of each adjacent node is rearranged, and the rearrangement result is passed to the management processing unit 228.

In the process of acquiring content, the operation of collecting the number of acquisitions related to each adjacent node is equivalent to the description in the second embodiment described above, and thus the description is omitted here. A method for managing adjacent nodes based on both the round trip time will be described.
FIG. 9 is a flowchart showing operations for managing adjacent nodes.

Note that among the steps shown in FIG. 9, the steps equivalent to those shown in FIGS. 17 and 2 are denoted by the reference numerals shown in FIGS. 17 and 2, and the description thereof is omitted.
As illustrated in FIG. 9, the management processing unit 228 sorts the list of adjacent nodes held in the adjacent node list 227 using the function of the list sort unit 229 based on the round trip time and the number of times of acquisition. The rankings given to the nodes as the respective sorting results are obtained (steps 321 and 322). Next, the management processing unit 228 selects, for example, k adjacent nodes having low ranks based on the round-trip time or the number of times obtained in this way (step 323), and then selects these adjacent nodes as candidate nodes. The connection destination is switched to the same number of candidate nodes held in the list 414 (step 304).

Thereafter, the connection confirmation process S1 surrounded by a chain line in FIG. 2 is executed, and in step 304, the connection with all adjacent nodes including the newly connected node is confirmed, and the process is terminated.
As described above, according to the number of times content is collected by the hit notification unit 222 and the list management unit 216 in the process of acquiring the content, and the round trip time measured by the round trip time measuring unit 215 in the process of the connection confirmation process. Autonomous distribution to which the fourth information distribution route dynamic control method according to the present invention is applied by selecting a node from which the management processing unit 228 releases the connection based on the rearranged list of adjacent nodes. A control node can be realized.

  In such an autonomous distributed control node, in the connection confirmation process, in addition to the adjacent node where the connection is detected, the adjacent node has a small number of times of content acquisition or a long time required for the connection confirmation. Can be selectively switched to candidate nodes. In other words, while maintaining the connection with the adjacent node that has the advantageous condition that the probability of content acquisition is high and the round trip time is short, for example, the connection with the adjacent node with extremely low characteristics related to one of the conditions is selected And can be switched to a candidate node that may have more favorable characteristics than this neighboring node.

In this way, each autonomous distributed control node selects adjacent nodes based on the rank according to the round-trip time and the rank according to the number of acquisitions, thereby increasing the possibility of content acquisition. Based on both the content acquisition and the information transfer time estimated from the round trip time, the information distribution route can be optimized. As a result, a virtual network is formed that connects nodes located at a relatively close distance physically with high probability in content acquisition, so that each node owns the content when acquiring the content. Reduces the number of nodes that transfer the content from the requesting node to the requesting node and shortens the distance to which the information is transferred, and as a whole the processing burden on each node that constitutes the virtual network in the content distribution process In addition, the total transfer time can be reduced.
[Fifth Embodiment]
FIG. 10 shows a fifth embodiment of the autonomous distributed control node according to the present invention.

10 that are the same as those shown in FIGS. 16 and 3 are denoted by the same reference numerals as those shown in FIGS. 16 and 3, and description thereof is omitted.
In the node 210 shown in FIG. 10, the candidate node list 231 includes, as information related to each candidate node, a node ID indicating each candidate node, and a node ID indicating an adjacent node related to the collection of information on the candidate node. The number of hits added each time content is acquired from the corresponding adjacent node is held. Further, the hit count updating unit 232 illustrated in FIG. 10 performs the hit count corresponding to the corresponding candidate node held in the candidate node list 231 based on the node ID received from the hit notification unit 222 as described later. Update. Further, the candidate list sorting unit 233 illustrated in FIG. 10 performs the sorting process of the candidate node list 231 in accordance with the update of the candidate node list 231 by the hit count updating unit 232 described above.

  For example, as shown in FIG. 11, nodes 210 (shown with reference A0 in FIG. 11) are adjacent to nodes indicated by reference signs A1 to A5, and in FIG. 11, reference signs C1 to C3 are added. When a plurality of nodes (not shown) are connected to these adjacent nodes to form a virtual network, the adjacent node list 221 and the candidate node list 231 of the node 210 are, for example, The result is as shown in FIGS.

In FIG. 12, the node IDs of the adjacent nodes A1 to A5 shown in FIG. 11 are indicated by reference signs A1 to A5, respectively, and the node IDs of the candidate nodes C1 to C3 shown in FIG. 11 are indicated by reference signs C1 to C3, respectively. Indicated.
Next, an operation related to updating and rearranging of the adjacent node list 221 and the candidate node list 231 in the process of acquiring contents will be described.

FIG. 13 is a flowchart showing the content acquisition operation.
Of the steps shown in FIG. 13, those equivalent to the steps shown in FIG. 18 or FIG. 4 are designated by the reference numerals shown in FIG. 18 or FIG.
When the content acquisition control unit 416 receives the content search result via the communication processing unit 412 (after the end of step 524), the hit notification unit 222 illustrated in FIG. 10 receives the content search result from the content acquisition control unit 416. The node ID indicating the transmission source adjacent node is received, and the node ID is notified to the list management unit 216 (step 311). In response to this, the list management unit 216 increments the number of acquisitions held in the adjacent node list 221 corresponding to the notified node ID, similarly to step 312. At this time, the hit count updating unit 232 shown in FIG. 10 increments the hit count included in the candidate node information including the node ID notified from the candidate node list 231 as the adjacent node ID (step 324). ). That is, for example, when the node ID indicating the adjacent node A3 is received from the hit notification unit 222, the list management unit 216 increments the corresponding acquisition count of the adjacent node list 221, while the hit count update unit 232 The number of hits corresponding to the candidate node (indicated by reference characters C2 and C3 in FIG. 12) including the node ID (A3) is incremented. Thereafter, in response to the notification that the update has been completed from the hit count update unit 232, the candidate list sort unit 233 performs a process of sorting the candidate node list 231 based on the number of hits (step 325). The candidate node list 231 rearranged by is used for processing by the adjacent node management unit 212.

  In this way, in the process of acquiring the content, the hit count update unit 232 updates the hit count corresponding to an appropriate candidate node in response to the notification from the hit notification unit 222, and the candidate list is updated according to the update result. The sorting unit 233 rearranges the candidate node list 231 and applies the candidate node list 231 to the processing of the adjacent node management unit 212, thereby applying the fifth information distribution route dynamic control method according to the present invention. An autonomous distributed control node can be realized.

Here, the number of hits corresponding to each candidate node held in the candidate node list 231 by the above-described processing indicates the high possibility that the candidate node is involved when the node 210 acquires the content.
Therefore, using the candidate node list 231 rearranged based on the number of hits in this way, the adjacent node management unit 212 controls the switching from the adjacent node to the candidate node according to the procedure shown in FIG. Thus, it is possible to preferentially select a candidate node that has a high possibility of being involved in content acquisition and set it as a new connection destination.

In other words, in such an autonomous distributed control node, each autonomous distributed control node selects an adjacent node based on the number of times of content acquisition, so that it is related to content acquisition estimated from the number of times of content acquisition. In the process of optimizing the distribution route of information based on the usefulness of adjacent nodes, the distribution route is optimized by preferentially selecting a candidate node that is likely to be involved in content acquisition as a new connection destination. Can be greatly accelerated.
[Sixth Embodiment]
Next, a method for applying the technique related to the rearrangement of the candidate node list described above to the autonomous distributed control node according to the third embodiment will be described.

FIG. 14 shows a sixth embodiment of the autonomous distributed control node according to the present invention.
14, the same components as those shown in FIGS. 16, 6, and 10 are denoted by the same reference numerals as those shown in FIGS. 16, 6, and 10. Is omitted.
The node 210 illustrated in FIG. 14 includes a candidate node list 231, a hit count update unit 232, and a candidate list sort unit 233 in addition to the units illustrated in FIG.

That is, in the process of acquiring content, the candidate node list 231 rearranged as described above is used for the adjacent node management process shown in FIG. With this configuration, it is possible to realize an autonomous distributed control node to which the sixth information distribution route dynamic control method according to the present invention is applied.
In such an autonomous distributed control node, based on a usefulness index that reflects the number of times content is acquired and the round-trip time, each autonomous distributed control node selects an adjacent node to optimize the information distribution route. In the process of achieving this, optimization of distribution channels can be greatly accelerated by preferentially selecting candidate nodes that are likely to be involved in content acquisition as new connection destinations.
[Seventh Embodiment]
Next, a method for applying the technique related to the rearrangement of the candidate node list described above to the autonomous distributed control node according to the fourth embodiment will be described.

FIG. 15 shows a sixth embodiment of the autonomous distributed control node according to the present invention.
15 that are the same as those shown in FIGS. 16, 8, and 10 are denoted by the same reference numerals as those shown in FIGS. 16, 8, and 10, and the description thereof is omitted. Is omitted.
The node 210 illustrated in FIG. 15 includes a candidate node list 231, a hit count update unit 232, and a candidate list sort unit 233 in addition to the units illustrated in FIG.

That is, in the process of acquiring content, the candidate node list 231 rearranged as described above is used for the adjacent node management process shown in FIG. By configuring in this way, it is possible to realize an autonomous distributed control node to which the seventh information distribution route dynamic control method according to the present invention is applied.
In such an autonomous distributed control node, in the process of optimizing the information distribution route, each autonomous distributed control node selects adjacent nodes based on the number of times of content acquisition and the round trip time. By optimizing candidate nodes that are likely to be involved in content acquisition as new connection destinations, optimization of distribution channels can be greatly accelerated.

  As described above, by applying the information distribution route dynamic control method according to the present invention to a virtual network composed of autonomous distributed control nodes, an overlay composed of autonomous distributed control nodes located in an upper layer of the IP network is provided. In the network, an appropriate information distribution topology can be formed without depending on the route in the hierarchy below the IP network.

  As a result, it is possible to set an optimal information distribution route that connects the node that requested the content and the node that owns the content across networks provided by each of the plurality of Internet service providers. Can be further activated, and the satisfaction of the user who has requested the content can be improved, which is very useful in the field of providing a variety of content distribution services.

It is a figure which shows 1st Embodiment of the autonomous distributed control node concerning this invention. It is a flowchart showing the operation | movement which manages an adjacent node. It is a figure which shows 2nd Embodiment of the autonomous distributed control node concerning this invention. It is a flowchart showing content acquisition operation | movement. It is a flowchart showing the operation | movement which manages an adjacent node. It is a figure which shows 3rd Embodiment of the autonomous distributed control node concerning this invention. It is a flowchart showing the operation | movement which manages an adjacent node. It is a figure which shows 4th Embodiment of the autonomous distributed control node concerning this invention. It is a flowchart showing the operation | movement which manages an adjacent node. It is a figure which shows 5th Embodiment of the autonomous distributed control node concerning this invention. It is a figure which shows the structural example of a virtual network. It is a figure which shows the example of an adjacent node list and a candidate node list. It is a flowchart showing content acquisition operation | movement. It is a figure which shows 6th Embodiment of the autonomous distributed control node concerning this invention. It is a figure which shows 7th Embodiment of the autonomous distributed control node concerning this invention. It is a figure which shows the structural example of the virtual network which consists of an autonomous distributed control node. It is a flowchart showing an adjacent node management operation. It is a flowchart showing content acquisition operation | movement. It is a flowchart showing a part of operation | movement which relays search information and a content. It is a flowchart showing another part of operation | movement which relays search information and a content.

Explanation of symbols

210 Autonomous Distributed Control Nodes 211, 221, 227, 413 Adjacent Node List 212, 415 Adjacent Node Management Unit 213 Connection Control Unit 214, 223, 225, 228 Management Processing Unit 215 Round Trip Time Measurement Unit 216 List Management Unit 222 Hit Notification Unit 226 Usefulness calculation unit 229 List sort unit 231, 414 Candidate node list 232 Hit frequency update unit 233 Candidate list sort unit 410 Autonomous distributed control node (node)
412 Communication processing unit 416 Content acquisition control unit 417 Content list 418 Search ID list 419 Content relay control unit

Claims (7)

Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination,
A measurement step of measuring a round trip time required for the data for connection confirmation to reciprocate between the adjacent nodes when confirming a connection state for each of the adjacent nodes being connected;
A reordering step for reordering the adjacent node list based on a round trip time obtained by measurement for each adjacent node;
Based on the adjacent node list obtained as a result of the rearrangement, a disconnecting step of selectively disconnecting a predetermined number of adjacent nodes from the longer corresponding round trip time;
And a switching step of selecting the predetermined number of candidate nodes from the candidate node list and connecting the selected nodes instead of the disconnected nodes. Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination,
A counting step for counting the number of times content is acquired from each adjacent node for each connected adjacent node;
Reordering step for reordering the adjacent node list based on the number of times obtained for each adjacent node;
Based on the adjacent node list obtained as a result of the rearrangement, a disconnection step of selectively disconnecting a predetermined number of adjacent nodes from the one having a smaller number of acquisitions correspondingly,
And a switching step of selecting the predetermined number of candidate nodes from the candidate node list and connecting the selected nodes instead of the disconnected nodes. Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination,
A counting step for counting the number of times content is acquired from each adjacent node for each connected adjacent node;
A measurement step of measuring a round trip time required for the data for connection confirmation to reciprocate between the adjacent nodes when confirming a connection state for each of the adjacent nodes being connected;
An evaluation step for evaluating the usefulness of connection with each adjacent node based on the number of times of acquisition and round trip time obtained for each adjacent node;
A reordering step of reordering the adjacent node list based on the usefulness obtained for each of the adjacent nodes;
Based on the adjacent node list obtained as a result of the rearrangement, a disconnecting step of selectively disconnecting a predetermined number of adjacent nodes from the corresponding less useful level,
And a switching step of selecting the predetermined number of candidate nodes from the candidate node list and connecting the selected nodes instead of the disconnected nodes. Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination,
A counting step for counting the number of times content is acquired from each adjacent node for each connected adjacent node;
A measurement step of measuring a round trip time required for the data for connection confirmation to reciprocate between the adjacent nodes when confirming a connection state for each of the adjacent nodes being connected;
A ranking step of assigning a rank related to the number of acquisitions and a rank related to the round-trip time to each adjacent node included in the adjacent node list based on the number of acquisitions and round-trip time obtained for each adjacent node;
A disconnecting step of selectively disconnecting a connection between an adjacent node whose score based on the ranking given regarding the number of acquisitions is equal to or less than a predetermined threshold and an adjacent node whose score based on the rank related to round trip time is equal to or less than a predetermined threshold;
A switching step of selecting from the candidate node list the number of candidate nodes that have been disconnected in the disconnecting step and connecting in place of the disconnected node; Control method. Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination,
Each time content is acquired from each of the adjacent nodes, the number of hits corresponding to the candidate node collected in the candidate node list as being connected to the adjacent node is updated, and the adjacent node in the adjacent node list is updated. An update step for updating the number of times the corresponding stored content is acquired;
A first sorting step for sorting the candidate node list based on the updated number of hits;
A second reordering step for reordering the adjacent node list based on the number of times obtained for each of the adjacent nodes;
Based on the adjacent node list obtained as a result of the rearrangement, a disconnection step of selectively disconnecting a predetermined number of adjacent nodes from the one having a smaller number of acquisitions correspondingly,
A switching step of selecting the predetermined number of candidate nodes in descending order of the number of hits from the candidate node list and connecting the selected node instead of the disconnected node. Control method. Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination,
Each time content is acquired from each of the adjacent nodes, the number of hits corresponding to the candidate node collected in the candidate node list as being connected to the adjacent node is updated, and the adjacent node in the adjacent node list is updated. An update step for updating the number of times the corresponding stored content is acquired;
A first sorting step for sorting the candidate node list based on the updated number of hits;
A measurement step of measuring a round trip time required for the data for connection confirmation to reciprocate between the adjacent nodes when confirming a connection state for each of the adjacent nodes being connected;
An evaluation step for evaluating the usefulness of connection with each adjacent node based on the number of times of acquisition and round trip time obtained for each adjacent node;
A second reordering step for reordering the adjacent node list based on the usefulness obtained for each of the adjacent nodes;
Based on the adjacent node list obtained as a result of the rearrangement, a disconnecting step of selectively disconnecting a predetermined number of adjacent nodes from the corresponding less useful level,
A switching step of selecting the predetermined number of candidate nodes in descending order of the number of hits from the candidate node list and connecting the selected node instead of the disconnected node. Control method. Information related to the nodes connected to the connected adjacent nodes shown in the adjacent node list is further collected in the candidate node list, and when connection with a new node becomes necessary, In the dynamic control method of the information distribution route in the autonomous distributed control node configured to maintain the connection with a certain number of nodes by selecting the connection destination,
Each time content is acquired from each of the adjacent nodes, the number of hits corresponding to the candidate node collected in the candidate node list as being connected to the adjacent node is updated, and the adjacent node in the adjacent node list is updated. An update step for updating the number of times the corresponding stored content is acquired;
A reordering step for reordering the candidate node list based on the updated number of hits;
A measurement step of measuring a round trip time required for the data for connection confirmation to reciprocate between the adjacent nodes when confirming a connection state for each of the adjacent nodes being connected;
A ranking step of assigning a rank related to the number of acquisitions and a rank related to the round-trip time to each adjacent node included in the adjacent node list based on the number of acquisitions and round-trip time obtained for each adjacent node;
A disconnecting step of selectively disconnecting a connection between an adjacent node whose score based on the ranking given regarding the number of acquisitions is equal to or less than a predetermined threshold and an adjacent node whose score based on the rank related to round trip time is equal to or less than a predetermined threshold;
A switching step of selecting the same number of candidate nodes as the adjacent nodes disconnected in the disconnecting step from the candidate node list in descending order of the number of hits, and connecting in place of the disconnected node Dynamic control method of information distribution path characterized by

Images