JP2010066930A - Content distributed storage system, content storage method, node device and node processing program - Google Patents

Abstract

A distributed content storage system, a content storage method, a node device, and a node device capable of automatically storing a replica of content in accordance with the recent popularity degree of the entire network in the node device from the beginning of participation in the system Provides a node processing program.
A node device selects a plurality of contents for a predetermined data capacity based on a content list when there is a predetermined free capacity in the storage area in the storage means when participating in a distributed content storage system. The copy data of the selected content is acquired and saved.
[Selection] Figure 9

Description

  The present invention relates to a technical field of a peer-to-peer (P2P) type communication system including a plurality of node devices that can communicate with each other via a network.

  In this type of peer-to-peer communication system, duplicate data (hereinafter referred to as “replica”) of a plurality of contents (contents such as movies and music data) is distributed and stored (arranged) in a plurality of node devices. A distributed content storage system that makes it possible to use replicas between devices is known, which improves fault tolerance and access dispersibility. The location of the replicas thus distributed and stored can be efficiently searched using a distributed hash table (hereinafter referred to as DHT (Distributed Hash Table)) as disclosed in, for example, Patent Document 1.

Then, when acquiring the desired content, the node device participating in the distributed content storage system sends a message (query) for searching (discovering) the location of the replica of the content to the other node device. As a result, the message is transferred by the relay node device to the managing node device of the content replica in accordance with the DHT. Then, the node device that has transmitted the message acquires information indicating the location of the replica from the node device of the management source that finally reaches the message. As a result, the node device that has transmitted the message can access (acquire) (download) the replica by accessing the node device that stores the replica of the content related to the search.
JP 2006-197400 A

  In a peer-to-peer communication system, a method in which a node device that has acquired a content replica stores the replica in a storage (for example, a hard disk) as a cache is common. In that case, there is no sufficient cache in the storage of the node device in the initial stage of participation in the system, and almost all content replicas must be obtained from other node devices each time. This is a disadvantage.

  Further, for example, when a replica in the storage is deleted by an instruction from the center or the like, a free space is generated in the storage. However, if this free space is large, it tends to be disadvantageous as described above.

  For example, considering that a peer-to-peer communication system is applied to a communication karaoke system, when a node device with an empty storage (or a large amount of storage) is connected to a peer-to-peer network, a content request is made until the cache is accumulated. Can only rely on downloads from other node devices, puts a load on the network, and cannot immediately reproduce the content.

  Furthermore, even if a replica of the content is stored in the storage in advance, it is advantageous to put a popular one in the storage as it is less downloaded from other node devices. How to do is not yet known.

  The present invention has been made in view of the above points and the like, and from the beginning of participation in the system, it is possible to automatically store a replica of content in accordance with the recent popularity of the entire network in the node device. An object is to provide a distributed content storage system, a content storage method, a node device, and a node processing program.

  In order to solve the above-described problem, the invention according to claim 1 is a distributed content storage system including a plurality of node devices that can communicate with each other via a network. The node device in the content distributed storage system that is stored in a distributed manner and can use the content among the node devices, and performs a participation process for participating in the content distributed storage system A storage unit having a storage area for storing the duplicated data, a content list acquisition unit that acquires a content list in which information about each content that can be used in the content distributed storage system is described, and a storage unit Empty to determine whether the storage area has a predetermined free space Content that selects a plurality of contents for a predetermined data capacity based on the content list when it is determined that there is a predetermined free capacity in the storage area when participating in the content distributed storage system Selection means and storage control means for acquiring copy data of the selected content from another node device or a server device having copy data of all contents, and storing the copy data in the storage area of the storage means And.

  According to the present invention, the node device selects a plurality of contents corresponding to a predetermined data capacity based on the content list when the storage area in the storage means has a predetermined free capacity when participating in the distributed content storage system. Since the copy data of the selected content is acquired and stored, the copy data of the content according to the recent popularity of the entire network is automatically stored in the node device from the beginning of participation in the distributed content storage system. Thus, the immediacy of content reproduction can be improved.

  According to a second aspect of the present invention, in the node device according to the first aspect, the content selection unit has a predetermined free space in the storage area by deleting the duplicated data stored in the storage area. If it is possible, content corresponding to a predetermined data capacity is selected based on the content list.

  According to this invention, the immediacy of content reproduction can always be improved.

  According to a third aspect of the present invention, in the node device according to the first or second aspect, the content selection unit preferentially selects content having a high probability of being used.

  According to the present invention, it is possible to preferentially acquire and store content having a high probability of being used.

  According to a fourth aspect of the present invention, in the node device according to any one of the first to third aspects, in the information related to each content, a plurality of data of the content is acquired by any of the plurality of node devices. Information indicating at least one of the number of times of acquisition and the number of times of reproduction of the duplicate data of the content by any of the plurality of node devices is included, and at least one of the number of times of acquisition and the number of times of reproduction is included. Probability calculation means for calculating a probability proportional to the number of times for each content is further provided, wherein the content selection means selects the content based on the probability calculated for each content.

  According to the present invention, it is possible to preferentially select popular content that has a high probability of being used, and content obtained by a plurality of node devices can be stochastically varied.

  According to a fifth aspect of the present invention, in the node device according to the fourth aspect, the information related to each content further includes information indicating at least one of the genre, the target generation, and the region attribute of the content. The probability calculation means increases the probability of content that matches at least one of a genre, a generation, and a region attribute set in advance in the node device with reference to the information about each content. It is characterized by that.

  According to the present invention, it is possible to easily acquire a replica of content that a user likes by specifying a genre or a region attribute.

  According to a sixth aspect of the present invention, in the node device according to any one of the first to fifth aspects, content acquisition means for acquiring duplicate data of content designated by the user through the input means from another node device. The copy data acquisition process by the content acquisition unit is prioritized when the copy data acquisition process by the content acquisition unit and the copy data acquisition process by the storage control unit overlap. And

  According to this invention, even when the copy data acquisition process by the content acquisition unit and the copy data acquisition process by the storage control unit overlap, it is avoided that the user satisfaction is lowered. can do.

  The invention of a node processing program according to a seventh aspect is characterized in that a computer is caused to function as the node device according to any one of the first to sixth aspects.

  The invention according to claim 8 is a content distributed storage system including a plurality of node devices that can communicate with each other via a network, wherein duplicate data of a plurality of contents is distributed and stored in a plurality of node devices, A content distributed storage system that enables the content to be used between the node devices, wherein the node device stores participation processing means for performing participation processing for participating in the content distributed storage system, and stores the duplicate data Storage means having a storage area for performing, content list acquisition means for acquiring a content list describing information about each content available in the content distributed storage system, and a predetermined free space in the storage area in the storage means Free capacity determining means for determining whether or not there is capacity; Content selection means for selecting, based on the content list, a plurality of contents corresponding to a predetermined data capacity when it is determined that the storage area has a predetermined free capacity when participating in the distributed storage system; Storage control means for acquiring copy data of the content obtained from another node device or a server device having copy data of all contents, and storing the copy data in the storage area of the storage means. And

  The invention according to claim 9 is a content distributed storage system comprising a plurality of node devices that can communicate with each other via a network, wherein duplicate data of a plurality of contents is distributed and stored in a plurality of node devices, A content storage method in a content distributed storage system that enables the content to be used between the node devices, wherein the node device executes a participation process for participating in the content distributed storage system; and A content list acquisition step in which a node device acquires a content list in which information about each content that can be used in the content distributed storage system is described; and a predetermined storage area in which the node device stores the duplicate data Free space determination process to determine whether there is free space and before Content selection for selecting, based on the content list, a plurality of contents corresponding to a predetermined data capacity when it is determined that the storage area has a predetermined free capacity when participating in the distributed content storage system And the node device acquires the copy data of the selected content from another node device or a server device having copy data of all the content, and stores the copy data in the storage area in the storage unit And a storage control step.

  According to the present invention, the node device selects, based on the content list, a plurality of contents corresponding to a predetermined data capacity when the storage area in the storage means has a predetermined free capacity when participating in the distributed content storage system. Since the copy data of the selected content is acquired and stored, the copy data of the content according to the recent popularity of the entire network is automatically stored in the node device from the beginning of participation in the distributed content storage system. Thus, the immediacy of content reproduction can be improved.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, the best embodiment of the invention will be described with reference to the drawings. The embodiment described below is an embodiment when the present invention is applied to a content distributed storage system.

[ 1. Outline and configuration of distributed content storage system ]
First, with reference to FIG. 1 etc., the structure and operation | movement outline | summary of the content distribution preservation | save system which concern on this embodiment are demonstrated.

  FIG. 1 is a diagram showing an example of a connection mode of each node device in the distributed content storage system according to this embodiment.

  As shown in the lower frame 101 of FIG. 1, IX (Internet eXchange) 3, ISP (Internet Service Provider) 4a, 4b, DSL (Digital Subscriber Line) line operators (devices) 5a, 5b, FTTH (Fiber To A network (real-world communication network) 8 such as the Internet is constructed by a line provider (device) 6 and a communication line (for example, a telephone line or an optical cable) 7. Note that a router for transferring data (packets) is appropriately inserted into the network 8 in the example of FIG.

  A plurality of node devices (hereinafter referred to as “nodes”) Nn (n = 1, 2, 3,...) Are connected to such a network 8. Each node Nn is assigned a unique manufacturing number and an IP (Internet Protocol) address. The content distributed storage system S according to the present embodiment is a peer-to-peer network system formed by participation of any of a plurality of nodes Nn, as shown in the upper frame 100 of FIG. It has become.

  A network 9 shown in the upper frame 100 of FIG. 1 is an overlay network 9 (logical network) that forms a virtual link formed using the existing network 8. Such an overlay network 9 is realized by a specific algorithm, for example, an algorithm using DHT. Each node Nn participating in the content distributed storage system S (in other words, the overlay network 9) is assigned a node ID, which is unique identification information having a predetermined number of digits.

  The node ID is, for example, a value (for example, the bit length is 160 bits) obtained by hashing an IP address or a manufacturing number individually assigned to each node Nn with a common hash function (for example, SHA-1). Yes, it will be distributed in one ID space without any bias.

  In addition, participation in the content distributed storage system S is performed by any node Nn (for example, the node N8) that has not participated in any node Nn (for example, a contact node that always participates in the system S) that has not participated. This is done by sending a participation message indicating a participation request.

  Each node Nn holds a routing table using DHT. This routing table defines the transfer destinations of various messages on the content distributed storage system S, specifically, a node including the node ID, IP address, and port number of a node Nn that is moderately separated in the ID space Multiple pieces of information are registered.

  One node Nn participating in the content distributed storage system S registers the minimum necessary node information of the nodes Nn among all the nodes Nn participating in the system S in the routing table, With respect to the node Nn that does not know (store) the node information, various messages are transferred between the nodes Nn to be delivered.

  Such a routing table using DHT is well known in Japanese Patent Application Laid-Open No. 2006-197400 and the like, and will not be described in detail.

  By the way, in the content distributed storage system S, replicas of various contents having different contents (for example, movies, music, etc.) are distributed (stored) in a plurality of nodes Nn in a predetermined file format. The replica can be used between Nn. For example, the node N5 stores a replica of the movie content with the title XXX, while the node N3 stores a replica of the movie content with the title YYY. (Hereinafter referred to as “content holding nodes”).

  In addition, information such as a content name (title) and a content ID (identification information unique to each content) is added to each of these content replicas. This content ID is generated, for example, by hashing the content name + arbitrary numerical value (or may be the first few bytes of the content data) with a hash function common to obtaining the node ID (node ID and Placed in the same ID space). Alternatively, the system administrator may give a unique ID value (same bit length as the node ID) for each content. In this case, the content catalog information in which the correspondence between the content name and the content ID is written is distributed to all the nodes Nn.

  In addition, the location of the replica of the content that is distributed and stored in this way, that is, the index information that includes the node information of the node Nn that stores the replica and the content ID of the content is the location of the replica of the content. Is stored (stored in the index cache) and managed by a node Nn (hereinafter referred to as “root node” or “root node of content (content ID)”) or the like.

  That is, the node information of the content holding node storing the content replica is managed by the root node so as to be provided in response to an inquiry from another node Nn.

  For example, the index information about the replica of the movie content with the title XXX is managed by the node N4 which is the root node of the content (content ID), and the index information about the replica of the movie content with the title YYY is Managed by the node N7 which is the root node of the content (content ID). Further, such a root node is determined to be, for example, a node Nn having a node ID closest to the content ID (for example, the higher-order digits match more).

  When a user of a certain node Nn wants to acquire a replica of the desired content, the node Nn that desires to acquire the replica (hereinafter referred to as “user node”) determines the content ID and the self of the content selected by the user. A content location inquiry (search) message (query) including the IP address of the client is generated and sent to another node Nn according to a routing table using its own DHT. That is, the user node sends a content location inquiry (search) message to the root node (to the root node) (inquires the location of the content replica to the root node). As a result, the content location inquiry (search) message finally arrives at the root node by DHT routing using the content ID as a key.

  The attribute information such as the content name and content ID of the content to be selected by the user in each node Nn is described in, for example, content catalog information distributed from the content providing server to all the nodes Nn.

  The content ID included in the content location inquiry (search) message may be generated by the user node by hashing the content name with the common hash function. The DHT routing is known in Japanese Patent Application Laid-Open No. 2006-197400 and the like, and thus detailed description thereof is omitted.

  The root node that receives the content location inquiry (search) message acquires index information corresponding to the content ID included in the content node from the index cache, and uses the index information as a user who is the transmission source of the content location inquiry message. Reply to the node. The user node that has acquired the index information in this way accesses the content holding node based on the IP address of a content holding node included in the index information, transmits a content transmission request message, and then acquires a content replica therefrom. It becomes possible to download (acquire). Note that the index information may include, for example, node information of a plurality of content holding nodes (when replicas of the same content are stored in a plurality of content holding nodes). In such a case, the user node can select one content holding node from among the plurality of content holding nodes and connect to the selected content holding node to download a content replica.

  The root node transmits a content transmission request message to the content holding node indicated by the IP address or the like included in the index information, whereby the user node downloads the replica from the content holding node. You can also. The user node can also acquire the index information from a cache node that caches the same index information as the root node until the content location inquiry message reaches the root node.

  As described above, the user node downloads a replica of the content selected by the user from the content holding node, and uses the replica as a cache, for example, a storage area for storing the replica (hereinafter referred to as a hard disk). , “Storage”), and thereafter, for example, every time a reproduction instruction is given from the user, the reproduction is performed from the storage. However, at the initial stage of participation in the distributed content storage system S, there is no sufficient cache in the storage, and most of the content desired by the user must be downloaded from the content holding node each time as described above. Therefore, in the present embodiment, each node Nn is a server that has a content holding node or a replica of all content when the storage has a predetermined free capacity when participating in the distributed content storage system S. Therefore, the immediacy of content reproduction is improved by downloading and storing a replica of popular content that can be used in the system S, for example, with a high probability of being used for a predetermined data capacity of the storage.

  Note that the user node that has stored the replica of the content acquired from the content holding node informs the root node that the replica has been stored (in other words, for other nodes Nn participating in the system S). A publish (registration notification) message including the content ID of the replica and its own node information, and sends the publish message to the root node (to the root node). As a result, the publish message arrives at the root node by DHT routing using the content ID as a key, like the content location inquiry (search) message. Then, the root node registers (stores in the index cache area) index information including a set of node information and content ID included in the received publish message. In this way, the user node becomes a new content holding node that holds a replica of the content.

  Note that the index information including the set of node information and content ID included in the publish message is also registered (cached) in the cache node in the transfer path to the root node.

[ 2. Configuration and function of node Nn ]
Next, the configuration and function of the node Nn will be described with reference to FIG.

  FIG. 2 is a diagram illustrating a schematic configuration example of the node Nn.

  As shown in FIG. 2, each node Nn has a control unit 11 as a computer composed of a CPU having a calculation function, a working RAM, a ROM for storing various data and programs, and various data and various programs. Included in a storage unit 12 (an example of a storage unit) composed of an HD (hard disk) for storing and storing (stored), a buffer memory 13 for temporarily storing a received content replica and the like, and a content replica A decoder unit 14 that decodes (decompresses or decodes) encoded video data (video information), audio data (audio information), and the like, and performs a predetermined drawing process on the decoded video data, etc. Video processing unit 15 to output as a signal, and video signal output from the video processing unit 15 Display unit 16 such as a CRT or a liquid crystal display for displaying an image based on the above, and an audio processing unit for converting the decoded audio data into an analog audio signal by D (Digital) / A (Analog) conversion and then amplifying it with an amplifier and outputting it 17, a speaker 18 that outputs an audio signal output from the sound processing unit 17 as a sound wave, a communication unit 19 for controlling communication of information between other nodes Nn and the like through the network 8, and an instruction from a user And an input unit (for example, a keyboard, a mouse, a remote controller, an operation panel, etc.) 20 serving as an input unit that receives an instruction signal corresponding to the instruction and supplies the instruction signal to the control unit 11. , Storage unit 12, buffer memory 13, decoder unit 14, communication unit 19, and input unit 20 are mutually connected via bus 21. It is connected. As the node Nn, a personal computer, an STB (Set Top Box), a TV receiver, or the like can be applied. For example, the node Nn is installed in a user's home, a karaoke shop, a hotel, or the like.

  Further, the storage unit 12 stores a routing table using DHT, index information, content catalog information, and an IP address and a port number of a contact node that is an access destination when participating in the content distributed storage system S. Has been.

  Further, the capacity (for example, 100 gigabytes) of the storage (storage area for storing a replica of the content) included in the storage unit 12 can be arbitrarily set by the user, for example.

  The storage unit 12 stores a popular content list (an example of a content list). This popular content list describes information about each content that can be used in the content distributed storage system S, for example, is popular (the popularity ranking is within a predetermined upper rank (for example, 20th)). Used to select a replica to download when joining S.

  FIG. 3 is a diagram illustrating an example of the contents of a popular content list. In the popular content list shown in FIG. 3, as an example of content-related information, a content name, the number of playback times, a playback point, a DL (download) number, a DL point, a target generation, and a region attribute are associated with each content ID. include. Note that the information regarding the content described in the popular content list shown in FIG. 3 is described in the order of content with the highest playback points (that is, with the highest popularity), but may be described in the order with the higher DL points.

  Here, the number of times of reproduction is information obtained by summing up the number of times that a replica of the content is reproduced by the node Nn within a certain period in the content distributed storage system S. The playback point (score) is a value proportional to the number of times of playback (the higher the number of times of playback, the higher the number of times of playback), and any method can be used as long as normalization is possible. For example, the playback ratio of the content in the list Can be calculated by the following calculation formula (1).

Playback point = number of playbacks of content ÷ total number of playbacks in popular content list × 100 (1)
The DL count is information obtained by counting the number of times a replica of the content has been downloaded (acquired) by the node Nn in the content distributed storage system S within a certain period. The DL point (score) is a value proportional to the number of DLs (the higher the number of DLs, the higher the number of DLs). As with the number of playbacks, any method can be used as long as normalization is possible. The DL ratio can be calculated by the following calculation formula (2).

DL points = DL number of content ÷ Total number of DLs in popular content list × 100 (2)
If it is desired to give weight to either the number of playbacks or the number of DLs, this can be handled by assigning weights using the above calculation formula.

  The genre (genre tag) is information added to the content in advance in accordance with the orientation of the content. For example, if the content is music data, songs such as “Rock”, “Pops”, “Jazz”, “For children”, “For young people”, “For middle-aged”, “For elderly”, “For foreigners”, etc. It is set according to the orientation. Also, if the content is video data, for example, “sports”, “comedy”, “action”, “foreign film”, “Japanese film”, “overseas drama”, “domestic drama”, “education” etc. It is set together.

  Such a genre is added to the content by the service operator when the content is input to the content distributed storage system S (for the first time, it is stored in the node Nn) (the genre is a node later) It may be set by Nn users). In the example of FIG. 3, target generations (“for young people”, “for middle-aged people”, and “for elderly people”) are described as examples of genres.

  Further, the regional attribute (regional characteristic) is information indicating a region where the content is likely to be used. Some contents are easily used in each region. For example, some of Okinawan songs and Osaka songs are popular nationwide, but it is thought that there are many opportunities to be sung and listened to, especially in that region.

  Further, in each node Nn, for example, at the time of initial setting, a genre, a region attribute, and a weight ratio thereof can be set, and the set genre, a region attribute, and a weight ratio thereof are stored in the storage unit 12. The

The genre weight ratio is, for example,
(For young people: For middle-aged people: For elderly people) = (1: 1: 1)
(For young people: For middle-aged people: For elderly people) = (2: 1: 1)
(For young people: For middle-aged people: For elderly people) = (1: 0: 0)
(For young people: For middle-aged people: For elderly people) = (1:10:10)
These can be set as follows, and this may be set to a genre point (point magnification) described later. Such a genre and weight ratio can be arbitrarily designated by the user. For example, in the node Nn installed in a karaoke shop for restaurants at night, by increasing the weight ratio of “for middle-aged and elderly”, the genre point of the content with “for middle-aged and elderly” added is increased to make it easier to download. it can. Further, the user may set a desired genre point of the content to which the genre is added by setting one or more desired genres without setting the weight ratio. For example, when the user sets only “for young people” as the genre, the genre point of the content to which “for young people” is added is set to, for example, “10”, and the genre point of the content to which “for young people” is not added is set. For example, by setting “1”, it is possible to increase the genre point of the content to which “for young people” is added and to facilitate downloading.

In addition, the weight ratio of the regional attribute is, for example,
(None (no regional attributes): Okinawa) = (1: 2)
(No: Hokkaido) = (1:10)
It can be set as follows, and this may be set to a region attribute point (point magnification) described later. As the set regional attribute, for example, installation location information (for example, a zip code or an address) set in the node Nn may be used, or a desired regional attribute may be designated by the user. Good. In addition, the weight ratio of the regional attribute can be arbitrarily designated by the user. For example, by increasing the weight ratio of the regional attribute “Okinawa”, it is possible to increase the genre point of the content to which “Okinawa” is added and to facilitate the download. Further, the user may set a desired region attribute without setting a weight ratio, so that the region attribute point of the content to which the region attribute is added is increased. . For example, when the user sets only “Hokkaido” as the region attribute, the region attribute point of the content with “Hokkaido” added is set to “10”, for example, and the region attribute point of the content without “Hokkaido” added is set. For example, by setting “1”, it is possible to increase the regional attribute point of the content to which “Hokkaido” is added and to facilitate the download.

  In the configuration as described above, the control unit 11 performs overall control by reading and executing a program (including the node processing program of the present invention) stored in the storage unit 12 or the like by the CPU. Processing as at least one of a node, a relay node, a root node, a cache node, and a content holding node is performed. Further, the control unit 11 functions as a participation processing unit, a free space determination unit, a content list acquisition unit, a probability calculation unit, a content selection unit, a storage control unit, and a content acquisition unit in the present invention.

  More specifically, the control unit 11 executes a participation process for participating in the distributed content storage system S, for example, when the own node is powered on. In such participation processing, for example, the control unit 11 connects to the contact node via the network 8 and transmits a participation message to the contact node. In response to this, a routing table using DHT is created based on information returned from the content node or the like, and participates in the distributed content storage system S. At this time, the control unit 11 determines whether or not the storage has a predetermined free capacity (for example, a data capacity capable of storing a plurality of content replicas). Based on the content list, a plurality of contents corresponding to a predetermined data capacity (for example, an empty capacity or a specified capacity) are selected.

  Here, there are a method of acquiring the popular content list from the management server when the node Nn participates, for example, and a method of acquiring the popular content list by creating the node Nn autonomously when participating, for example.

  FIG. 4 is a conceptual diagram showing how the node Nn acquires the popular content list from the management server. In this method, as shown in FIG. 4, when each node Nn reproduces a content replica and downloads a content replica from a content holding node or the like, its log information (including content ID) is displayed. Then, the information related to each content that can be used in the content distributed storage system S is transmitted to a management server (for example, a content providing server) of the center. The management server can calculate the number of reproductions and the number of DLs of each content by counting the log information. Further, for example, setting location information (for example, a zip code or an address) is set in each node Nn, and each node Nn includes the location information in log information and transmits it to the management server. It is possible to obtain the regional attribute of each content by summing up information on the usage region trends of the content. In this way, the management server creates a popular content list in which information related to content that is popular during an arbitrary period is described, and provides the list to each node Nn. Thus, according to the method in which the management server creates the popular content list, the accuracy of the popular content list can be improved. The playback point and DL point described in the popular content list may be calculated by the management server, or may be calculated by the control unit 11 of the node Nn that has acquired the popular content list.

  On the other hand, FIG. 5 is a conceptual diagram showing a method of acquiring by the node Nn creating autonomously. In this method, each node Nn stores content reproduction history and download history data for a certain period or a certain amount of data, and if there is a history data acquisition request from another node Nn, the history data Is provided to the node Nn. Thereby, each node Nn can calculate the number of reproductions and the number of DLs of each content by collecting history data from, for example, a plurality of other nodes Nn selected at random, and collecting the data. Note that the greater the number of nodes Nn from which history data is collected, the closer the aggregation result is to the case of aggregation by the management server. In addition, in the method shown in FIG. 5, it is not easy to calculate the area attribute of each content by collecting information on the usage area trend as compared with the method shown in FIG. 4, but each node Nn makes a history data acquisition request. It is more effective if it is configured so that history data is acquired only from nodes Nn in a certain area by specifying the installation location when performing the other node Nn. For example, priority is given to the acquisition of historical data from the node Nn installed in a location close to the installation location of the node (for example, the prefecture is the same or the postal code is close, etc.). It becomes easier to obtain information about certain content. In this way, each node Nn creates a popular content list in which information about content that is popular in an arbitrary period is described. Thus, according to the method in which each node Nn creates a popular content list, the load on the management server can be reduced. The playback points and DL points described in the popular content list are calculated by the control unit 11 of each node Nn.

  Then, when selecting content based on the popular content list, the control unit 11 creates a download list based on the popular content list and the setting of the weight ratio of the genre and the region attribute.

  6 to 8 are diagrams showing examples of the contents of the download list created based on the popular content list shown in FIG. As illustrated, in the download list, in addition to the items described in the popular content list, the total points and download probabilities of each content are described. The total points and the download probabilities may be calculated by any method as long as a standardized value can be calculated. For example, the above-described playback points, DL points, genre points, and regional attribute points are used. Can be calculated by the following equations (3) and (4).

Total points = playback points x DL points x genre points x regional attribute points (3)
Download probability = total points / total total points in popular content list × 100 (4)
According to the equations (3) and (4), it can be said that the download probability is a probability proportional to at least one of the number of reproductions and the number of DLs, and further, a genre preset in the own node. And the download probability of content that matches at least one of the regional attributes varies depending on, for example, the weight ratio.

  In the example of FIG. 6, the weight ratio of the genre is set as (for young people: for middle-aged and older people: for elderly people) = (1: 0: 0), and the region attribute is not particularly specified. In this case, the genre point of the content whose genre is “for young people” is set to “1” (the genre point of the content for middle-aged and elderly people is “0”), and the regional attribute is not particularly specified. The regional attribute point is set to “1” in all contents in the popular content list, and the total point and the download probability are calculated.

  In the example of FIG. 7, the weight ratio of the genre is set as (for young people: for middle-aged and older people: for elderly people) = (1: 0: 0), and the regional attribute is (none: Okinawa) = (1: 2). Shows the contents of the download list. In this case, the genre point of content whose genre is “young” is set to “1”, and the regional attribute point of content whose regional attribute is “Okinawa” is set to “2” (“1” except for Okinawa). Thus, the total points and the download probability are calculated.

  In the example of FIG. 8, a case where the genre and the region attribute are not particularly specified is shown. In this case, the genre point and the region attribute point are set to “1” in all the contents in the popular content list, and the total point and the download probability are calculated.

  Then, the control unit 11 selects content based on the download probability calculated for each content. For example, the control unit 11 determines whether or not to select for each content, using an arithmetic expression in which the selection ratio increases as the download probability increases. For example, in the example of FIG. 8, since the download probability of “content bbb” is about 20%, it is selected at a rate of once every five times, but the download probability of “content ccc” is about 5%. Therefore, it will be selected at a rate of once in 20 times. That is, content with a high probability of being used is preferentially selected.

  The control unit 11 performs such content selection processing until, for example, there is no free storage capacity or until the storage capacity reaches the capacity specified by the user, and the selected content replica is used as a root node. By inquiring about the location, the content is downloaded from the content holding node (or downloaded from the content providing server), and the replica is stored in the storage. It should be noted that downloading of replicas may be performed every time content is selected, or may be performed collectively after selection of all content is completed.

  In addition, the replica once stored in the storage may be deleted by, for example, an instruction from the center or an instruction from the user to create a free space. For such a case, the control unit 11 determines whether or not the storage has a preset free capacity (for example, a capacity capable of storing at least one content replica) in the storage, and the free capacity is determined. When it is determined that the content has been created, the content of the predetermined data capacity (for example, the free capacity or the specified capacity) is selected and downloaded based on the popular content list, and the replica is stored in the storage as described above. Let

  By the way, as described above, the node Nn automatically downloads a content replica when there is free space in the storage. However, the automatic download process and the user use the node Nn (for example, singing karaoke, video And a request download process that downloads a replica of the content requested by the user (that is, designated through the input unit 20) from the content holding node or the like (batting). Is done. In such a case, the load on the control unit 11 increases, and it may take longer than usual to download the content by the request download process, and the user's satisfaction may be reduced. Therefore, in the present embodiment, when the automatic download process overlaps with the request download process, the control unit 11 executes the load avoidance process to prioritize the request download process. Specific examples of such load avoidance processing include the following three methods (a) to (c).

  (A) The control unit 11 temporarily stops the automatic download process and preferentially downloads a replica of the content requested by the request download process. Then, the control unit 11 resumes the stopped automatic download process when the download of the replica by the request download process is completed.

  (B) The replica download bit rate by the automatic download process is lowered, and the replica download bit rate by the request download process is raised.

  (C) A time interval is set for the replica download by the automatic download process so as not to overlap with the replica download by the request download process.

  The node processing program may be downloaded from a predetermined server on the network 8, for example, or recorded on a recording medium such as a CD-ROM and read via the drive of the recording medium. You may make it.

[ 3. Operation of node Nn ]
Next, the operation of the node Nn according to the present embodiment will be described with reference to FIG.

  FIG. 9 is a flowchart showing processing of the control unit 11 in the node Nn.

  The process shown in FIG. 9 is started, for example, when the power is turned on at the node Nn. First, the control unit 11 performs initial setting according to an instruction from the user via the input unit 20 (step S1). In this initial setting, as described above, the genre and area attributes and their weight ratios can be set, and the set genre and area attributes and their weight ratios are stored in the storage unit 12. This initial setting can be changed later.

  Next, the control unit 11 performs a participation process (step S2). In the participation process, the control unit 11 performs routing using DHT by using node information included in a message returned by transmitting a participation message (including node information of the own node) to the contact node. Generate a table. As a result, the node Nn participates in the content distributed storage system S.

  Next, the control unit 11 determines whether or not the storage in the storage unit 12 has a predetermined free space (step S3). If it is determined that there is a free space (step S3: YES), the control unit 11 proceeds to step S4. If it is determined that there is no free space (step S3: NO), the process proceeds to step S14.

  In step S4, the control unit 11 acquires a popular content list as described above.

  Next, the control unit 11 creates a download list as described above based on the popular content list and the setting of the weight ratio of the genre and the region attribute (step S5).

  Next, the control unit 11 specifies, for example, one of the highest (most popular) contents described in the download list (step S6).

  Next, the control unit 11 determines whether or not the specified content replica is already stored in the storage (step S7). If the replica is already stored in the storage (step S7: YES), step S12 is performed. If it is not stored in the storage yet (step S7: NO), the process proceeds to step S8.

  In step S8, the control unit 11 reads the download probability of the identified content from the download list, and determines whether or not to select the content based on the probability as described above. When the control unit 11 determines to select the content (step S8: YES), the control unit 11 proceeds to step S9. When the control unit 11 determines not to select the content (step S8: YES), the control unit 11 proceeds to step S12. move on.

  In step S9, the control unit 11 determines whether or not the content request from the user has been within the past predetermined time (for example, several seconds before). If there is no request (step S9: NO), the control unit 11 proceeds to step S10. If there is such a request (step S9: YES), the process proceeds to step S11.

  In step S10, the control unit 11 performs automatic download processing of the selected content. In such an automatic download process, for example, the control unit 11 generates a content location inquiry message including the content ID of the selected content, its own IP address, and the like, and uses this according to the routing table using its own DHT. Sending to the node Nn (that is, sending to the root node). Thus, the content location inquiry message finally arrives at the root node by DHT routing using the content ID as a key, and the root node that has received the message stores the index information corresponding to the content ID included in the index cache. And the index information is returned to the node Nn that is the transmission source of the content location inquiry message. The control unit 11 of the node Nn that has acquired the index information in this way accesses the content holding node based on the IP address or the like of a certain content holding node included in the index information, and transmits a content transmission request message. Download a content replica and save it to storage. If the index information cannot be acquired from the root node within a certain time due to the load condition of the network, the control unit 11 of the node Nn accesses the content providing server, downloads the replica of the content from there, and stores it in the storage. save.

  In step S11, the control unit 11 performs the load avoidance process described above (for example, any one of the three methods (a) to (c) described above). As a result, the content replica selected in step S8 and the content replica requested by the user are downloaded from the content holding node or the content providing server and stored in the storage.

  Subsequently, the control part 11 deletes the information regarding the content specified by said step S6 from a download list (step S12).

  Next, the control unit 11 determines whether or not the free space of the storage has reached a predetermined capacity (step S13). If the free space of the storage has reached the predetermined capacity (step S13: YES), the control unit 11 proceeds to step S14. If the free capacity of the storage does not reach the predetermined capacity (in other words, there is still a free capacity in the storage where the replica of the content can be stored) (step S13: NO), the process returns to step S6. In this way, the control unit 11 again identifies one top (most popular) content described in the download list and repeats the same processing as described above. Here, since the information regarding the content already specified in step S6 has been deleted from the download list, it is not specified redundantly in step S6. By such processing, the selected content replica is downloaded and stored in the storage until, for example, the free space of the storage runs out or the storage capacity reaches the capacity designated by the user.

  In step S <b> 14, the control unit 11 performs other processing such as various message transmission / reception processing and processing in accordance with an instruction from the user (for example, content download processing, content reproduction processing, content deletion processing, etc.).

  In step S15, the control unit 11 determines whether or not there is a power-off command. If there is no power-off command (step S15: NO), the control unit 11 returns to step S3 and performs the same processing as above (replica). Is deleted, the process after step S4 is performed again), and when there is a power-off command (step S15: YES), the process ends.

  As described above, according to the above embodiment, the node Nn, when participating in the distributed content storage system S, stores a plurality of contents corresponding to a predetermined data capacity as popular contents when there is a predetermined free capacity in the storage. Since the selected content replica is downloaded from the content holding node or the like and stored in the storage, the replica of the selected content is configured to be stored in the storage. It is possible to automatically store a replica of the content according to the popularity etc. in the node Nn. Thereby, the immediacy of content reproduction can be improved.

  Further, even when the storage stored in the storage is deleted and the storage becomes free, the node Nn selects, for example, content corresponding to the free capacity based on the popular content list, and the selected content replica. Is downloaded from the content holding node or the like and the replica is stored in the storage, so that the immediacy of content reproduction can always be improved.

  In addition, the node Nn calculates, for each content, a download probability that is proportional to at least one of content download count and playback count in the popular content list, and downloads based on the download probability calculated for each content. Since the content is selected as the content to be selected, the popular content having a high probability of being used can be preferentially selected, and the content downloaded at the plurality of nodes Nn can be stochastically separated.

  Further, since the node Nn is configured to change the download probability according to the setting of the genre, the region attribute, and the weight ratio thereof, it is possible to easily download the replica of the content that the user likes by specifying the genre and the region attribute. it can. In addition, it can be said that only a replica of content to which a certain genre is added is downloaded among the contents in the popular content list.

  Furthermore, when the automatic download process and the request download process overlap, the node Nn can avoid a decrease in user satisfaction by performing a load avoidance process to prioritize the request download process.

  In addition, although the content distributed storage system S in the said embodiment was demonstrated on the assumption that it was formed by the algorithm using DHT, this invention is not limited to this. In addition, the distributed content storage system S in the above embodiment holds the cache of its own node as much as possible according to the degree of popularity of the entire network when participating in the system (network) or when storage becomes available. For example, the present invention can be applied to a content distribution system that is more advantageous, for example, a music distribution system at a karaoke store, a video distribution system at a hotel, or the like.

It is a figure which shows an example of the connection aspect of each node apparatus in the content distributed storage system which concerns on this embodiment. It is a figure which shows the example of an outline structure of the node Nn. It is a figure which shows an example of the content of a popular content list. It is a conceptual diagram which shows the method by which the node Nn acquires a popular content list from a management server. It is a conceptual diagram which shows the method acquired by the node Nn creating autonomously. It is a figure which shows an example of the content of the download list produced based on the popular content list shown in FIG. It is a figure which shows an example of the content of the download list produced based on the popular content list shown in FIG. It is a figure which shows an example of the content of the download list produced based on the popular content list shown in FIG. It is a flowchart which shows the process of the control part 11 in the node Nn.

Explanation of symbols

8 Network 9 Overlay Network 11 Control Unit 12 Storage Unit 13 Buffer Memory 14 Decoder Unit 15 Video Processing Unit 16 Display Unit 17 Audio Processing Unit 18 Speaker 19 Communication Unit 20 Input Unit 21 Bus Nn Node S Content Distributed Storage System

Claims (9)

A distributed content storage system comprising a plurality of node devices that can communicate with each other via a network, wherein copy data of a plurality of contents is distributed and stored in a plurality of node devices, and the content is used between the node devices The node device in the distributed content storage system that is enabled,
Participation processing means for performing participation processing for participating in the content distributed storage system;
Storage means having a storage area for storing the replicated data;
Content list acquisition means for acquiring a content list in which information about each content available in the content distributed storage system is described;
Free capacity determination means for determining whether or not there is a predetermined free capacity in the storage area in the storage means;
Content selection means for selecting, based on the content list, a plurality of contents corresponding to a predetermined data capacity when it is determined that the storage area has a predetermined free capacity when participating in the content distributed storage system;
A storage control unit that acquires the copy data of the selected content from another node device or a server device having copy data of all the content, and stores the copy data in the storage area of the storage unit;
A node device comprising: The node device according to claim 1,
The content selection means selects a predetermined amount of content based on the content list when a predetermined free space is created in the storage region by deleting the duplicate data stored in the storage region. A node device characterized by: The node device according to claim 1 or 2,
The node device according to claim 1, wherein the content selection means preferentially selects content having a high probability of being used. In the node apparatus as described in any one of Claims 1 thru | or 3,
The information related to each content includes at least one of the number of times the plurality of pieces of content data are acquired by any of a plurality of node devices and the number of times of reproduction of the copy data of the content by any of a plurality of node devices. Contains one piece of information,
Probability calculating means for calculating a probability proportional to at least one of the number of acquisitions and the number of reproductions for each content,
The node device, wherein the content selection means selects the content based on a probability calculated for each content. The node device according to claim 4, wherein
The information related to each content further includes information indicating at least one of the genre, the target generation, and the region attribute of the content,
The probability calculation means increases the probability of content that matches at least one of a genre, a generation, and a region attribute set in advance in the node device with reference to information on each content. Node device. In the node apparatus as described in any one of Claims 1 thru | or 5,
It further comprises content acquisition means for acquiring copy data of content designated by the user through the input means from another node device,
A node that prioritizes the copy data acquisition process by the content acquisition unit when the copy data acquisition process by the content acquisition unit and the copy data acquisition process by the storage control unit overlap apparatus.   A node processing program for causing a computer to function as the node device according to any one of claims 1 to 6. A distributed content storage system comprising a plurality of node devices that can communicate with each other via a network, wherein copy data of a plurality of contents is distributed and stored in a plurality of node devices, and the content is used between the node devices A content distribution and storage system that enables
The node device is
Participation processing means for performing participation processing for participating in the content distributed storage system;
Storage means having a storage area for storing the replicated data;
Content list acquisition means for acquiring a content list in which information about each content available in the content distributed storage system is described;
Free capacity determination means for determining whether or not there is a predetermined free capacity in the storage area in the storage means;
Content selection means for selecting, based on the content list, a plurality of contents corresponding to a predetermined data capacity when it is determined that the storage area has a predetermined free capacity when participating in the content distributed storage system;
A storage control unit that acquires the copy data of the selected content from another node device or a server device having copy data of all the content, and stores the copy data in the storage area of the storage unit;
A content distributed storage system comprising: A distributed content storage system comprising a plurality of node devices that can communicate with each other via a network, wherein copy data of a plurality of contents is distributed and stored in a plurality of node devices, and the content is used between the node devices A content storage method in a distributed content storage system,
A participation process step in which the node device executes a participation process for participating in the distributed content storage system;
A content list acquisition step in which the node device acquires a content list in which information about each content available in the content distributed storage system is described;
The node device determines whether or not there is a predetermined free space in a storage area for storing the replicated data; and
Content for selecting a plurality of contents corresponding to a predetermined data capacity based on the content list when it is determined that the storage area has a predetermined free capacity when participating in the distributed content storage system. A selection process;
Storage control in which the node device acquires copy data of the selected content from another node device or a server device having copy data of all content, and stores the copy data in the storage area in the storage unit Process,
The content storage method characterized by including.

Images