JP2007213149A - Distribution system, controller, distribution destination device, node device, program for control, program for distribution destination device and program for node device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a distribution system for improving reliability and convenience to a whole distribution system by starting the reproduction of a content in an early stage while validly utilizing a recording device in a node belonging to a distribution system even when distribution requests to content are concentrated. <P>SOLUTION: This distribution system is configured by connecting a plurality of nodes through a network to distribute content stored in a content holder CH to a requester RQ at the destination of distribution, and provided with a content holder CH for storing whole contents; an another content holder CH different from the content holder CH and for storing an opening replica including at least the opening section of the content ; and a route node RN for controlling from which of the content holder CH for storing the whole content and the content holder CH for storing the opening replica the distribution of content data should be started. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention belongs to a technical field of a distribution system, a control device, a distribution destination device, a node device, a control program, a distribution destination device program, and a node device program, and more specifically, a movie via a network such as the Internet. A distribution system that distributes content (distribution information) such as a control device that controls distribution of distribution information in the distribution system, a distribution destination device that receives distribution of distribution information in the distribution system, and a node included in the distribution system TECHNICAL FIELD Technical Field of a Device, a Control Program Used for Controlling Distribution Information in the Control Device, a Distribution Device Device Program that Controls Reception of Distribution Information in the Distribution Device, and a Node Device Program that Controls Operation in the Node Device Belonging to.

  In recent years, so-called content distribution in which a terminal device accesses a server or the like storing the above content via a network such as the Internet, and distributes the content desired to be viewed on the terminal device to the terminal device. Research and development is thriving.

  Here, the basic configuration of a conventional distribution system that distributes content is, for example, connected from a terminal device that desires content distribution to a server or the like that stores the content via the network, and establishes the connection. The terminal device is configured to receive distribution of the desired content, that is, transmission of the content.

  Here, in the case of content with a high distribution request (that is, highly popular), when the distribution request for the content is concentrated at once on a server or the like that stores the content (so-called access concentration occurs) Can be considered. As a countermeasure against access concentration in the conventional distribution system, for example, by distributing a plurality of (permanently) servers or the like themselves in the distribution system, the load is distributed or distributed. For specific content with a large number of requests, there has been a method in which a server or the like that temporarily stores (caches) the content is arranged in the distribution system.

  In view of this, a technique has been proposed in which a so-called distributed hash table (DHT (Distributed Hash Table)) is used to logically construct an overlay network to efficiently distribute and distribute information to achieve load distribution.

  In this system, each terminal device does not recognize link information (for example, IP (Internet Protocol) address) to all terminal devices participating in the overlay network. Only the link information to a part of terminal devices obtained at the time or the like is held, and the data is inquired based on the link information.

In the overlay network, load sharing needs to be appropriately performed even if the terminal device frequently participates in the network and withdraws (leaves) from the network. The following Non-Patent Document 1 discloses a technique for appropriately distributing load even when participation and withdrawal (withdrawal) are frequently performed in an overlay network.
"Examination of lightweight load balancing method for distributed hash table" The Institute of Electronics, Information and Communication Engineers IEICE technical report

  Here, in a so-called peer-to-peer type content distribution system in which the technology of Non-Patent Document 1 is used in a content distribution system, each terminal device is a content distribution device that holds content data, and each terminal device is a so-called content. It becomes a distribution request client.

  Even in such a system, when popularity is concentrated on a specific content, concentration of access to a specific terminal device becomes a problem. Therefore, as a countermeasure, a recording device (each of which is included in each terminal device constituting the distribution system) For example, a replica (duplicate) of electronic data corresponding to the content of the distribution request target is recorded in a hard disk device or the like in accordance with the frequency of the distribution request, thereby deteriorating the content distribution environment due to the access concentration (more specifically, In particular, the delay of the start of distribution, etc.) was prevented. In this case, it is effective for avoiding concentration of access if it is configured to keep as many kinds of content replicas as possible within a range not exceeding the capacity of the storage device of the terminal device.

  However, as a result, each terminal device often runs in a situation where the storage capacity in the storage device in each terminal device is almost exhausted and a small amount of free space is left. For example, when 16 6-gigabyte video content replicas are stored in a terminal device equipped with a 100-gigabyte hard disk, the storage capacity of the terminal device is up to 96 gigabytes, and the remaining 4 gigabytes of room are available.

  At this time, for example, if there is a recording capacity of 4 gigabytes in a certain terminal device, but the information amount of the entire content to be recorded on the replica is larger than 4 gigabytes, a replica of the content is created in the terminal device Cannot use the available recording capacity in the terminal device effectively, and it also hinders the processing to avoid the deterioration of the content distribution environment due to the concentration of access by creating a replica. .

  Therefore, the present application has been made in view of the above circumstances, and the purpose thereof is to effectively utilize the recording device in the terminal device belonging to the distribution system even when a large number of distribution requests for specific content are concentrated at once. However, a distribution system capable of contributing to improvement in reliability and convenience for the entire distribution system by starting reproduction of the specific content in the distribution destination terminal device at an early stage, and distribution of content in the distribution system A control device for controlling content, a distribution destination device that receives distribution of content in the distribution system, a terminal device included in the distribution system, a control program used for content control in the control device, and a content in the distribution destination device Program for destination device for controlling reception of data and operation in the terminal device To provide a control node device program.

  In order to solve the above-mentioned problem, the invention according to claim 1 is configured such that a plurality of node devices are connected via a network, and distribution information stored in any one of the node devices is stored. In a distribution system that distributes to a distribution destination device that is any other node device, an entire storage device that is the node device including distribution information storage means such as a recording unit that stores the entire distribution information. Partial distribution information storage means such as a recording unit that stores partial distribution information that is a part of the distribution information that is at least the beginning portion of the one distribution information and that is a node device different from the overall storage device. Distribution of the distribution information to the distribution destination device is started from either the partial storage device, which is the node device, and the total storage device or the partial storage device And a control unit of the control unit such as a said node device comprises control means for controlling the.

  Therefore, the partial storage device that stores partial distribution information including at least the beginning portion of the distribution information and the total storage device that stores the entire distribution information are included in the distribution system as separate node devices. Even if distribution of distribution information from the overall storage device to the distribution destination device is delayed due to a large number of distribution requests to the device, etc., partial distribution information before the distribution information in the entire storage device Is used for distribution, reproduction of the distribution information using the beginning portion can be quickly started in the distribution destination apparatus.

  In order to solve the above problem, the invention described in claim 2 is the control device included in the distribution system described in claim 1, and includes the control means.

  Therefore, the partial storage device that stores the partial distribution information including at least the beginning portion of the distribution information and the total storage device that stores the entire distribution information are included in the distribution system as separate node devices. Since it controls whether to start distributing distribution information, even if distribution information distribution from the entire storage device to the distribution destination device is delayed due to many distribution requests to the entire storage device, By using the partial distribution information for distribution before the distribution information in the entire storage device, reproduction of the distribution information using the beginning portion can be started quickly in the distribution destination device.

  In order to solve the above-mentioned problem, the invention according to claim 3 is the control device according to claim 2, wherein the control means is such that the number of distribution requests for distribution information is equal to or greater than a preset threshold value. The partial distribution information is controlled to be distributed from the partial storage device to the distribution destination device, and when the number of distribution requests is less than the threshold, the distribution information is transferred from the overall storage device to the distribution destination device. Control to deliver.

  Therefore, since the distribution source of the distribution information is determined based on the comparison with the preset threshold value, the distribution source can be separated with a simple configuration.

  In order to solve the above-described problem, the invention according to claim 4 is the control device according to claim 3, wherein the control means sets the partial storage device when the number of distribution requests is equal to or greater than the threshold. The response information indicating is transmitted to the distribution destination device, and when the number of distribution requests is less than the threshold, the response information indicating the entire storage device is transmitted to the distribution destination device.

  Therefore, since the response information indicating the distribution source is transmitted from the control device to the distribution destination device when the distribution source is determined, the distribution source can be easily recognized by the distribution destination device.

  In order to solve the above problem, the invention according to claim 5 is the control device according to claim 2, further comprising management means such as a control unit for managing the total storage device and the partial storage device, The control unit controls the partial storage device to distribute the partial distribution information to a distribution destination device when the partial storage device is managed by the management unit, and the partial storage device manages the management unit. If not, control is performed so that the distribution information is distributed from the overall storage device to the distribution destination device.

  Therefore, when the partial storage device is under the control of the control device, the partial distribution information is distributed from the partial storage device. By using a partial storage device that can wait for a distribution request from the destination device, it can handle distribution requests from multiple destination devices, compared to when there is only a total storage device under the control of the control device. Thus, the partial distribution information can be distributed to the distribution destination device to accelerate the reproduction start of the distribution information in the distribution destination device.

  In order to solve the above-mentioned problem, the invention according to claim 6 is the control device according to claim 5, wherein the control means is configured to control the partial storage device when the partial storage device is managed by the management means. Response information indicating a storage device is transmitted to the distribution destination device, and when the partial storage device is not managed by the management unit, response information indicating the entire storage device is transmitted to the distribution destination device. The

  Therefore, when the distribution source under management is determined, the response information indicating the distribution source is transmitted from the control device to the distribution destination device, so that the distribution source device can easily recognize the distribution source.

  In order to solve the above-described problem, the invention according to claim 7 is the control device according to claim 3 or 4, wherein request information for requesting delivery of the delivery information is acquired from the delivery destination device. An acquisition unit such as a control unit, and a control unit that newly accumulates the partial distribution information such as a control unit in another node device when the number of distribution requests based on the acquired request information is equal to or greater than the threshold. The partial distribution information accumulation control means.

  Therefore, the number of distribution requests for the entire storage device is confirmed, and control is performed so that partial distribution information is newly stored when the number of distribution requests is equal to or greater than the threshold value. It is possible to quickly respond to a distribution request from a distribution destination apparatus while effectively utilizing an empty storage area in the apparatus.

  In order to solve the above-mentioned problem, the invention according to claim 8 is the control device according to claim 7, wherein the partial distribution information accumulation control unit is capable of accumulating the partial distribution information. A node search means such as a control unit for searching for a device in the distribution system; and a part of the distribution information stored in the total storage device in the searched node device from the total storage device Copy control means such as a control unit for copying as partial distribution information.

  Therefore, when new partial distribution information is stored, the partial distribution information itself is not newly generated, but a portion corresponding to the partial distribution information in the distribution information stored in the entire storage device is copied. Since it is accumulated, new partial distribution information can be accumulated quickly without causing a heavy load.

  In order to solve the above-mentioned problem, the invention according to claim 9 is the control device according to claim 8, wherein the node search means determines whether the partial distribution information can be stored in another node device. An inquiry unit such as a control unit that inquires, and a specifying unit such as a control unit that specifies the node device capable of storing the partial distribution information based on a response result to the inquiry by the inquiry unit. The copy control means transmits overall device information indicating the overall storage device to the specified node device, and causes the specified node device to receive the part from the overall storage device as the partial distribution information. Configured as follows.

  Therefore, since the entire device information is transmitted to the node device specified by the inquiry and the partial distribution information is stored in the specified node device, the necessary partial distribution information can be stored with a simple configuration. .

  In order to solve the above-described problem, the invention according to claim 10 is the control device according to any one of claims 2 to 9, wherein the number of distribution requests for the distribution information is less than a preset threshold value. Then, the image processing apparatus further includes an erasure control unit such as a control unit for erasing the partial distribution information stored in the partial distribution information storage unit from the partial distribution information storage unit.

  Therefore, when the busyness with respect to the distribution processing of the distribution information in the entire storage device decreases, control is performed so that the partial distribution information that is no longer needed is deleted from within the partial storage device, so the partial storage in which the partial distribution information is deleted The storage area in the apparatus can be diverted to other uses, and the distribution request from the distribution destination apparatus can be quickly handled while effectively utilizing the storage capacity of the entire distribution system.

  In order to solve the above-described problem, the invention according to claim 11 is the delivery destination device included in the delivery system according to claim 1, wherein the delivery information for which delivery to the delivery destination device is desired. When the partial device information indicating the partial storage device is included in the sending means such as a communication unit for transmitting the distribution request information indicating, and the response information transmitted corresponding to the distribution request information, A partial distribution receiving unit such as a communication unit that receives the distribution of the partial distribution information from the partial storage device indicated by the partial device information; and a reproduction unit such as a decoder that reproduces the distributed partial distribution information. .

  Therefore, when the partial response information is received from the corresponding partial storage device when the partial response information is included in the received response information, the distribution request to the entire storage device existing in the distribution system is large. However, the reproduction of the distribution information can be quickly started in the distribution destination device.

  In order to solve the above-described problem, the invention according to claim 12 is the delivery destination device according to claim 11, wherein the partial delivery information after a preset time has elapsed from the reproduction start time of the partial delivery information. A re-sending means such as a communication unit that re-sends the distribution request information before the end of reproduction, and the overall storage device in the reply information transmitted in response to the re-sent distribution request information. When the entire device information to be included is included, further comprising: an entire distribution receiving unit such as a communication unit that receives distribution of the distribution information from the entire storage device indicated by the entire device information. After the reproduction of the partial distribution information is completed, the distribution information is reproduced continuously from the portion of the distributed distribution information following the partial distribution information.

  Therefore, the distribution request information is retransmitted from the start of the reproduction of the partial distribution information to the end of the reproduction after the elapse of a preset time, and the entire distribution information distributed corresponding to the retransmitted distribution request information is received. Since the reproduction is continued after the partial distribution information, even when the distribution request to the entire storage device is large, the entire reproduction can be performed while promptly starting the reproduction of one distribution information.

  In order to solve the above problem, an invention according to claim 13 is the delivery destination device included in the delivery system according to claim 1, wherein the delivery information for which delivery to the delivery destination device is desired. A transmission means such as a communication unit for transmitting distribution request information for requesting at least one of the distribution information corresponding to the entire distribution information or at least one of the partial distribution information, and transmitted corresponding to the distribution request information. A distribution receiving means such as a communication unit that receives distribution of the distribution information from either the whole storage device or the partial storage device indicated by the received response information, and a decoder unit that reproduces the distributed distribution information. And reproducing means.

  Therefore, distribution information (either distribution information as a whole or partial distribution information) is received from the corresponding partial storage device or total storage device when the received response information includes partial device information or total device information. Therefore, if the distribution request to the entire storage device existing in the distribution system is large, the distribution information is distributed from the available partial storage device, and the reproduction of the distribution information is quickly started in the distribution destination device. can do.

  Therefore, even when the distribution request to the entire storage device that stores the entire distribution information that should be distributed in the distribution system is busy, the reproduction of the distribution information in the distribution destination device can be started. Therefore, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  In order to solve the above-described problem, the invention described in claim 14 is any one of the node devices included in the distribution system according to claim 1, wherein the storage device such as a recording unit and the partial distribution are provided. A receiving unit such as a communication unit that receives inquiry information for inquiring whether information can be stored, and the storage unit has a free area that is equal to or greater than the information amount of the partial distribution information and less than the total information amount of the distribution information. And a response result sending means such as a communication unit for sending a response result indicating that the partial delivery information can be stored in the delivery system when the whole delivery information is not stored in the storage means. And comprising.

  Therefore, when the storage means in any of the node devices in the distribution system has an empty area that is equal to or greater than the information amount of the partial distribution information and less than the information amount of the entire distribution information, and the entire distribution information is not accumulated. Since the response result indicating that the partial distribution information can be stored is sent out, the partial distribution information is accumulated by effectively utilizing the empty area in each node device in the distribution system, so that the entire distribution system is free. The capacity can be used effectively.

  In order to solve the above-described problem, the invention described in claim 15 provides a computer included in any one of the node devices included in the distribution system described in claim 1. Function as the control device described in the section.

  Therefore, when the computer is caused to function as the control device according to claim 2, a partial storage device that stores partial distribution information including at least the beginning portion of the distribution information and a total storage device that stores the entire distribution information include: This is because it is included in the distribution system as a separate node device, and the computer functions to control from which distribution information distribution is started. Even if distribution of distribution information from the entire storage device to the distribution destination device is delayed, the first part is used by using partial distribution information for distribution before the distribution information in the entire storage device. The reproduction of the distribution information can be started quickly at the distribution destination device.

  When the computer functions as the control device according to claim 3, in addition to the function as the control device according to claim 2, the distribution source of the distribution information is set based on comparison with a preset threshold value. Since the computer functions as determined, the distribution source can be separated with a simple configuration.

  Further, when the computer functions as the control device according to claim 4, in addition to the function as the control device according to claim 3, when the distribution source is determined, the control device returns a response indicating the distribution source. Since the computer functions to transmit information to the distribution destination apparatus, the distribution destination apparatus can easily recognize the distribution source.

  Furthermore, when the computer functions as the control device according to claim 5, in addition to the function as the control device according to claim 2, when the partial storage device is under the control of the control device, Since the computer functions to distribute the partial distribution information from the partial storage device, the distribution is completed in a short time compared to the case where the entire distribution information is distributed, and the distribution request waiting state from other distribution destination devices By using a partial storage device that can be configured, the distribution destination device can receive partial distribution information in response to distribution requests from a plurality of distribution destination devices, compared to the case where there is only a whole storage device under the control of the control device. And the reproduction start of the distribution information in the distribution destination device can be accelerated.

  When the computer functions as the control device according to claim 6, in addition to the function as the control device according to claim 5, when the distribution source under management is determined, the distribution is performed from the control device. Since the computer functions so as to send response information indicating the origin to the distribution destination apparatus, the distribution destination apparatus can easily recognize the distribution source.

  Further, when the computer functions as the control device according to claim 7, in addition to the function as the control device according to claim 3 or 4, the number of distribution requests to the entire storage device is confirmed, Since the computer functions so as to control to newly store partial distribution information when the threshold is equal to or greater than the threshold value, the free storage area in other node devices is made effective in accordance with the busyness of the distribution request to the entire storage device. It is possible to respond promptly to a distribution request from a distribution destination device while using it.

  Furthermore, when the computer functions as the control device according to claim 8, in addition to the function as the control device according to claim 7, when the partial distribution information is accumulated, the partial distribution information is stored. The computer functions to copy and store the portion corresponding to the partial distribution information in the distribution information stored in the entire storage device, rather than generating a new one, resulting in high load. New partial distribution information can be accumulated quickly.

  When the computer is caused to function as the control device according to claim 9, in addition to the function as the control device according to claim 8, the entire device information is transmitted to the node device specified by the inquiry. In addition, since the computer functions to store the partial distribution information in the identified node device, the necessary partial distribution information can be stored with a simple configuration.

  Further, when the computer is caused to function as the control device according to claim 10, in addition to the case where the computer is caused to function as the control device according to any one of claims 2 to 9, distribution of distribution information in the entire storage device When the busyness with respect to processing decreases, the computer functions so as to control the partial distribution information that is no longer needed to be deleted from the partial storage device, so the storage in the partial storage device from which the partial distribution information has been deleted The area can be diverted to other uses, and the distribution request from the distribution destination apparatus can be promptly handled while effectively utilizing the storage capacity of the entire distribution system.

  In order to solve the above-described problem, the invention described in claim 16 provides a computer included in any one of the node devices included in the distribution system described in claim 1. Function as the delivery destination device described in the section.

  Therefore, when the computer is caused to function as the delivery destination device according to claim 11, the delivery of the partial delivery information is received from the corresponding partial storage device when the partial device information is included in the received delivery information. Since the computer functions in this manner, even when the distribution request to the entire storage device existing in the distribution system is large, reproduction of the distribution information can be quickly started in the distribution destination device.

  Further, when the computer is caused to function as the delivery destination device according to claim 12, in addition to the case where the computer is caused to function as the delivery destination device according to claim 11, the computer is set in advance from the start of reproduction of the partial delivery information. The distribution request information is retransmitted between the elapse of the specified time and the end of the reproduction, the entire distribution information distributed in response to the retransmitted distribution request information is received, and the reproduction is continued following the partial distribution information. In addition, since the computer functions, even when the distribution request to the entire storage device is large, it is possible to reproduce the whole while quickly starting reproduction of one piece of distribution information.

  Furthermore, when the computer is caused to function as the distribution device according to claim 13, distribution is performed from the corresponding partial storage device or total storage device when the received response information includes partial device information or total device information. Since the computer functions to receive distribution of information (either distribution information as a whole or partial distribution information), if the distribution request to the entire storage device existing in the distribution system is large, the partial storage device that is free By receiving the distribution of the partial distribution information, the reproduction of the distribution information can be quickly started in the distribution destination device.

  Therefore, even when the distribution request to the entire storage device that stores the entire distribution information that should be distributed in the distribution system is busy, the reproduction of the distribution information in the distribution destination device can be started. Therefore, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  In order to solve the above-described problem, the invention described in claim 17 is configured such that a computer included in any one of the node devices included in the distribution system described in claim 1 is used as the node device described in claim 14. Make it work.

  Therefore, when the storage means in any of the node devices in the distribution system has an empty area that is equal to or greater than the information amount of the partial distribution information and less than the information amount of the entire distribution information, and the entire distribution information is not accumulated. Since the computer functions so as to send a response result indicating that partial distribution information can be stored, it is possible to accumulate the partial distribution information by effectively utilizing the free area in each node device in the distribution system. Therefore, it is possible to effectively utilize the free capacity of the distribution system as a whole.

  As described above, according to the first aspect of the present invention, the partial storage device that stores the partial distribution information including at least the beginning portion of the distribution information and the total storage device that stores the entire distribution information are separated. Since it is included in the distribution system as a node device, even if the distribution of distribution information from the entire storage device to the distribution destination device is delayed due to many distribution requests to the entire storage device, By using the partial distribution information for distribution before the distribution information in the entire storage device, reproduction of the distribution information using the beginning portion can be started quickly in the distribution destination device.

  According to the second aspect of the present invention, the partial storage device that stores the partial distribution information including at least the beginning portion of the distribution information and the total storage device that stores the entire distribution information are distributed as separate node devices. Since it is included in the system and from which distribution information distribution is started, the distribution information from the entire storage device to the distribution destination device due to many distribution requests to the entire storage device, etc. Even in the case where the delivery is delayed, by using the partial delivery information for delivery prior to the delivery information in the overall storage device, the delivery destination device can quickly reproduce the delivery information using the beginning portion. Can start.

  Therefore, according to the first and second aspects of the present invention, even in the distribution system, even when the distribution request to the entire storage device that stores the entire distribution information to be distributed is busy, the distribution destination device Since the reproduction of the distribution information can be started, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  According to the third aspect of the invention, in addition to the effect of the second aspect of the invention, the distribution source of the distribution information is determined based on a comparison with a preset threshold value. Can be separated.

  According to the invention described in claim 4, in addition to the effect of the invention described in claim 3, when the distribution source is determined, the control device transmits response information indicating the distribution source to the distribution destination device. The delivery source can be easily recognized in the destination device.

  According to the invention described in claim 5, in addition to the effect of the invention described in claim 2, when the partial storage device is under the control of the control device, the partial distribution information is distributed from the partial storage device. By using a partial storage device that can complete delivery in a short time and wait for a delivery request from another delivery destination device compared to delivering the entire delivery information, it is under the control of the control device. Compared to the case where there is only a whole storage device, partial distribution information can be distributed to the distribution destination device in response to distribution requests from a plurality of distribution destination devices, and the reproduction start of the distribution information in the distribution destination device can be accelerated. .

  According to the sixth aspect of the invention, in addition to the effect of the fifth aspect of the invention, when the distribution source under management is determined, response information indicating the distribution source is transmitted from the control device to the distribution destination device. Therefore, the distribution source can be easily recognized in the distribution destination device.

  According to the invention described in claim 7, in addition to the effect of the invention described in claim 3 or 4, the number of distribution requests to the entire storage device is confirmed, and when it is equal to or greater than the threshold, the partial distribution information is newly stored. Therefore, it is possible to respond quickly to the distribution request from the distribution destination device while effectively utilizing the free storage area in the other node devices in accordance with the busyness of the distribution request to the entire storage device.

  According to the invention described in claim 8, in addition to the effect of the invention described in claim 7, when new partial distribution information is accumulated, the partial distribution information itself is not newly generated, Since the portion corresponding to the partial distribution information in the distribution information stored in the storage device is copied and stored, new partial distribution information can be stored quickly without causing a high load.

  According to the ninth aspect of the invention, in addition to the effect of the eighth aspect of the invention, the entire device information is transmitted to the node device specified by the inquiry, and the partial device information is included in the specified node device. Since the distribution information is accumulated, the necessary partial distribution information can be accumulated with a simple configuration.

  According to the invention described in claim 10, in addition to the effect of the invention described in any one of claims 2 to 9, when the busyness with respect to the distribution processing of the distribution information in the entire storage device is reduced, it is unnecessary. Since the partial distribution information is controlled so as to be deleted from the partial storage device, the storage area in the partial storage device from which the partial distribution information has been deleted can be used for other purposes and stored as a whole distribution system. It is possible to respond quickly to a distribution request from a distribution destination device while effectively utilizing the capacity.

  According to the eleventh aspect of the present invention, since partial distribution information is received from the corresponding partial storage device when partial device information is included in the received response information, the entire storage existing in the distribution system is received. Even when the distribution request to the device is large, reproduction of the distribution information can be quickly started in the distribution destination device.

  Therefore, even when the distribution request to the entire storage device that stores the entire distribution information that should be distributed in the distribution system is busy, the reproduction of the distribution information in the distribution destination device can be started. Therefore, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  According to the twelfth aspect of the invention, in addition to the effect of the eleventh aspect of the invention, the distribution request information is retransmitted from the start of reproduction of the partial distribution information to the end of the reproduction after the preset time has elapsed. Since the entire distribution information distributed corresponding to the retransmitted distribution request information is received and the reproduction is continued following the partial distribution information, even if the distribution request to the entire storage device is large, one It is possible to reproduce the entire distribution information while quickly starting reproduction.

  According to the thirteenth aspect of the present invention, distribution information (entire distribution information or partial distribution information) is received from a corresponding partial storage device or total storage device when partial device information or total device information is included in the received response information. If any of the distribution requests to the entire storage device existing in the distribution system is large, the distribution device receives the partial distribution information from the available partial storage device and places it in the distribution destination device. Thus, the reproduction of the distribution information can be started quickly.

  Therefore, even when the distribution request to the entire storage device that stores the entire distribution information that should be distributed in the distribution system is busy, the reproduction of the distribution information in the distribution destination device can be started. Therefore, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  According to the fourteenth aspect of the present invention, the storage means in any of the node devices in the distribution system has an empty area that is equal to or greater than the information amount of the partial distribution information and less than the total information amount of the distribution information, and is distributed. When the entire information is not accumulated, a response result indicating that the partial distribution information can be stored is transmitted, so that the partial distribution information is accumulated by effectively utilizing the free space in each node device in the distribution system. Thus, the free capacity of the distribution system as a whole can be used effectively.

  According to the fifteenth aspect of the present invention, when the computer included in any of the node devices included in the distribution system according to the first aspect functions as the control device according to the second aspect, the beginning of the distribution information A partial storage device that stores partial distribution information including at least a portion and an entire storage device that stores the entire distribution information are included in the distribution system as separate node devices, and distribution information distribution starts from either of them. Even if the distribution of distribution information from the entire storage device to the distribution destination device is delayed due to a large number of distribution requests to the entire storage device, etc. By using the partial distribution information for distribution before the distribution information in the entire storage device, the distribution destination device can reproduce the distribution information using the beginning portion. It is possible to start quickly.

  Therefore, even when the distribution request to the entire storage device that stores the entire distribution information that should be distributed in the distribution system is busy, the reproduction of the distribution information in the distribution destination device can be started. Therefore, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  When the computer functions as the control device according to claim 3, in addition to the function as the control device according to claim 2, the distribution of distribution information based on comparison with a preset threshold value Since the computer functions so as to determine the origin, the distribution source can be separated with a simple configuration.

  Further, when the computer functions as the control device according to claim 4, in addition to the function as the control device according to claim 3, when the distribution source is determined, the control device returns a response indicating the distribution source. Since the computer functions to transmit information to the distribution destination apparatus, the distribution destination apparatus can easily recognize the distribution source.

  Furthermore, when the computer functions as the control device according to claim 5, in addition to the function as the control device according to claim 2, when the partial storage device is under the control of the control device, Since the computer functions to distribute the partial distribution information from the partial storage device, the distribution is completed in a short time compared to the case where the entire distribution information is distributed, and the distribution request waiting state from other distribution destination devices By using a partial storage device that can be configured, the distribution destination device can receive partial distribution information in response to distribution requests from a plurality of distribution destination devices, compared to the case where there is only a whole storage device under the control of the control device. And the reproduction start of the distribution information in the distribution destination device can be accelerated.

  When the computer functions as the control device according to claim 6, in addition to the function as the control device according to claim 5, when the distribution source under management is determined, the distribution is performed from the control device. Since the computer functions so as to send response information indicating the origin to the distribution destination apparatus, the distribution destination apparatus can easily recognize the distribution source.

  Further, when the computer functions as the control device according to claim 7, in addition to the function as the control device according to claim 3 or 4, the number of distribution requests to the entire storage device is confirmed, Since the computer functions so as to control to newly store partial distribution information when the threshold is equal to or greater than the threshold value, the free storage area in other node devices is made effective in accordance with the busyness of the distribution request to the entire storage device. It is possible to respond promptly to a distribution request from a distribution destination device while using it.

  Furthermore, when the computer functions as the control device according to claim 8, in addition to the function as the control device according to claim 7, when the partial distribution information is accumulated, the partial distribution information is stored. The computer functions to copy and store the portion corresponding to the partial distribution information in the distribution information stored in the entire storage device, rather than generating a new one, resulting in high load. New partial distribution information can be accumulated quickly.

  When the computer is caused to function as the control device according to claim 9, in addition to the function as the control device according to claim 8, the entire device information is transmitted to the node device specified by the inquiry. In addition, since the computer functions to store the partial distribution information in the identified node device, the necessary partial distribution information can be stored with a simple configuration.

  Further, when the computer is caused to function as the control device according to claim 10, in addition to the case where the computer is caused to function as the control device according to any one of claims 2 to 9, distribution of distribution information in the entire storage device When the busyness with respect to processing decreases, the computer functions so as to control the partial distribution information that is no longer needed to be deleted from the partial storage device, so the storage in the partial storage device from which the partial distribution information has been deleted The area can be diverted to other uses, and the distribution request from the distribution destination apparatus can be promptly handled while effectively utilizing the storage capacity of the entire distribution system.

  According to the invention described in claim 16, when the computer included in any of the node devices included in the distribution system according to claim 1 functions as the distribution destination device according to claim 11, the received response Since the computer functions so as to receive distribution of partial distribution information from a corresponding partial storage device when partial device information is included in the information, the distribution request to the entire storage device existing in the distribution system is large. Even in some cases, the reproduction of the distribution information can be quickly started in the distribution destination apparatus.

  Therefore, even when the distribution request to the entire storage device that stores the entire distribution information that should be distributed in the distribution system is busy, the reproduction of the distribution information in the distribution destination device can be started. Therefore, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  Further, when the computer is caused to function as the delivery destination device according to claim 12, in addition to the case where the computer is caused to function as the delivery destination device according to claim 11, the computer is set in advance from the start of reproduction of the partial delivery information. The distribution request information is retransmitted between the elapse of the specified time and the end of the reproduction, the entire distribution information distributed in response to the retransmitted distribution request information is received, and the reproduction is continued following the partial distribution information. In addition, since the computer functions, even when the distribution request to the entire storage device is large, it is possible to reproduce the whole while quickly starting reproduction of one piece of distribution information.

  Further, when the computer is caused to function as the delivery destination device according to claim 13, when the received response information includes the partial device information or the entire device information, the corresponding partial storage device or the entire storage device is used. Since the computer functions to receive distribution information (either distribution information as a whole or partial distribution information), if there are a large number of distribution requests to the entire storage device existing in the distribution system, the partial storage that is free By receiving the distribution of the partial distribution information from the apparatus, the reproduction of the distribution information can be quickly started in the distribution destination apparatus.

  Therefore, even when the distribution request to the entire storage device that stores the entire distribution information that should be distributed in the distribution system is busy, the reproduction of the distribution information in the distribution destination device can be started. Therefore, it is possible to contribute to improvement of reliability and convenience for the entire distribution system.

  According to the invention described in claim 17, when a computer included in any of the node devices included in the distribution system according to claim 1 is caused to function as the node device according to claim 14, The storage means in any one of the node devices stores the partial distribution information when there is an empty area that is not less than the information amount of the partial distribution information and less than the total amount of the distribution information and the entire distribution information is not stored. Since the computer functions to send a response result indicating that it is possible, the distribution system as a whole can be obtained by storing the partial distribution information by effectively utilizing the free space in each node device in the distribution system. Free space can be used effectively.

  Next, the best mode for carrying out the present invention will be described with reference to the drawings. The embodiment described below is an embodiment when the present invention is applied to the P2P distribution system that distributes the content using a network such as the Internet. In the following description, the terminal device is generally referred to as a “node”.

Initially the principle of (I) the present invention, before the specifically described embodiments of the present invention, per the principles of the present invention, P2P type distribution system according to the embodiment (hereinafter, simply referred to as a delivery system) with Overview explain.

(A) Overview of Distribution System First, an overview of a distribution system according to an embodiment will be described with reference to FIGS. 1 and 2. 1 and 2 are schematic diagrams showing an outline of the distribution system.

  In general, when a specific node connected to the network acquires content stored in another node from the other node, it is necessary to know information about which node stores which content. There is. For example, when the network is the Internet, it recognizes a set of index information (for example, a title name of a movie) that can identify the content and the IP address of the node storing the content, and uses the IP address as a clue for the desired content. It is necessary to give and receive. Therefore, in a distribution system in which an unspecified number of nodes share content such as the distribution system according to the embodiment, each node recognizes index information of all the contents and the IP address of the node storing the contents. It is necessary to keep it.

  However, when the number of nodes connected to the network increases, the index information of all contents and the nodes storing the contents are stored because of the limitation of the physical recordable amount in each node. It is not realistic for each node to record the IP address of each node (for example, when 1 million nodes are connected to the network, the IP addresses for 1 million nodes are all connected to each other). It's not realistic to record it).

  Also, when each node records the index of all contents and the IP address of the node storing it, the power switch at each node in the network is frequently turned on / off (for example, the node is set by a personal computer). In this case, the power switch can be frequently operated). However, the IP address and the like recorded in each node are frequently updated, and it is practically difficult to operate the entire network.

  Therefore, in order to cope with the problems as described above, only the IP address of the minimum necessary node is recorded, and for other nodes that do not recognize the IP address, the contents to be distributed to the node are stored. A distribution system having a mechanism of transferring and delivering via another node has been studied, and one of them is a distribution system using a so-called DHT (Distributed Hash Table).

  Next, an outline of the distribution system using the DHT will be described.

In the distribution system using the DHT, a node ID (Identification) for identifying each node is assigned to each node. At this time, the node ID is assigned a unique number for each node, that is, a number different from other nodes. This number has a number of bits (bit length) that can accommodate the maximum number of nodes in the network. More specifically, for example, when a 128-bit node ID is used, a maximum of 2 ¹²⁸ = 340 × 10 ³⁶ nodes can be connected in one network. The node ID is generally a hash value for a unique value for each node, such as an IP address assigned to the node itself, a so-called MAC (Media Access Control) address, or a serial number of the node itself. Use the value obtained by applying the function.

  Next, in the distribution system using the DHT, a unique content ID different from other contents is assigned to the content itself distributed by the distribution system. The bit length of the content ID is also the same as the bit length of the node ID. In general, the content ID includes title data indicating the title of the content, attribute data indicating the attributes of the data constituting the content, data for several bytes from the top of the data configuring the content, and the like. A value obtained by applying a hash function is used.

  Here, in relation to the fact that corresponding node IDs and content IDs are expressed with the same bit length, the nodes and content are assumed to be scattered in the same ring-shaped virtual ID space. Can think. That is, as shown in FIG. 1A in which one node is displayed with ● and one content is displayed with ○, the ring Rc to which the content belongs and the ring Rn to which the node belongs are virtually concentrically assumed. Further, if it is defined that the value of each ID increases counterclockwise in each ring Rc and Rn, it can be assumed that each node or each content exists without overlapping on each ring Rc and Rn. If each ID is expressed in 128-bit length as in the above example, the number of digits is too large. In FIG. 1A, for the sake of simplicity of explanation, the bit length of each ID is set to 32 bits. Expressed (the same applies hereinafter). Then, as described above, due to the use of the hash function in determining the value of each ID, each node and each content are present in a substantially distributed manner on each of the above rings Rc and Rn. Become.

  On the other hand, in the distribution system using the DHT, “a node that manages content to which a certain content ID is assigned is a node to which a node ID having a value closest to the value of the content ID is assigned”. ing.

  Here, “near” may be any definition as long as it is consistent in various provisions related to the distribution system. Specifically, for example, “a value that does not exceed the value of the content ID”. And having the smallest difference between values as IDs ”.

  More specifically, for example, as illustrated in FIG. 1A, a node having a node ID value “A0334055” and a node having a node ID value “A03340FF” are adjacent to each other on the ring Rn. If it exists, the content ID value “A0334080” is managed by the node whose node ID value is “A0334055”. In FIG. 1A, the terminal information for managing the content of “◯” is indicated by an arrow extending from each “◯”. In this way, various contents can be managed by being distributed among a large number of nodes.

  At this time, “management” does not mean that the content assigned with the content ID is recorded therein, but recognizes the location of the node where the content is recorded (for example, an IP address). It means that The node where the content is actually recorded may be different from the managed node, or the content to be managed may be recorded in the managed node.

  The above-described content management node is referred to as a “root node”. The root node records at least index information consisting of a pair of a content ID indicating the content that is to be managed and the IP address of the node in which the content indicated by the content ID is recorded. Is recorded so that it can be referred to by other nodes in the distribution system. In some cases, the index information includes the title of the content, the attribute (genre), and the like.

  In addition, when the content IDs indicating different contents are close to coincidence, and there is no other node having a node ID close to that value, a plurality of indexes corresponding to a plurality of contents by one root node Information will be recorded. Furthermore, when the same content is recorded in a plurality of different nodes and the node recording the same content is close to one root node by chance, The index information of a plurality of nodes each recording content is recorded. An example of the index information EX in the root node described above is shown in FIG.

  Further, a node that actually records each content itself is referred to as a “content holder”. At this time, since the content holder itself is essentially a node, the node ID exists on the ring Rn in FIG.

  Next, in the distribution system using the DHT, the DHT (routing table) itself, which is one of the devices for eliminating the necessity for each node to record the IP addresses of all the nodes as described above, will be described. To do.

  In the DHT according to the embodiment, the ID space illustrated in FIG. 1 is subdivided while increasing the number of levels for each stepped level (for example, in the case of level 1, the ID space is divided into four). Routing information that describes a pair of a node ID of an arbitrary node and its IP address is described for each area (for each area subdivided in stages). For example, when the distribution source inquiry information for requesting the content distribution source information is transferred to the root node in which the index information is recorded, the inquiry information is transferred to the target root node while referring to the routing information. To do. That is, each time the level in DHT increases, the node ID of the routing destination matches the node ID of the root node to be reached one digit at a time, and the request information finally reaches the root node to be reached. Will do.

  In the distribution system using DHT, the inquiry information (query) as described above, distribution request information for requesting distribution of the content itself, a registration message described later, and the like are simply referred to as “message”. When the DHT having the above-described mechanism is used, the message can be efficiently transferred to the target node in the ID space as illustrated in FIG. In the following description, when a new content is recorded in the content holder, a message generated in the content holder is referred to as a registration message so that it can be discovered (published) from other nodes. A message transmitted from a requester described later to a root node or a cache node described later is referred to as a distribution source inquiry message.

  Here, as described above, since the value of the node ID of the root node is closest to the value of the node ID of the content holder that manages the content by the root node, the distribution source inquiry message is the distribution request. Immediately before the content holder requested by the message is reached, the root node that manages the content to be distributed is reached. Then, the root node that has received the distribution source inquiry message returns the IP address of the content holder storing the content requested by the distribution source inquiry message to the transmission source of the distribution source inquiry message. That is, from the index information (see FIG. 1B) recorded in the root node, the title indicating the content requested by the distribution source inquiry message, the attribute information, and the IP of the node as its content holder The address or the like is returned to the node that is the source of the distribution source inquiry message. By utilizing such DHT, the location of the desired content can be efficiently recognized at the distribution request source, for example. The node that is the content distribution request source described above is hereinafter referred to as a “requester”.

  Next, a description will be given of content registration processing when new content is accumulated in a certain node.

  When new content is stored in a node, that is, the content holder for that content, the content holder will publish that the content has been stored to other nodes in the distribution system.

  That is, when new content is stored in a certain node, the node that is a content holder for the stored content calculates a content ID corresponding to the content based on the title of the stored content.

  Next, the content holder transmits a registration message with a node having a node ID having the same value as the calculated content ID as a destination (regardless of whether or not the node having the node ID actually exists). This registration message includes the title indicating the content, the attribute information, the IP address of the node as the content holder, and the like, and is sent to the node as the destination of each level via each routing destination according to the description of the DHT. Sequentially transferred.

  Then, when the node having the closest node ID is reached, the node recognizes that there is no node in the ID space to which the registration message is to be transferred afterwards ( Recognizing that the node to be transferred is itself), the node to which the registration message arrives at this point becomes the root node for managing new content. Then, the node that has become the root node records the content ID and the IP address indicating the content holder, the attribute information, and the like included in the registration message as index information EX illustrated in FIG. 1B.

  Next, the concept of “cache node” in the distribution system according to the embodiment will be described with reference to FIG. In FIG. 2 and FIG. 3 to be described later, the root node is indicated by a double circle, the content holder is indicated by a circle marked with “X”, and the requester is indicated by a circle marked with “|” inside, which will be described later. A cache node is indicated by a circle with + in the inside, and other general nodes are indicated by a circle.

  When the distribution source inquiry message Qr from the requesters RQ1 to RQ3 is transferred to the root node RN (or content holder CH) by the routing using the DHT described above, the transfer route (in FIGS. 2A and 2B) 2 (a) is similar to the path followed by the registration message Qp (shown by the solid line in FIGS. 2 (a) and 2 (b)) as shown in FIG. If the index information EX exemplified in FIG. 1B is also recorded in each node on the path (indicated by reference numeral CN in FIG. 2A), it corresponds to the distribution source inquiry message Qr passing through the path. Before the delivery source inquiry message Qr reaches the destination route node RN or the content holder CH. The response corresponding to the message Qr can be returned to the requesters RQ1 to RQ3.

  That is, the requesters RQ1 to RQ3 that request distribution of the content published in the distribution system transmit the distribution source inquiry message Qr that requests the IP address of the node (that is, the content holder CH) in which the content is recorded. Then, the distribution source inquiry message Qr is transferred to the routing destination described in the DHT in each node and approaches the root node RN. Finally, the route merges with the route of the registration message Qp from the content holder CH and reaches the root node RN.

  Here, when the path on the ring Rc shown in FIG. 2A is replaced with a tree structure, it becomes as shown in FIG. 2B, and this is generally called a “spanning tree”.

  As shown in FIG. 2A, the route of the distribution source inquiry message Qr transmitted from each requester RQ and the route of the registration message Qp transmitted from the content holder CH are in any node on these routes. Eventually cross. Accordingly, the index information EX included in the registration message Qp is recorded in a temporarily recordable temporary recording area at each node on the route for reaching the root registration message Qp to the root node RN ( That is, the registration message Qp is transferred (temporarily cached). Then, a reply to the distribution source inquiry message Qr from the requester RQ is made before the distribution source inquiry message Qr reaches the root node RN from the node where the distribution source inquiry message Qr on the route to the root node RN has arrived. It can be done. A node that is a node other than the root node RN and temporarily stores the index information EX in this manner is hereinafter referred to as a “cache node” CN.

(B) Principle of the present invention Next, the principle of the present invention based on the above-described distribution system will be described with reference to FIG.

  The content recorded in the content holder CH in the distribution system described above is originally the entire content (that is, the entire movie if the content is, for example, a single movie. Hereinafter, the entire content will be described). Is referred to as “whole content”).

  However, in the present invention, in addition to the whole content, it corresponds to a time set in advance from the beginning of the whole content (for example, in the case of the movie, for the first 5 minutes or 10 minutes of the movie). A copy of the portion (replica) is recorded in a node other than the content holder CH that records the entire content (a replica of only the beginning portion is hereinafter referred to as “beginning replica”).

  It is also possible to use a content other than the content for a preset time as described above as the opening portion. That is, for example, when the entire content is album content including a plurality of songs, the first song of the album content may be handled as the opening portion.

  If the entire content is popular content and the distribution request messages for the entire content are concentrated at a time, etc., the distribution of the entire content itself to the requester is delayed. Instead, the corresponding initial replica (ie, containing the same initial portion) is delivered to the requester. Thereby, the distribution start of the content in the requester is quickly performed.

  In addition, even if the distribution of the entire content is not delayed, for example, when it is desired to continuously play back the beginning of various contents, the distribution of the opening replica in the distribution request message itself from the requester Is first described, and only the first replica is first delivered to the requester.

  Therefore, as the index information in the distribution system according to the present invention, the information included in the index information EX shown in FIG. 1B (for example, FIG. In addition to the title, genre, or director name exemplified in b), or instead of the part other than the IP address, the content recorded in the content holder CH having the IP address is the above-mentioned initial replica. An initial replica flag indicating whether or not there is described. In FIG. 3, index information EXX corresponding to a plurality of contents is recorded in a root node RN as one information processing apparatus. Further, for the entire contents corresponding to one content and the initial replica, The index information EXX which shows the content holder CH which records an opening replica each has illustrated the state currently recorded including the said opening replica flag.

  Further, the index information EXX is recorded not only in the root node RN shown in FIG. 3 but also in the cache node CN as the information processing apparatus described above, and by referring to this, the requester RQ The return information RT corresponding to the distribution source inquiry message Qr can be returned to the requester RQ without the distribution source inquiry message Qr as the request information reaching the root node RN.

  Thus, the index information EXX including the opening replica flag according to the present invention is recorded by the root node RN and the cache node CN, respectively, and the entire content corresponding to each content and the opening replica are stored in separate content holders CH. By recording the content in the requester, it is possible to more quickly start the playback of the content desired to be played back by the requester (for the time being, before starting to acquire the entire content corresponding to the content) and to improve the convenience of the distribution system as a whole. Can also be improved.

(II) First Embodiment Next, a first embodiment according to the present invention in accordance with the above-described principle will be specifically described with reference to FIGS. Note that, in the distribution system according to the first embodiment described below, for the sake of simplification of description, it is assumed that the cache node CN does not exist in the distribution system.

  First, a schematic configuration of the node according to the first embodiment will be described with reference to FIG. FIG. 4 is a block diagram showing a schematic configuration of the node according to the first embodiment. In the first embodiment, the content holder CH, the requester RQ, the root node RN, and other nodes basically have the same hardware configuration, so that they represent general nodes on behalf of them. The outline of the configuration of N will be described with reference to FIG.

  As shown in FIG. 4, a node N included in the distribution system according to the first embodiment includes a CPU having a calculation function, a working RAM (Random Access Memory), and a ROM (Read Only) for recording various data and programs. Memory) and the like, and the entire content or the content data as the initial replica itself, the index information EXX, the DHT, and other necessary programs, etc. The configured recording unit 12 (the content data is recorded only in the recording unit 12 in the node N as the content holder CH before distribution), and the buffer memory 13 that temporarily stores the received content data Encoded video data (video information) and audio included in the content data. A decoder unit 14 that decodes (decompresses and decodes) data such as audio data (audio information), a video processing unit 15 that performs a predetermined drawing process on the decoded video data and outputs the video signal; A display unit 16 such as a CRT (Cathode Ray Tube) or a liquid crystal display that displays video based on the video signal output from the video processing unit 15, and the decoded audio data as an analog audio signal is converted into a D / A (Digital / (Analog) After the conversion, the sound processing unit 17 that amplifies and outputs the signal by an amplifier, the speaker 18 that outputs the audio signal output from the sound processing unit 17 as a sound wave, and the other node device 1 through the network 8 And a communication unit 20 for controlling communication of information between them and receiving instructions from respective users And an input unit (for example, a keyboard, a mouse, or an operation panel) 21 that outputs a corresponding instruction signal to the control unit 11, and includes a control unit 11, a recording unit 12, a buffer memory 13, a decoder unit 14, The communication unit 20 is connected via a bus 22 so as to be able to exchange data with each other.

  Then, when the CPU in the control unit 11 executes various programs recorded in the recording unit 12 or the like, the control unit 11 performs a requester RQ, a cache node CN, a root node RN, a content holder CH, or other general nodes. The overall operation as any one of N is comprehensively controlled.

  When the node N is a content holder CH as a partial storage device according to the present invention, the storage unit 12 functions as a partial distribution information storage unit. In addition, when the node N is a content holder CH as an overall storage device according to the present invention, the storage unit 12 functions as an overall distribution information storage unit. Further, when the node N is a root node RN as a control device according to the present invention, the control unit 11 is a control means, a management means, an acquisition means, a partial distribution information storage control means, a node search means, a copy control means, Functions as inquiry means, identification means, and erasure control means. Furthermore, when the node N is a requester RQ as a distribution destination apparatus according to the present invention, the communication unit 20 functions as a transmission unit, a re-transmission unit, a partial distribution reception unit, and an entire distribution reception unit, and the decoder unit 14 Functions as a reproducing means. Further, when the node N is a content holder CH as a partial storage device according to the present invention, the recording unit 12 functions as a storage unit, and the communication unit 20 functions as a reception unit and a response result transmission unit.

  Next, the operation of each node N according to the first embodiment will be described separately for the root node RN and the requester RQ.

(A) Operation in Root Node RN First, the operation in the root node RN according to the first embodiment will be described with reference to FIGS. FIG. 5 is a flowchart showing a process performed by the control unit 11 in the route node RN when the route node RN according to the first embodiment receives the delivery request message. FIGS. It is a flowchart which shows each detailed operation | movement in the control part 11 of root node RN which concerns on embodiment.

  First, in the process corresponding to the flowchart shown in FIG. 5, the route node RN according to the first embodiment sends a distribution source inquiry message (hereinafter, the distribution source inquiry message is simply referred to as “query”) from the requester RQ described later. This is started when it is confirmed that the root node RN is the root node for the inquired content ID.

  Here, the information included in the query specifically includes an IP address indicating the requester RQ itself that transmits the query, a content ID indicating content desired to be reproduced in the requester RQ, and the like. Yes.

  When the processing shown in FIG. 5 is started, first, the index information EXX is referred to, whether or not the requested content ID is included in the index information EXX, that is, the root node RN is the request target. It is confirmed whether or not a certain content is recognized (step S1). When the content is recognized (step S1; YES), the process proceeds to the following step S2. On the other hand, when the content is not recognized (step S1; NO), it is the root node RN. However, since it is impossible to store information related to content that does not exist in the distribution system, a query response message indicating that is returned to the requester (step S12), and the route node RQ according to the first embodiment is Complete the process.

  Next, in cooperation with the content holder CH that stores the entire content corresponding to the distribution request indicated by the query, the number of distribution requests for the stored entire content (request busyness) Is confirmed (steps S2, S3). The specific request busyness determination process executed in step S2 will be described in detail later.

  When it is determined that the distribution request for the content is busy (step S3; YES), the received query is a second or subsequent query transmitted from the same requester RQ (that is, the entire content). It is determined whether or not it is a query requesting reproduction, and the second and subsequent queries for the same content from the same requester RQ are referred to as “re-queries” (step S4).

  Here, as a specific determination method for determining whether or not the received query is a re-query, for example, in the root node RN, the same query is transmitted twice from the same node within a predetermined time. (For example, the specific value of the predetermined time is a time obtained by adding a preset time α to the reproduction time of the initial replica itself), or the query In the message, a flag for identifying whether the query is the first query or the second query may be provided.

  If the query is not a re-query (step S4; NO), then the first replica corresponding to the content requested to be distributed is the other one in the distribution system in which the root node RN participates. It is determined by referring to the initial replica flag included in the index information EXX stored in the root node RN whether or not it is stored in the node N (step S5).

  If it is determined in step S5 that the corresponding initial replica is stored in any other node N (step S5; YES), the initial address including the IP address of the node N storing the initial replica is included. The replica storage node information is returned as a query response message to the requester RQ (requester RQ that transmitted the query received in step S1) (step S7), and the processing as the root node RQ according to the first embodiment is completed. To do.

  Here, as the contents of the query response message, specifically, a flag indicating whether or not the content requested to be distributed has been found in the distribution system, and a content holder CH that stores the content requested to be distributed. And a flag for identifying whether the content holder CH holds the entire content or only the initial replica is described in the query response message. Further, since each content holder CH of the content is busy, and there is no content holder CH that can immediately distribute the content (specifically, described later), it is requested that the content holder CH wait until the distribution becomes possible. (If any content holder CH is busy, the requester RQ cannot immediately start playing the content, so the requester RQ has to wait, so this requester RQ has to wait.) The query response message in this case is also referred to as “standby message” in the following.) In addition to this, the query response message may include additional information such as the node ID and AS (Autonomous System) number of the content holder CH, distribution capability information, and the creation date and time of the stored replica.

  On the other hand, when it is determined in step S5 that the corresponding initial replica is not stored in any other node N (step S5; NO), a new initial in any other node N is determined. After the replica is accumulated (step S6), the initial replica accumulation node information including the IP address of the node N newly accumulating the initial replica is returned to the requester RQ as a query response message (step S7). The process as the root node RN according to the embodiment is completed. The specific processing for creating the initial replica executed in step S6 will be described in detail later.

  On the other hand, if it is determined in step S4 that the received query is a re-query (step S4; YES), the content that has been requested to be distributed as a whole but is stored in the entire content. Since distribution from the holder CH is impossible (see “Step S3; YES” above), a query response message (waiting until the entire content can be distributed at present) (in the query response message) Is returned to the requester RQ (step S8), and the processing as the root node RN according to the first embodiment is completed.

  Next, in the determination in step S3, the distribution request for the content requested to be distributed is not currently busy (that is, distribution is performed in one or more of the content holders CH that store the entire content corresponding to the content). When it is determined that the request is not busy (step S3; NO), it is no longer meaningful to store the initial replica in the distribution system, so the initial replica exists in the distribution system. It is determined with reference to the initial replica flag included in the index information EXX (step S9). When the initial replica exists (step S9; YES), a deletion process is performed to delete one of the existing initial replicas from the distribution system (step S10). Thereafter, the entire content storage node information including the IP address of the node N that stores the entire content and the distribution request is not busy is returned to the requester RQ as a query response message (step S11). The process as the root node RQ is completed. The specific processing for deleting the initial replica executed in step S10 will be described in detail later.

  On the other hand, if it is determined in step S9 that the initial replica does not exist in the distribution system (step S9; NO), the process proceeds to step S11 as it is and the process is executed.

  Next, the requested busyness determination process according to the first embodiment, which is executed in step S2, will be specifically described with reference to FIG.

  In the request busyness determination process, the busyness of the distribution request for the entire content is obtained by the cooperation of the root node RN and the content holder CH that stores the entire content corresponding to the content requested to be distributed. Is determined.

  That is, as shown in FIG. 5B, in the process of step S2, a determination request message for requesting determination of the busyness is first transmitted from the root node RN to the content holder CH storing the entire content ( Step S21). At this time, the location (IP address) of the content holder CH is acquired with reference to the index information EXX stored in the root node RN.

  Then, the control unit 11 of the content holder CH that has received the determination request message confirms the current status of the distribution request for the content (step S23), and whether or not the status is busy according to preset criteria. Is determined (step S24).

  As a result, when it is determined that the distribution request status for the content is not busy (step S24; NO), a result message indicating that is generated and transmitted (returned) to the root node RN (step S26).

  On the other hand, if it is determined in step S24 that the distribution request status for the content is busy (step S24; YES), a result message indicating that is generated and transmitted (returned) to the root node RN. (Step S25).

  Then, the root node RN that has received the result message of any content (step S27) moves to step S3 shown in FIG. 5A as the result of busyness determination of the distribution request in the content holder CH with the content. .

  In addition, specifically, as the reference set in advance in step S24, for example, the determination that the number of distribution requests for the content is set in advance within a preset determination reference time (for example, 300 seconds). When the number of times reaches a reference number (for example, 5 times) or more, it may be determined that the distribution request for the content is busy.

  Here, specifically with respect to the determination reference time, the determination reference time may be matched with the average transmission time of the content to be distributed. That is, in the case of a so-called streaming playback method, the determination reference time matches the content playback time.

  On the other hand, the number of times of the determination reference is specifically set in consideration of how many contents can be simultaneously distributed in correspondence with the bandwidth of the communication line.

  For example, if a 1 megabps (bit per second) line is used, 128 kbps music content can be transmitted in a calculation of 7.8 lines, but the number is set to about 5 in consideration of overhead such as a message header. And if the average reproduction time of those music contents is 5 minutes, it will determine with the delivery request | requirement 5 times or more within 300 seconds being busy.

  For example, if video content of 2 hours is distributed at 5 Mbps on a 100 Mbps line, assuming that the effective speed on the 100 Mbps line is 20 to 50 Mbps, four lines can be distributed simultaneously. In this case, it is considered busy with four or more distribution requests within 7,200 seconds.

  In addition to this, it is also possible to provide a reference value for the sum of the transmission speeds of the contents currently being distributed, instead of the number of times within the reference time, and to consider it busy if the reference value is exceeded. Specifically, for a 100 Mbps line (effective speed 20 to 50 Mbps), 20 Mbps is set as a reference value. And if you are delivering 2 content with a transfer rate of 5 Mbps, 2 content with a transfer rate of 4 Mbps, and 1 content with a transfer rate of 1 Mbps, it will not be determined to be busy. If there is one more distribution request for content with a transfer rate of 3 megabps, it is determined as “busy”.

  Next, the initial replica creation process according to the first embodiment executed in step S6 will be specifically described with reference to FIG.

  In the initial replica creation process, the initial replica is stored in one of the nodes N by the cooperation of the root node RN and the node N in the distribution system in which neither the entire content nor the initial replica is currently stored. Is done. Note that such a node N refers to the index information EXX stored in the root node RN, and the node N having an IP address not described in the index information EXX performs the initial replica creation process as step S6. It is targeted.

  That is, as shown in FIG. 6, in the process of step S6, first, in the root node RN, an initial setting process for determining the information amount of the initial replica itself to be newly created is executed (step S51). A beginning storage request message is transmitted to each target node N in the distribution system, inquiring whether or not a beginning replica that is a target of the initial setting process can be newly stored (step S52).

  Here, specifically regarding the selection of the transmission destination of the beginning accumulation request message, for example, it is preferable to transfer to the random node ID, and even if the node N having the node ID does not exist good. Then, when the initial storage request message is transferred to the node N having the node ID having the closest value to the node ID that is the first destination, the transfer is stopped there.

  Specifically, as the contents of the opening accumulation request message, the opening accumulation request message includes a node ID of a corresponding destination, an IP address of a node N that is a transmission source of the opening accumulation request message, The size of the entire content (that is, the total information amount as the entire content) and the size of the opening replica (that is, the total information amount as the opening replica) are described. The size of the initial replica is described in the initial storage request message if, for example, the distribution system as a whole specifies that the size of the initial replica is always 1/20 of the size of the entire content. It does not have to be done.

  On the other hand, the control unit 11 of the node N that has received the beginning storage request message determines whether or not a beginning replica can be newly stored in the node N (steps S61 and S62).

  Here, specifically, for example, for the purpose of effectively utilizing the storage unit 12 in each node N as the entire distribution system, the amount of information of the entire content in the storage unit 12 of the determination target node N is specifically described. If there is the above free storage area, it is determined that accumulation is impossible.

  If it is determined in step S62 that the initial replica cannot be stored in the node N (step S62; NO), an initial storage request response message indicating that is transmitted to the root node RN (step S68). ). On the other hand, when it is determined that the beginning replica can be accumulated in the node N (step S62; YES), an opening accumulation request response message indicating that is transmitted to the root node RN (step S63). A message including copy source information corresponding to the head accumulation request response message is received from the root node RN (step S64). At this time, in the message including the copy source information, when a new replica is created, the IP of the content holder CH that stores the entire content to be copied from which the content data corresponding to the start portion is copied is stored. An address etc. are included.

  After receiving the message including the copy source information, the content data corresponding to the initial replica is copied from the copy source (steps S65, S66; NO), and if the copy is completed (step S66; YES), a completion message indicating the completion is transmitted to the root node RN (step S67).

  On the other hand, after transmitting the opening accumulation request message in step S52, the root node RN waits for the opening accumulation request message from the node N (see step S68 or S63), and stores the opening replica. It is determined whether or not a possible node N has been found (steps S53, S53; NO).

  When a node N capable of storing the initial replica is found (step S53; YES), a message including the copy source information is transmitted to the node N (step S54), and then the completion of copying is indicated. Wait until a completion message is received (step S56; see step S67 above).

  Here, as a method for selecting the node N to be included in the copy source information, the root node RN recognizes the IP address of the node storing the entire content as the index information EXX. It is preferable that the node N as the transmission source is selected at random and the IP address is notified to the node N that requests the initial replica accumulation.

  If the completion message has been transmitted (step S56; YES), the IP address of the node N that has transmitted the completion message is indexed while the initial replica flag (see FIG. 3) is set to “1”. It adds as information EXX (step S57), and transfers to step S7 shown to Fig.5 (a).

  Next, the initial replica deletion process according to the first embodiment executed in step S10 will be specifically described with reference to FIG.

  In the initial replica deletion process, the initial replica is deleted from any one of the nodes N by the cooperation of the root node RN and any one of the nodes N currently storing the initial replica. Note that the node N that is the target of deletion of the initial replica is targeted at the node N whose initial replica flag is “1” in the index information EXX stored in the root node RN.

  That is, as shown in FIG. 7, in the process of step S81, first, in the root node RN, the node N (first replica content holder CH) storing the first replica to be deleted is specified.

  At this time, specifically, as the specifying method, it is preferable to select and specify at random from the nodes N whose initial replica flag is “1”. In addition to this, for example, the node N that has accumulated the initial replica for the longest time may be preferentially selected. In this case, it is necessary to add information indicating the time when the initial replica is accumulated in the index information EXX.

  Then, a deletion message for deleting the initial replica stored in the node N is transmitted to the identified node N (step S82). Thereafter, the root node RN waits for a deletion completion message to be described later (step S84).

  On the other hand, when the control unit 11 of any node N storing the initial replica to be deleted receives the deletion message, the control unit 11 stores the initial replica stored in the node N from the storage unit 12. Delete (step S87). If the deletion is completed, a completion message indicating the completion is transmitted to the root node RN (step S88).

  On the other hand, when the root node RN receives the completion message from the node N (step S84; YES), the root node RN determines the IP address of the node N that has transmitted the completion message from the stored index information EXX. After deleting (step S85), the process proceeds to step S11 shown in FIG.

(B) Operation in Requester RQ Next, the operation of the requester RQ that acquires the content by transmitting the query or the requery to the root node RN according to the first embodiment described with reference to FIGS. Will be described with reference to FIG. FIG. 8 is a flowchart showing content request processing in the requester RQ according to the first embodiment.

  In the requester RQ according to the first embodiment, when a content desired to be reproduced is selected by a user's operation, the control unit 11 starts processing corresponding to the flowchart of FIG.

  That is, as shown in FIG. 8, first, the query including the content ID indicating the selected content is transmitted to the distribution system (step S90), and a reply message corresponding to the query is received (step S91). (Refer FIG.5 (a) step S7 or S11).

  Next, it is confirmed whether or not the query response message includes information indicating that “the corresponding content cannot be found” (step S91-1). If the information is not included (step S91-1; no corresponding content is found), a message to that effect is displayed on the display unit 16 in the requester RQ (step S102).

  On the other hand, if it is determined in step S91-1 that the information is included (step S91-1; corresponding content discovery), then node information indicating the node N that stores the initial replica in the query response message. Or the like is included (step S92). At this time, when node information indicating the node N that stores the initial replica is not included (step S92; NO), the content holder CH that stores the entire content corresponding to the content that is desired to be reproduced Since the distribution request status in is not busy (see step S3 etc. in FIG. 5 (a)), a message to request transmission of the entire content is transmitted to the content holder CH that stores the entire content, The corresponding content data corresponding to the entire content is acquired and reproduced (step S100), and the processing as the requester RQ according to the first embodiment is completed.

  Specifically, as the processing in step S100, a message requesting acquisition of the entire content (including information indicating that transmission from the beginning of the entire content is desired) is transmitted to the content holder CH, and the message is handled. Then, the entire content transmitted is reproduced from the beginning.

  On the other hand, when the determination in step S92 includes node information indicating the node N storing the initial replica (step S92; YES), the initial replica is acquired based on the node information, and the initial replica is obtained. The reproduction is started from the top of the replica (step S93).

  Next, during the playback of the first replica, the playback of the first replica is performed only for the time when the acquisition of the corresponding entire content and preparation for playback are completed in consideration of the time expected to be acquired again. It waits until the time traced back from the end time (step S94). In the standby in step S94, the fact that the node information indicating the node N storing the initial replica is included in the query response message means that the content holder CH that stores the entire content corresponding to the content to be reproduced is used. This means that the distribution request status is busy, and it is expected that the busy distribution request status will be improved by waiting as much as possible in step S94.

  When the standby time in step S94 elapses, the re-query for acquiring the entire content corresponding to the first replica being reproduced is immediately transmitted to the distribution system (step S95), and the response corresponding to the re-query is sent. Receive the message.

  Then, it is determined whether or not the received query response message is a standby message for which a standby flag is set (see step S8 in FIG. 5A) (step S96). (Step S96; YES), the system waits for a preset standby time (Step S99), returns to Step S95 again, and repeats the above-described processing.

  On the other hand, if it is determined in step S96 that the received message is not a standby message (step S96; NO), a content holder CH (FIG. 5A) that accumulates the entire content included in the received query response message. Data from the portion corresponding to the beginning replica of the entire content is acquired from step S11) (step S97).

  Then, after preparation for reproduction, reproduction of the entire content acquired in step S97 is executed from the portion of the content following the initial replica being reproduced in step S93 (step S98), and the content holder CH according to the first embodiment is reproduced. To complete the process.

  Here, as the processing of steps S97 and S98, processing for starting reproduction of the entire content from the content portion following the initial replica will be specifically described.

  In this process, the control unit 11 in the requester RQ partially transfers the acquired (encoded state) data to the decoder unit 14 at a transfer speed corresponding to the playback speed of the content itself. At this time, in the case where the content is reproduced from the portion following the opening replica, after the data corresponding to the last portion of the opening replica is transferred to the decoder unit 14, the portion corresponding to the subsequent data in the acquired entire content Thereafter, the data is transferred to the decoder unit 14.

  At this time, the entire content is acquired from the head portion from the content holder CH storing the entire content, and the requester RQ is configured to discard the portion corresponding to the initial replica in the acquired entire content. However, this configuration is inefficient unless the line speed is significantly higher. Therefore, in the process of step S97 in the first embodiment, whether to acquire the entire content from the beginning or whether to acquire only from the portion following the opening replica is necessary. It is desirable that the request message for content acquisition is described together with information indicating a specific desired acquisition range and then transmitted to the corresponding content holder CH.

  As described above, according to the operations of the root node RN and the requester RQ according to the first embodiment, the node N that stores the initial replica (content holder CH of the initial replica), and the content holder CH that stores the entire content Are included in the distribution system as separate nodes N, the distribution of content from the content holder CH to the requester RQ is delayed due to many distribution requests for the content holder CH storing the entire content. Even in such a case, by using the opening replica for distribution before the entire content, reproduction of the content using the opening portion can be quickly started in the requester RQ.

  Accordingly, in the distribution system, even when the distribution request for the content holder CH that stores the entire content to be originally distributed is busy, the reproduction of the content in the requester RQ can be started. It can contribute to improvement of reliability and convenience for the whole.

  Further, since the content distribution source is determined based on comparison with a preset busyness criterion, the distribution source can be separated with a simple configuration.

  Further, since the query reply message indicating the distribution source is transmitted from the root node RN to the requester RQ when the distribution source is determined, the distribution source can be easily recognized by the requestor RQ.

  Furthermore, since the number of distribution requests for the content holder CH that stores the entire content is confirmed, and control is performed so that the initial replica is newly stored when it exceeds the reference value, the distribution request for the entire content is busy. It is possible to respond quickly to a distribution request from the requester RQ while effectively utilizing the free storage area in the other nodes N according to the degree.

  In addition, when a new initial replica is stored, a replica corresponding to the initial replica in the entire content is copied and stored instead of newly generating the entire content itself. New initial replicas can be stored in

  Further, node information indicating the content holder CH that stores the entire content is transmitted to the node N specified by the inquiry, and the initial replica is copied and stored, so that the required initial replica can be stored with a simple configuration. Can do.

  Furthermore, when the distribution request status in the content holder CH that accumulates the entire content is no longer busy, control is performed so that the unnecessary initial replica is deleted from the distribution system, so that the node N where the initial replica is deleted The storage unit 12 can be diverted to other uses, and the storage capacity can be effectively utilized as the entire distribution system.

  Also, as the requester RQ, when the received query response message includes information indicating the initial replica, the initial replica is received from the corresponding content holder CH, so that the entire content existing in the distribution system is accumulated. Even if the distribution request for the content holder CH to be performed is large, the reproduction of the content can be started quickly by placing it in the requester RQ.

  Accordingly, in the distribution system, even when the distribution request for the content holder CH that stores the entire content to be originally distributed is busy, the reproduction of the content in the requester RQ can be started. It can contribute to improvement of reliability and convenience for the whole.

  Further, as the requester RQ, a requery is transmitted after a preset time has elapsed from the start of reproduction of the initial replica until the end of the reproduction, and the entire content distributed in response to the transmitted requery is received. Since the reproduction is continued after the opening replica, even when there is a large distribution request for the content holder CH that accumulates the entire content, it is possible to reproduce the entire content while quickly starting the reproduction of one content.

  Furthermore, in the node N that has accumulated the initial replica, the storage unit 12 of the node N has a free area that is greater than the information amount of the initial replica and less than the information amount of the entire content, and the entire content is still If the initial replica is not stored in the node N, the initial storage request reply message indicating that the initial replica can be stored is transmitted to the root node RN, so that the free space in each node N in the distribution system is made effective. By utilizing and accumulating the initial replica, the free capacity of the entire distribution system can be effectively utilized.

(III) Second Embodiment Next, a second embodiment which is another embodiment according to the present invention based on the above-described principle will be specifically described with reference to FIGS. In the distribution system according to the second embodiment described below, for the sake of simplicity of explanation, it is assumed that the cache node CN does not exist in the distribution system, as in the case of the first embodiment. To do.

  Moreover, since the configuration of the nodes included in the distribution system according to the second embodiment is the same as that of the node N according to the first embodiment, the member numbers shown in FIG. Similarly, detailed description will be omitted.

  In the distribution system according to the first embodiment described above, a query is transmitted from the requester RQ, and at the root node RN, depending on the busyness of the distribution request status in the content holder CH that stores the entire content, the initial replica or the entire content However, in the second embodiment described below, both the entire content and the initial replica are already sent to different nodes N in the distribution system. When the query is transmitted in the requester RQ, information indicating whether the entire content or the initial replica is desired is included in the query and transmitted. Then, the root node RN that has received it returns information indicating the distribution source to the requester RQ in accordance with the content of the received query.

  Next, the operation of each node N according to the second embodiment will be described separately for the root node RN and the requester RQ.

(A) Operation in Root Node RN First, the operation in the root node RN according to the second embodiment will be described with reference to FIG. FIG. 9 is a flowchart showing a process for receiving a distribution request message in the root node RN according to the second embodiment, and the same process as in FIG. 5A used in the description according to the first embodiment is described. Similar step numbers are assigned and detailed description is omitted.

  In the processing corresponding to the flowchart shown in FIG. 9, the control unit 11 of the root node RN according to the second embodiment sends a distribution source inquiry message (hereinafter, the distribution source inquiry message is simply referred to as “query”) from the requester RQ. It starts when it is received.

  Here, as the information included in the query, specifically, in addition to the IP address indicating the requester RQ itself that transmits the query, the content ID indicating the content desired to be reproduced in the requester RQ, and the like, In the requester RQ, information indicating whether only the acquisition of the entire content is desired, only the acquisition of the opening replica is desired, or both the acquisition of the entire content and the acquisition of the opening replica is desired is included. It is.

  When the process shown in FIG. 9 is started, the index information EXX is referred to in the same manner as the process of step S1, and it is confirmed whether or not the requested content ID is included in the index information EXX (step S1). S1). When the content is recognized (step S1; YES), the process proceeds to the following step S30. On the other hand, when the content is not recognized (step S1; NO), it is the root node RN. However, since it is impossible to store information related to content that does not exist in the distribution system, a query response message indicating that is returned to the requester (step S34), and the route node RQ according to the second embodiment Complete the process.

  Next, the contents of the transmitted query are confirmed (step S30). As a result, if both the acquisition of the entire content and the acquisition of the initial replica are desired in the requester RQ that has transmitted it in the query (step S30; overall + initial), the root node RN at that time Referring to the index information EXX stored therein, both the node information indicating the content holder CH storing the entire content and the node information indicating the content holder CH storing the initial replica are both A query response message is generated and returned to the requester RQ (step S31), and the processing as the root node RQ according to the second embodiment is completed.

  On the other hand, if it is determined in step S30 that only the acquisition of the entire content is desired in the query in the requester RQ that has transmitted the query (step S30; overall), the information is stored in the root node RN at that time. Referring to the index information EXX, a response message including node information indicating the content holder CH that stores the entire content is generated and returned to the requester RQ (step S32), and the second embodiment The processing as the root node RQ related to is completed.

  Furthermore, if it is determined in step S30 that only the initial replica is desired in the query in the requester RQ that has transmitted it (step S30; initial), then it is stored in the root node RN at that time. Referring to the index information EXX, a response message including node information indicating the content holder CH storing the initial replica is generated and returned to the requester RQ (step S33), and the second embodiment The processing as the root node RQ related to is completed.

(B) Operation in Requester RQ Next, the operation of the requester RQ that attempts to acquire the content by transmitting the query to the root node RN according to the second embodiment described with reference to FIG. 12 will be described. FIG. 10 is a flowchart showing a first example of content request processing in the requester RQ according to the second embodiment, and FIGS. 11 and 12 are second examples of content request processing in the requester RQ according to the second embodiment. It is a flowchart to show.

  First, the first example will be described with reference to FIG. In the first example, the query that is first transmitted to the root node RN includes information indicating that both the acquisition of the entire content and the acquisition of the initial replica are desired in the requester RQ. This is an example.

  That is, in the first example, when the content desired to be reproduced is selected by the user's operation, the control unit 11 starts processing corresponding to the flowchart of FIG.

  As shown in FIG. 10A, first, the query including the content ID indicating the selected content is shown, the node information indicating the content holder CH that stores the entire content, and the content holder CH that stores the initial replica. Both of the node information are transmitted together with information indicating that the node information is desired (step S40), and a reply message corresponding to the query is received (step S41; see step S31 in FIG. 9).

  Then, it is determined whether or not node information indicating the node N that stores the initial replica is included in the response message (step S42). At this time, if node information indicating the node N that stores the initial replica is not included (step S42; NO), the content holder CH that stores the entire content corresponding to the content desired to be played back However, since it does not exist in the distribution system, the content data corresponding to the entire content is acquired from the content holder CH that stores the entire content, and is reproduced from the beginning (step S47). The processing as the requester RQ related to is completed.

  On the other hand, in the determination of step S42, as exemplified in FIG. 10B as a response message RTM, for example, the node information indicating the node N that stores the initial replica (that is, the node whose initial replica flag is “1”) Information) is included (step S42; YES), the initial replica is acquired based on the node information, and reproduction of the initial replica is started from the head (step S43).

  Next, during the reproduction of the opening replica, in consideration of the time estimated to be necessary for acquiring the whole content again with the same purpose (see step S94 in FIG. 8) as in the first embodiment, The system waits for a time traced back from the reproduction end time of the initial replica for the time when the acquisition and reproduction preparation of the corresponding entire content is completed (step S44).

  When the standby time in step S44 elapses, the entire content is acquired immediately from the content holder CH that accumulates the entire content included in the received response message (see step S41) (step S45). Then, after the reproduction preparation, the entire content is reproduced from the content portion following the initial replica being reproduced in step S43 (step S46), and the processing as the content holder CH according to the second embodiment is completed. Note that the processing for starting reproduction of the entire content from the content portion following the initial replica in step S46 is specifically the same as the processing described with reference to FIG. 8 as the processing in steps S97 and S98. Detailed description is omitted.

  Here, content desired to be reproduced is selected by the user's operation, and both node information indicating the content holder CH that stores the entire content and node information indicating the content holder CH that stores the initial replica are desired. When the selection is made, the control unit 11 may start processing corresponding to the flowchart of FIG. At this time, if it is predicted that the user wants to see the whole story as well as the beginning, it is preferable that the user selects both.

  Next, a second example of the overall operation in the requester RQ according to the second embodiment will be described with reference to FIGS. 11 and 12. In FIG. 11, the same processing steps as those in the first example shown in FIG. 10A are denoted by the same step numbers, and detailed description thereof is omitted.

  Further, the second example is an example in a case where information indicating that acquisition of the initial replica is desired in the requester RQ is included in the query transmitted to the root node RN first.

  That is, in the second example, when the content desired to be reproduced is selected by the user's operation, the control unit 11 starts processing corresponding to the flowchart of FIG.

  As shown in FIG. 11, first, the above query including the content ID indicating the selected content is transmitted into the distribution system together with information indicating that only node information indicating the content holder CH storing the initial replica is desired. (Step S50), a reply message corresponding to the query is received (Step S41; see Step S33 in FIG. 9).

  Then, it is determined whether or not the query response message includes node information indicating the node N that stores the initial replica (step S42). At this time, when the node information indicating the node N storing the initial replica is not included (step S42; NO), the content reproduction from the initial replica is no longer possible, and the process proceeds to step S51 described later. To do.

On the other hand, in the judgment of step S42, for example, as illustrated as a response message RT _P in FIG. 12 (a), when the node information indicating a node N for storing the beginning replica is present (step S42; YES), The initial replica is acquired based on the node information, and reproduction of the initial replica is started (step S43).

  Next, during playback of the initial replica, acquisition of the corresponding entire content and preparation for playback are completed in consideration of the time required to acquire the entire content again in the same manner as in the first example. It waits until the time traced back from the reproduction end time of the initial replica for the time to be performed (step S44).

  When the standby time in step S44 elapses, a new query is transmitted in the distribution system immediately with information indicating that only node information indicating the content holder CH that stores the entire content is desired (step S51). A response message corresponding to the query is received (see step S32 in FIG. 9). At this time, the reply message includes node information indicating the node N that accumulates the entire content (the initial replica flag is “0”) as exemplified by the reply message RTa in FIG. Next, it is determined whether or not the initial replica has been reproduced at that time (step S52). When the initial replica has already been reproduced (step S52; YES), the entire content is Content data corresponding to the entire content is acquired from the content holder CH to be stored (step S45), and the entire content is reproduced from the portion of the content following the initial replica being reproduced in step S43 (step S46). The process as the content holder CH according to the second embodiment is completed.

  On the other hand, if it is determined in step S52 that the reproduction of the beginning replica has not been started (step S52; NO), content data corresponding to the entire content is acquired from the content holder CH that stores the entire content (step S53). Then, the reproduction is executed from the beginning of the acquired entire content (step S54), and the processing as the content holder CH according to the second embodiment is completed.

  Here, when the content desired to be reproduced is selected by the user's operation and the node information indicating the content holder CH storing the initial replica is further selected, the control unit 11 in FIG. Processing corresponding to the flowchart may be started. At this time, if there is a request that the user wants to start watching as soon as possible because only the beginning is sufficient, the user may select only the beginning replica.

  As described above, according to the operation of the root node RN and the requester RQ according to the second embodiment, when the node N (content holder CH) storing the initial replica is under the management of the root node RN, the content Since the initial replica is distributed from the holder CH, it is possible for the content holder CH to complete the distribution in a short time and wait for a query from another requester RQ as compared with the case where the entire content is distributed. Compared to the case where there is only a content holder CH that stores the entire content under the management of the node RN, the initial content is distributed to the requester RQ in response to a distribution request from a plurality of requesters RQ, and the content is reproduced at the requester RQ. Start early.

  Further, since the reply message indicating the distribution source is transmitted from the root node RN to the requester RQ when the distribution source under management is determined, the distribution source can be easily recognized in each requester RQ.

  As for the new accumulation of the initial replica in the distribution system according to the second embodiment, for example, when the number of distribution requests for the entire content is increasing, the initial content may be created by copying to an arbitrary node N. preferable.

  In addition, regarding the deletion of the initial replica from the distribution system, for example, when there is no distribution request for a certain period, it is preferable to delete the initial replica from the distribution system.

(IV) Modification Next, a modification according to the present invention will be described with reference to FIG.

  In the first embodiment and the second embodiment described above, the embodiment has been described on the assumption that the cache node CN is not included in the distribution system, but the present invention uses FIG. Of course, the present invention can be applied to a distribution system including a cache node CN for explaining the operation.

  FIG. 13 is a flowchart showing a process for receiving a distribution request message in the cache node CN according to the modified embodiment, and the same step number is used for the same process as in FIG. 5A according to the first embodiment. A detailed description will be omitted.

  In the processing corresponding to the flowchart shown in FIG. 13, the cache node CN receives a distribution source inquiry message from the requester RQ (hereinafter, the distribution source inquiry message is simply referred to as “query”), and the cache node CN has its own index cache. Is started when an index matching the inquired content ID is found.

  That is, in the process of step S70, in order to confirm whether or not the initial replica not described in the index information EXX (see FIG. 3) in the cache node CN is accumulated in the distribution system, the root node RN is checked. The initial replica information request message is transmitted, and all index information of the initial replica stored in the root node RN is acquired as a response.

  Then, as the process of step S70-1, with reference to the index information acquired in the process of step S70, when new initial replicas are registered in the root node RN, the position information of all the initial replicas as long as it can be stored. Cache (temporarily store) the IP address of the content holder CH that stores the initial replica.

  This cache information optimization process uses a known logic such as LRU (Least Recently Used) that deletes information in the order of the longest unused time in order to reflect the dynamic change of the distribution system in the cache information. In addition to deleting any cache information with a long unused time by this normal optimization process when the generation of the initial replica is permitted in the distribution system, which is normally provided in the node CN The process of preferentially recording the position information of the content holder CH that stores the beginning replica is immediately executed.

  Thereafter, the processing of steps S2 to S5 (or S8) shown in FIG. 5A is executed, and it is confirmed in the determination of step S5 that the corresponding initial replica is not accumulated in any other node N. When it is done (step S5; NO), a message recommending that a new initial replica be stored in any of the other nodes N is transmitted to the root node RN at that time (step S71), and the new The initial replica storage node information including the IP address of the node N storing the initial replica is returned as a query response message to the requester RQ (step S7), and the processing as the cache node CN according to the modified form is completed.

  On the other hand, when it is determined in step S3 that the distribution request for the requested content is not busy at present (step S3; NO), the process of step S9 shown in FIG. 5A is executed. If the initial replica exists in the determination in step S9 (step S9; YES), a message recommending that one of the existing initial replicas be deleted from the distribution system is the route at that time. After transmitting to the node RN (step S72), the entire content storage node information including the IP address of the node N that stores the entire content and the distribution request is not busy is returned as a response message to the requester RQ. (Step S11), the processing as the cache node CN according to the modified form is completed. .

  Note that the root node RN that has received the message according to step S71 or S72 described above corresponds to the process in step S6 shown in FIG. 5A (see FIG. 6) or FIG. The process of step S10 shown (refer FIG. 7) will be performed.

  Even with the distribution system including the cache node CN according to the modified embodiment described above, the same operational effects as those of the first embodiment can be obtained.

  In addition, the program corresponding to the flowcharts shown in FIGS. 5 to 11 and FIG. 13 is recorded on an information recording medium such as a flexible disk or a hard disk, or acquired and recorded via the Internet or the like. By reading and executing these with a general-purpose computer, it is also possible to cause the computer to function as the control unit 11 according to the embodiment.

  Furthermore, in the above-described embodiment, etc., the content holder CH itself is inquired whether the content holder CH is busy, but in addition to this, the root node RN or the cache node CN depends on the number of queries per predetermined time. You may decide whether you are busy.

  Furthermore, in the above-described embodiment, the entire content is automatically acquired after receiving the opening replica. However, the user of the requester RQ views the entire content after viewing the opening replica. When a desired operation is performed, the entire content query may be transmitted.

  As described above, the present invention can be used in the field of content distribution via a network, and particularly when applied to the field of content distribution in a download format, a particularly remarkable effect can be obtained.

It is a schematic diagram (I) which shows the outline | summary of the delivery system which concerns on embodiment, (a) is a schematic diagram which shows ID space in the said delivery system, (b) is a figure which illustrates index information. It is a schematic diagram (II) which shows the outline | summary of the delivery system which concerns on embodiment, (a) is a schematic diagram which shows the transmission state of the message in the said delivery system, (b) is transmission of the said message represented as a spanning tree. It is a figure which shows a state. It is a figure which shows the principle of this invention. It is a block diagram which shows the outline structure of the node which concerns on embodiment common about a content holder etc. It is a flowchart which shows the process with respect to reception of the delivery request message in the root node which concerns on 1st Embodiment, (a) is a flowchart which shows the said whole operation | movement, (b) is the request busyness determination process which concerns on 1st Embodiment. It is a flowchart which shows. It is a flowchart which shows the beginning replica production | generation operation | movement in the root node which concerns on 1st Embodiment. It is a flowchart which shows the beginning replica deletion operation | movement in the root node which concerns on 1st Embodiment. It is a flowchart which shows the content request process in the requester which concerns on 1st Embodiment. It is a flowchart which shows the process with respect to reception of the delivery request message in the root node which concerns on 2nd Embodiment. 6 is a flowchart showing a first example of processing for receiving a distribution request message in the requester according to the second embodiment, (a) is a flowchart showing the first example, and (b) is a content of a response message to be received. FIG. It is a flowchart which shows the 2nd example of the process with respect to reception of the delivery request message in the requester concerning 2nd Embodiment. It is a figure which illustrates the content of the reply message received in the 2nd example in the requester which concerns on 2nd Embodiment, (a) is a figure which shows a 1st example, (b) is a figure which shows a 2nd example. It is. It is a flowchart which shows the process with respect to reception of the delivery request message in the cache node which concerns on a deformation | transformation form.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Control part 12 Recording part 13 Buffer memory 14 Decoder part 15 Image | video processing part 16 Display part 17 Audio | voice processing part 18 Speaker 20 Communication part 20
21 Input unit 22 Bus Rc, Rn Ring EX, EXX Index information CH Content holder RQ, RQ1, RQ2, RQ3 Requester RN Root node CN Cache node Qp Registration message Qr Distributor inquiry message N Node RT reply information RTM, RT _P , RT _a reply message

Claims (17)

Distribution in which a plurality of node devices are connected via a network, and distribution information stored in any one of the node devices is distributed to a distribution destination device that is any other node device. In the system,
An overall storage device which is the node device comprising distribution information storage means for storing the entire one piece of distribution information;
The node device that is different from the overall storage device and includes partial distribution information storage means for storing partial distribution information that is a part of the distribution information including at least the beginning portion of the one distribution information A partial storage device which is
A control device which is the node device comprising control means for controlling whether the distribution information is started to be distributed to the distribution destination device from the whole storage device or the partial storage device;
A distribution system comprising: The control device included in the distribution system according to claim 1,
A control apparatus comprising the control means. The control device according to claim 2,
The control means controls to distribute the partial distribution information from the partial storage device to the distribution destination device when the number of distribution requests for the distribution information is equal to or greater than a preset threshold value.
When the number of distribution requests is less than the threshold, the control apparatus controls to distribute the distribution information from the overall storage apparatus to the distribution destination apparatus. The control device according to claim 3,
When the number of distribution requests is equal to or greater than the threshold, the control means transmits response information indicating the partial storage device to the distribution destination device, and when the number of distribution requests is less than the threshold, the total storage device The control apparatus is characterized in that the response information indicating is transmitted to the distribution destination apparatus. The control device according to claim 2,
Management means for managing the entire storage device and the partial storage device,
The control unit controls the partial storage device to distribute the partial distribution information to a distribution destination device when the partial storage device is managed by the management unit, and the partial storage device manages the management unit. If not, the control device controls to distribute the distribution information from the overall storage device to the distribution destination device. The control device according to claim 5,
When the partial storage device is managed by the management unit, the control unit transmits response information indicating the partial storage device to the distribution destination device, and the partial storage device is not managed by the management unit And a response information indicating the entire storage device is transmitted to the distribution destination device. The control device according to claim 3 or 4,
Obtaining means for obtaining request information for requesting delivery of the delivery information from the delivery destination device;
Partial distribution information storage control means for newly storing the partial distribution information in another node device when the number of distribution requests based on the acquired request information is equal to or greater than the threshold;
A control device comprising: The control device according to claim 7,
The partial delivery information accumulation control means includes:
Node search means for searching the distribution system for the node device capable of storing the partial distribution information;
Copy control means for copying a part of the distribution information stored in the total storage device as the partial distribution information from the total storage device in the searched node device;
A control device comprising: The control device according to claim 8, wherein
The node search means includes:
Inquiry means for inquiring whether or not the partial delivery information can be stored in another node device;
Identifying means for identifying the node device capable of storing the partial delivery information based on a response result to the inquiry by the inquiry means;
With
The copy control means transmits overall device information indicating the overall storage device to the specified node device, and causes the specified node device to receive the part from the overall storage device as the partial distribution information. A control device characterized by. In the control device according to any one of claims 2 to 9,
Erasure control means for erasing the partial distribution information stored in the partial distribution information storage means from the partial distribution information storage means when the number of distribution requests for the distribution information becomes less than a preset threshold value. A control device further comprising: The distribution destination device included in the distribution system according to claim 1,
Sending means for sending delivery request information indicating the delivery information desired to be delivered to the delivery destination device;
When partial device information indicating the partial storage device is included in the response information transmitted in response to the distribution request information, the partial distribution information is sent from the partial storage device indicated by the partial device information. Partial distribution receiving means for receiving distribution;
Reproducing means for reproducing the distributed partial distribution information;
A delivery destination device comprising: The delivery destination device according to claim 11,
Re-transmission means for re-transmitting the distribution request information after a preset time has elapsed from the reproduction start time of the partial distribution information until before the reproduction of the partial distribution information is completed;
When the whole device information indicating the whole storage device is included in the response information transmitted in response to the delivery request information sent again, the distribution from the whole storage device indicated by the whole device information An overall distribution receiving means for receiving information distribution;
Further comprising
The distribution destination device, wherein after the reproduction of the partial distribution information is completed, the reproduction unit reproduces the distribution information continuously from the distributed distribution information portion following the partial distribution information. The distribution destination device included in the distribution system according to claim 1,
Sending means for sending delivery request information for requesting at least one of the whole delivery information corresponding to the delivery information desired to be delivered to the delivery destination device or at least one of the partial delivery information;
Distribution receiving means for receiving distribution of the distribution information from either the entire storage device or the partial storage device indicated by the response information transmitted in response to the distribution request information;
Reproducing means for reproducing the distributed distribution information;
A delivery destination device comprising: The node device included in the distribution system according to claim 1,
Storage means;
Receiving means for receiving inquiry information for inquiring whether or not the partial delivery information can be stored;
The partial distribution when the storage means has an empty area that is not less than the information amount of the partial distribution information and less than the total information amount of the distribution information, and the distribution information is not stored in the storage means. Response result sending means for sending a response result indicating that information can be stored into the distribution system;
A node device comprising:   A control program causing a computer included in any one of the node devices included in the distribution system according to claim 1 to function as the control device according to any one of claims 2 to 10.   14. A distribution destination device that causes a computer included in any one of the node devices included in the distribution system according to claim 1 to function as the distribution destination device according to any one of claims 11 to 13. Program.   15. A program for a node device, which causes a computer included in any one of the node devices included in the distribution system according to claim 1 to function as the node device according to claim 14.

Images