JP2012070327A - Distribution system, distribution server management apparatus, distribution server management program, distribution server, and distribution program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve efficiency of distribution by distribution servers.SOLUTION: A distribution system of this invention is provided with: plural distribution servers each having first storage means and second storage means and distributing a content stored in one of the storage means: and a distribution server management apparatus which selects a server for distributing the content. The distribution server management apparatus comprises server selection means which on the basis of information on the storage and distribution states of the distribution servers, selects a distribution server having a capacity larger than a storage amount required for storing the content included in content information and having the first storage means of a minimum number of distributions, as the distribution server for distributing the content.

Description

  The present invention relates to a distribution system, a distribution server management device, a distribution server management program, a distribution server, and a distribution program. For example, in a system that distributes moving images through an IP network, a program being broadcast is viewed from a portion that has already been broadcast. It can be applied to a start-over type storage and distribution system.

Conventional video distribution service systems have the following main problems.
1. The server must handle a very large amount of delivery per hour.
2. Servers must deliver to a wide range of delivery areas with as little delay as possible.
3. Servers must manage content that is huge in number and capacity.

In order to solve the above problems, the following measures have been conventionally taken.
1. Increase the number of servers. As a result, the distribution amount can be distributed to the servers, so even a huge amount of distribution can be dealt with.
2. For example, a distribution server group is created for each region. As a result, users in each region use one of the distribution servers in the group, and the distribution route becomes shorter.
3. The server that distributes content is divided into a center server that holds and manages all content and a local server that holds only high-demand content. As a result, the content can be centrally managed, and the cost required for the recording medium can be suppressed.

  Patent Document 1 discloses a tree-structured content delivery that uses a center server as a root and a local server as a node in order to efficiently distribute content from a center server that holds and manages all content to a local server. A content distribution system constituting a network is described.

  FIG. 2 is a configuration diagram showing a system configuration of the technique described in Patent Document 1. As shown in FIG. The system described in Patent Document 1 distributes content managed by the center server 7 through a path shown in FIG. In addition, the system described in Patent Document 1 does not have the local servers 8-1 and 8-2 and the local servers 9-1 to 9-4 hold the entire content, but the local server 8 according to the average viewing time and the like. -1 and 8-2, and the local servers 9-1 to 9-4 hold a part of the content, thereby increasing the usage efficiency of the recording medium.

JP 2010-034871 A

  Incidentally, there is a start-over type moving image distribution system as one of moving image distribution services. A start-over type moving image distribution system refers to a storage distribution system that stores programs that are being broadcast and enables viewing from a portion that has already been broadcast.

The start-over type video distribution system has the following features.
1. The system distributes broadcast programs that are currently being broadcast.
2. Viewing requests are accepted only during the program broadcast time, and viewing is limited to the viewing deadline.
3. The number of distributions at the start of the program is considerably higher than normal VOD (video on demand) content.
4). As shown in FIG. 3, the number of distributions has the largest increase at the start of the program, and then maintains that number. After the program ends, the number of distributions decreases rapidly.
5. Since there is a broadcast schedule, there is a program that follows the program after the current broadcast program ends. Therefore, the accumulation / delivery request is continuously generated according to the broadcast schedule.

The start-over type video distribution system has the following problems.
1. During program broadcasting, it is necessary to secure a considerably larger number of distributions than VOD content.
2. Accumulated distribution requests continue to occur, and after the program ends, the number of distributions of the program suddenly decreases to zero when the viewing period expires. Therefore, each program is most suitable at the start of broadcasting so that the number of distributions at each server can be maximized. Must be assigned to the server.

  Even if it is going to solve the above subjects with conventional technology, the following problems may arise.

1. The conventional video delivery system has either a content viewing deadline or a long viewing deadline, and the number of distributions does not increase in a certain period. For this reason, the conventional VOD system does not have a large distribution capacity that can cope with intensive distribution at one time.

2. The content is centrally managed, and is distributed from the center server to the local server according to the audience rating and the like for load distribution. Therefore, the conventional moving image distribution system distributes the load at a relatively moderate speed. However, since the start-over type broadcast program storage and distribution system requires real-time performance, the conventional load distribution method cannot cope with it.

  Therefore, since the number of programs to be distributed greatly varies depending on the distribution time, a distribution system, a distribution server management device, a distribution server management program, a distribution server, and a distribution program that can increase the distribution efficiency of the memory server are required. Yes.

  In order to solve such a problem, the distribution system of the first aspect of the present invention has (A) first storage means and second storage means for storing contents to be distributed to the client, and upon receiving a distribution request from the client. Selecting a plurality of distribution servers that distribute the content stored in one of the first storage unit and the second storage unit, and (B) a distribution server that distributes the content among the plurality of distribution servers, and selecting the selected distribution server A distribution server management device that notifies the address information of the distribution server to the client, and the distribution server management device (1) stores the storage state and the distribution state of the first storage unit and the second storage unit in each distribution server. Accumulated distribution status information acquisition means for periodically acquiring distribution status information from all distribution servers, and (2) content information related to content from outside Content information acquisition means to obtain, (3) information storage means for storing accumulated distribution status information and content information of a plurality of distribution servers, and (4) content information based on accumulated distribution status information of a plurality of distribution servers And (5) a server that selects a distribution server having a first storage unit that has a capacity that exceeds the required storage amount of the content and that has the smallest number of distributions as the content to be distributed. A storage distribution request unit that gives content information about the content to the distribution server selected by the selection unit and requests storage distribution of the content; and (6) a server that receives a notification of a content distribution destination from the client, Server information notification to notify the client of the address information of the distribution server selected by the selection means And having a stage.

  The distribution server management apparatus according to the second aspect of the present invention includes first storage means and second storage means for storing content to be distributed to the client. Upon receiving a distribution request from the client, the first storage means and the second storage In a distribution server management apparatus that selects a distribution server that distributes content from among a plurality of distribution servers that distribute content stored in any of the means, and notifies the client of address information of the selected distribution server. 1) stored distribution status information acquisition means for periodically acquiring stored distribution status information indicating the storage status and distribution status of the first storage means and the second storage means in each distribution server; and (2) Content information acquisition means for acquiring content information related to the content from the outside, and (3) accumulated distribution status information of a plurality of distribution servers and Information storage means for storing content information, and (4) a capacity that exceeds the storage required amount of the content included in the content information based on the storage distribution status information of a plurality of distribution servers, and the number of distributions A server selection unit that selects the distribution server having the minimum first storage unit as the one to distribute the content; and (5) content information related to the content is given to the distribution server selected by the server selection unit, (6) server information for notifying the client of the address information of the distribution server selected by the server selection means upon receiving a content distribution destination notification request from the client; And a notification means.

  The distribution server management program according to the third aspect of the present invention includes first storage means and second storage means for storing contents to be distributed to the client. When receiving a distribution request from the client, the first storage means and the second storage In a distribution server management program for selecting a distribution server that distributes content from among a plurality of distribution servers that distribute content stored in any of the means, and notifying a client of address information of the selected distribution server, a computer (1) Accumulated distribution status information acquisition means for periodically acquiring accumulated distribution status information indicating the accumulation status and distribution status of the first accumulation means and the second accumulation means in each distribution server from all distribution servers; 2) Content information acquisition means for acquiring content information related to content from the outside, (3) Multiple distribution services (4) Based on the storage / delivery state information of a plurality of distribution servers, the information storage means for storing the storage / distribution state information and content information of the server has a capacity that exceeds the required storage amount of the content included in the content information. A server selection unit that selects a distribution server having a first storage unit with a minimum number of distributions as a distribution server, and (5) content related to the content for the distribution server selected by the server selection unit. (6) When receiving a notification of content distribution destination from the client, the client is notified of the address information of the distribution server selected by the server selection unit. It is made to function as a server information notification means.

  The first storage means and the second storage means for storing the content to be distributed to the distribution server client of the fourth aspect of the present invention, and when receiving a distribution request from the client, one of the first storage means and the second storage means A distribution server that distributes the content stored in (1), a distribution switching control unit that controls distribution of the content stored in either the first storage unit or the second storage unit, and (2) In response to a request from the distribution server management device, the storage management state is monitored including the storage state and distribution state of the first storage unit and the second storage unit, and the stored distribution state information is returned to the distribution server management device. Means.

  A distribution program according to a fifth aspect of the present invention includes first storage means and second storage means for storing content to be distributed to a client. Upon receiving a distribution request from a client, the first storage means and the second storage means A distribution switching control unit that distributes content stored in any one of the following: (1) distribution switching control unit that controls distribution of content stored in either the first storage unit or the second storage unit (2) Upon receiving a request from the distribution server management device, the storage distribution status including the storage status and distribution status of the first storage means and the second storage means is monitored, and the stored distribution status information is sent to the distribution server management device. It is made to function as a state management means to respond.

  According to the present invention, since the number of distributions of content to be distributed greatly varies depending on the distribution time, the distribution efficiency of the memory server can be improved.

It is a whole lineblock diagram showing the whole distribution system composition of a 1st embodiment. It is a block diagram which shows the structural example of the conventional delivery system. It is explanatory drawing explaining the time transition of the number of delivery by start over. 1 is an internal configuration diagram illustrating an internal configuration of a memory server 30 according to a first embodiment. It is an internal block diagram which shows the internal structure of the memory server manager 10 of 1st Embodiment. It is a sequence diagram explaining the process of the broadcast program storage function in the delivery system of 1st Embodiment. It is explanatory drawing explaining the selection method of the memory server of 1st Embodiment. It is a sequence diagram explaining the process of the stored program delivery function in the delivery system of 1st Embodiment. It is explanatory drawing explaining the switching process of the delivery system in the memory server of 1st Embodiment. It is a flowchart explaining the selection method of the memory server of 2nd Embodiment. It is a block diagram which shows the structural example of the memory free space transition table of 2nd Embodiment. It is explanatory drawing explaining the relationship between memory free capacity transition of 2nd Embodiment, and the accumulation amount required for accumulation | storage of an accumulation | storage program. It is explanatory drawing explaining the integration process of the memory area | region of the memory of 2nd Embodiment. It is a sequence diagram explaining the process of the broadcast program storage function in the delivery system of 3rd Embodiment.

(A) First Embodiment Hereinafter, a first embodiment of a distribution system, a distribution server management device, a distribution server management program, a distribution server, and a distribution program according to the present invention will be described with reference to the drawings.

  The first embodiment exemplifies an embodiment in which the present invention is applied to, for example, a start-over type broadcast program storage and distribution system.

(A-1) Configuration of the First Embodiment (A-1-1) Overall Configuration FIG. 1 is an overall configuration diagram showing the overall configuration of the distribution system of the first embodiment. In FIG. 1, a distribution system 1 according to the first embodiment includes a memory server manager 10, an accumulated distribution management list 20, and a plurality of (N in FIG. 1, N is a positive integer) memory servers 30-1 to 30-. N, the client 40, and the dispatcher 50 are comprised at least.

  The memory server manager 10 manages the memory servers 30-1 to 30-N that distribute broadcast programs (contents).

  The memory server manager 10 receives a program storage / delivery request from a system administrator terminal (not shown), and registers the program information received from the system administrator terminal in the storage / distribution management list 20. After registering the program information, the memory server manager 10 returns a response to the program storage / delivery request from the system administrator terminal.

  Here, the program information is information related to the broadcast program, and includes, for example, a program number for identifying the program, a start time of the broadcast program, an end time of the broadcast program, a viewing time limit (viewing period) for viewing the broadcast program, and a program. There is an accumulation amount for accumulation.

  A system administrator terminal (not shown) makes a program storage and distribution request to the memory server manager 10 for a program to be broadcast in accordance with a preset program broadcast schedule.

  Further, the memory server manager 10 periodically makes a memory server status acquisition request to the memory servers 30-1 to 30-N, and acquires memory server status information from each of the memory servers 30-1 to 30-N. Is. When the memory server manager 10 receives the memory server status information from each of the memory servers 30-1 to 30-N, the memory server manager 10 stores the memory server status information of each of the memory servers 30-1 to 30-N in the accumulated distribution management list 20. Update.

  Here, the memory server state information is a memory server number for identifying each of the memory servers 30-1 to 30-N, an accumulation state indicating accumulation start or accumulation end, a distribution state indicating memory distribution or HDD (hard disk drive) distribution. , The number of memory distribution or HDD distribution, and the memory usage (for example, the planned memory capacity, the current storage capacity, etc.).

  Further, the memory server manager 10 refers to the memory server status information of all the memory servers 30-1 to 30-N in the storage / delivery management list 20 immediately before the start time of the program, thereby referring to the memory servers 30-1 to 30-. The memory server that broadcasts the program to the client 40 is selected from N.

  Further, when the memory server manager 10 receives a program distribution memory server information acquisition request from the dispatcher 50, the memory server manager 10 refers to the stored distribution management list 20 and specifies the memory server number of the memory server 30 selected to distribute the program. It responds to 50.

  The storage / delivery management list 20 manages information related to broadcast programs requested to be stored / distributed and memory server status information of the memory servers 30-1 to 30-N. The storage / delivery management list 20 manages the memory server status information of the memory servers 30-1 to 30-N that deliver the program in association with each other. For example, the storage / delivery management list 20 manages a program number, start number, end program, viewing time limit, memory server number, storage state, distribution state, number of distributions, memory capacity to be used, current storage capacity, and the like.

  The memory server 30 (30-1 to 30-N) includes a memory 31 and an HDD 32, stores broadcast programs received from a server (not shown) in both the memory 31 and the HDD 32, and stores them in the memory 31 and the HDD 32. Such a program is distributed to the client 40.

  The memory servers 30-1 to 30-N that have received a storage / delivery start request from the memory server manager 10 have a program related to the request from the start time of the program to the end time of the program and the viewing time limit of the program. Broadcast programs. Further, the memory servers 30-1 to 30-N accept a distribution request from the client 40 by the end time, and distribute the program to the client 40 that has requested distribution.

  The memory servers 30-1 to 30 -N set a threshold value α for the number of distributions of the memory 31. When the number of distributions on the memory 31 exceeds the threshold value α, the memory servers 30-1 to 30-N perform memory distribution for distributing to the clients 40 using data stored in the memory 31. When the number of distributions in the memory 31 falls below the threshold value α, the memory servers 30-1 to 30-N switch to the HDD distribution for distributing the stored data in the HDD 32 to the client 40, and store the program data in the memory 31. Stop and delete the data stored in the memory 31. Thereby, program data (for example, program data of the next different program) reserved for subsequent recording can be stored in the memory 31.

  The client 40 receives a program distributed from the memory servers 30-1 to 30-N. When receiving the program distribution, the client 40 makes a memory server allocation request for the distribution destination of the program to the dispatcher 50. As a result, when the client 40 receives the memory server number of the memory server 30 that is the distribution destination from the dispatcher 50, it makes a distribution request to the memory server 30 of that memory server number.

  When the dispatcher 50 receives a memory server allocation request from the client 40, the dispatcher 50 sends a program distribution memory server information acquisition request for the program to the memory server manager 10. Further, when the dispatcher 50 acquires the program distribution memory server information for distributing the program from the memory server manager 10, the dispatcher 50 gives the memory server information to the client 40.

(A-1-2) Internal Configuration of Memory Server 30 FIG. 4 is an internal configuration diagram showing the internal configuration of the memory server 30. In FIG. 4, the memory server 30 includes at least a memory 31, an HDD 32, a control unit 33, and a communication unit 34.

  The memory 31 is a storage unit that accumulates program data received from a server (not shown) via the communication unit 34. The memory 31 distributes the stored data to the client 40 via the communication unit 34 under the control of the control unit 33.

  The HDD 32 is a hard disk drive. The HDD 32 also stores program data received from a server (not shown) via the communication unit 34. Further, the HDD 32 distributes the stored data to the client 40 via the communication unit 34 under the control of the control unit 33.

  The control unit 33 manages processing in the memory server 30. The main functions of the control unit 33 include at least a program accumulation / delivery processing unit 331, a state management unit 332, and a distribution method switching control unit 333.

  The program storage / distribution processing unit 331 performs processing for distributing stored program data. When the program storage / delivery processing unit 331 receives a storage / delivery start request including the program number, start time, end time, and viewing time limit from the memory server manager 10, the program storage / delivery processing unit 331 sends the program data of the program to the memory 31 and the HDD 32. Make a storage allocation.

  The program storage / delivery processing unit 331 starts storing program data in the memory 31 and the HDD 32 at the start time of the program, and sends a storage / distribution start response to the memory server manager 10.

  Further, when receiving a distribution request including a program number, an expiration date, and a reproduction method from the client 40, the program accumulation / delivery processing unit 331 designates the number of the program number based on the data accumulated in the memory 31 and the HDD 32. It is processed into data that can be played back by the playback method and delivered to the client 40.

  Note that the program storage / delivery processing unit 331 accepts the delivery request from the client 40 until the program end time, and does not accept the delivery request after the end time has elapsed.

  Further, the program storage / delivery processing unit 331 deletes the data stored in the memory 31 and the HDD 32 when the viewing time limit is reached.

  The state management unit 332 monitors the state of the memory 31 and the HDD 32 of the memory server 30. In response to the memory server status acquisition request from the memory server manager 10, the status management unit 332 responds to the memory server manager 10 with memory server status information including the statuses of the memory 31 and the HDD 32.

  Specifically, the state management unit 332 stores the memory 31 and the HDD 32 used for program distribution (current storage or storage end, etc.), whether the program distribution method is memory distribution or HDD distribution, 31 and the number of distributions indicating the number of program streams being distributed by the HDD 32, the memory capacity to be used as the usage status of the memory 31, the current storage capacity of the memory 31 and the HDD 32, and the like are monitored.

  Then, the state management unit 332 monitors its own memory server number, the storage state of the memory 31 and the HDD 32, the distribution state, the number of distributions of the memory 31 and the HDD 32, the scheduled memory capacity, the current storage capacity, etc. Is provided to the memory server manager 10 as memory server status information.

  The distribution method switching control unit 333 controls switching between memory distribution and HDD distribution.

  The distribution method switching control unit 333 sets a threshold value α for the number of distributions in the memory 31. The distribution method switching control unit 333 compares the number of distributions of the memory 31 monitored by the state management unit 332 with the threshold value α. When the number of distributions exceeds the threshold value α, the distribution method switching control unit 333 performs memory distribution. When the number of distributions falls below the threshold value α, the distribution method switching control unit 333 ends the accumulation in the memory 31 and the memory 31. Delete the data stored in.

  At the start of the program, there are many distribution requests from the client 40, and the number of distributions may be large. In the first embodiment, memory distribution is performed at the start of a program so that it is possible to cope with a large number of distributions at the start of the program. In addition, since the memory 31 has a relatively small capacity and the next program storage and distribution is also refrained, when the number of distributions falls below the threshold value α (for example, when the number of distributions decreases at the end of the program), switching to HDD distribution is performed. The memory 31 contributes to storing subsequent programs.

  The communication unit 34 is a processing unit or device that communicates with a network. The communication unit 34 exchanges information with a server (not shown), the memory server manager 10, and the client 40 via a network.

(A-1-3) Internal Configuration of Memory Server Manager 10 FIG. 5 is an internal configuration diagram showing the internal configuration of the memory server manager 10. In FIG. 5, the memory server manager 10 includes at least a control unit 11 and a communication unit 12.

  The control unit 11 is responsible for processing in the memory server manager 10. The main functions of the control unit 11 include a program information registration unit 111, a memory server state acquisition unit 112, a memory server selection unit 113, an accumulation / distribution request unit 114, and a program distribution memory server information response unit 115.

  When receiving a program accumulation request from a system administrator terminal (not shown), the program information registration unit 111 registers program information in the accumulation / distribution management list 20. In addition, the program information registration unit 111 responds to the system management terminal with program registration success / failure information indicating whether or not the registration of the program information in the stored distribution management list 20 is successful.

  The memory server status acquisition unit 112 periodically requests acquisition of memory server status information to all the memory servers 30-1 to 30-N. When the memory server status acquisition unit 112 acquires the memory server status information from the memory servers 30-1 to 30 -N, the memory server status acquisition unit 112 registers the memory server status information in the accumulated distribution management list 20.

  The memory server selection unit 113 refers to the accumulated distribution management list 20 and selects the memory server 30 that performs program distribution.

  Here, the memory server selection method can be determined by the memory server selection unit 113 based on the memory server status information of the memory servers 30-1 to 30 -N registered in the storage / delivery management list 20. Various methods can be applied.

  For example, the memory server selection unit 113 creates a list of the memory servers 30 having a sufficient capacity for storing programs based on the usage status of the memory 31 (expected memory capacity to be used, current storage capacity). And the memory server selection part 113 can apply the method of selecting the memory server 30 with the smallest delivery number of the memory 31 based on the list | wrist. It should be noted that the capacity sufficient to store a program means that the free memory capacity exceeds the storage amount necessary for storing program data.

  The storage / delivery start request unit 114 makes a storage / distribution start request including the program number, the start time, the end time, and the viewing time limit of the program to the memory server 30 selected by the memory server selection unit 113.

  When the program distribution memory server information response unit 115 receives an acquisition request for program distribution memory server information from the dispatcher 50, the memory server number (for example, address) of the memory server 30 selected by the memory server selection unit 113 for the requested program. Information) to the dispatcher 50.

  The communication unit 12 is a processing unit or device that performs communication processing via a network. The communication unit 12 exchanges information with a system management terminal (not shown), the memory servers 30-1 to 30-N, and the dispatcher 50 via a network.

(A-2) Operation of the First Embodiment Next, the operation of processing in the start-over type distribution system 1 of the first embodiment will be described with reference to the drawings.

  The distribution system 1 of the first embodiment has a broadcast program accumulation function and an accumulated program distribution function.

  The broadcast program storage function is a function requested by the system administrator terminal to the memory server manager 10. The memory server manager 10 starts accumulating programs specified by the system administrator terminal.

  In the stored program distribution function, the memory server manager 10 allocates a memory server 30 that distributes a program through the dispatcher 50 that has received a request from the client 40. The allocated memory server 30 has a function of distributing the stored program to the client 40.

(A-2-1) Broadcast Program Storage Function FIG. 6 is a sequence diagram illustrating processing of the broadcast program storage function in the distribution system 1 according to the first embodiment.

  In FIG. 6, the system administrator terminal 60 makes a program storage / delivery request to the memory server manager 10 (S101). At this time, the system administrator terminal gives the memory server manager 10 information including the program number, the broadcast start time, end time, and viewing time limit of the program.

  The memory server manager 10 registers the program information received from the system administrator terminal 60 in the accumulated distribution management list 20 (S102). If the registration of the program information is successful, the memory server manager 10 responds to the system administrator terminal 60 in response to the program storage / delivery request (S103).

  The memory server manager 10 monitors the broadcast start time of each program registered in the accumulated distribution management list 20. When the broadcast start time of the program approaches (S104), the memory server manager 10 makes a memory server status acquisition request to all the memory servers 30-1 to 30-N (S105).

  When the memory servers 30-1 to 30-N receive the memory server status acquisition request from the memory server manager 10, the memory 31 and the HDD 32 are currently storing data, such as the storage status of the memory 31 and the HDD 32, for example. Whether the distribution is completed, the status of the distribution method, the distribution number of programs currently distributed on the memory 31 and the HDD 32, the scheduled memory capacity of the memory 31 and the HDD 32, the current storage capacity, and the like are monitored.

  Then, the memory servers 30-1 to 30-N respond to the memory server manager 10 with information including the monitored information and its own memory server number (S106).

  In FIG. 6, the memory server status acquisition requests and responses to all the memory servers 30-1 to 30-N are shown only once, but the memory server manager 10 is assumed to perform periodically.

  When the memory server manager 10 acquires the memory server status information from each of the memory servers 30-1 to 30-N, the memory server status information of each of the memory servers 30-1 to 30-N is stored in the storage / delivery management list 20. Update.

  The memory server manager 10 selects the memory server 30 that performs program distribution based on the memory server status information of each of the memory servers 30-1 to 30-N in the accumulated distribution management list 20 (S107).

  Here, the concept of the selection method of the memory server 30 will be described with reference to FIG.

  The memory server manager 10 makes a selection based on the memory server status information in the storage / delivery management list 20, but FIG. 7 illustrates and describes the status of the memory 1 of the memory server 30 for convenience of explanation. FIG. 7 illustrates the memory server states of the four memory servers 30-1 to 30-4.

  In FIG. 7, in the memory 31 of each of the memory servers 30-1 to 30-4, the hatched portion indicates the free memory capacity. The memory server manager 10 acquires a capacity necessary for data storage of the program in response to a program storage / delivery request from the system administrator terminal 60.

  The memory server manager 10 determines the free capacity of the memory 31 from the memory server status information of each of the memory servers 30-1 to 30-4. Then, the memory servers 30 having sufficient free space are listed. In the case of FIG. 7, it is assumed that the memory servers 30-2 and 30-3 have sufficient free space in the memory 31.

  The memory server manager 10 selects the memory server 30 with the smallest number of distributions as the memory server that distributes the program from among those having sufficient free space in the memory 31. In the case of FIG. 7, since the number of distributions of the memory server 30-2 is the smallest, the memory server 30-2 is selected as a program distribution server. Note that the number of distributions of the memory server 30 can be the sum of the number of streams being distributed in the program on the memory 31.

  When the stored program distribution server is selected as described above, the memory server manager 10 makes a stored distribution start request to the selected memory server 30 (S108). At this time, the memory server manager 10 provides the selected memory server 30 with information including the program number, program broadcast start time, end time, and viewing time limit.

  In response to this, the memory server 30 allocates the capacities of the memory 31 and the HDD 32 that store program data related to the storage and distribution start request. Only the capacity of program data to be stored is secured in the memory 31 and the HDD 32.

  When the program broadcast starts, the memory server 30 receives the received program data, and stores the received program data in the memory 31 and the HDD 32. Thereafter, when the storage / delivery start process is performed (S109), the memory server 30 makes a response to the storage / distribution start request to the memory server manager 10 (S110). At this time, the memory server 30 may give its own memory server state to the memory server manager 10 as distribution starts.

  When the memory server manager 10 receives the accumulated distribution start request response from the memory server 30 that has started the distribution process, the memory server manager 10 updates the contents of the accumulated distribution management list 20 based on the received memory server status information (S111).

(A-2-2) Stored Program Distribution Function FIG. 8 is a sequence diagram illustrating processing of the stored program distribution function in the distribution system 1 according to the first embodiment. Distribution of the stored program is performed when there is a distribution request from the client 40 until the end of the program.

  In FIG. 8, the client 40 makes a memory server allocation request with the program number of the program as an argument to the dispatcher 50 (S201). That is, the client 40 designates the program number of the program that the user views and makes a memory server allocation request to the dispatcher 50.

  The dispatcher 50 makes a memory server information acquisition request to the memory server manager 10 in order to obtain information from which memory server 30 the program with the program number designated by the client 40 is distributed (S202). .

  When the memory server manager 10 receives the memory server allocation request from the dispatcher 50, the memory server manager 10 refers to the stored / distributed management list 20 and selects a memory server 30 that can be distributed for the program number (S203).

  A method similar to the method described with reference to FIG. 7 can be applied to the selection of the memory server 30. That is, a list of memory servers having a sufficient memory capacity is listed, and the memory server 300 with the smallest number of distributions is selected from the list.

  As described above, each time a memory server allocation request is received, the memory server 30 to which the program is distributed is selected using the above list, and the memory server 30 having a small number of distributions is selected at the time of the request. Can do. As a result, the load related to the distribution of the memory server 30 can be distributed.

  When the memory server manager 10 selects the memory server 30 to distribute the program, the memory server manager 10 returns information including the memory server number (for example, address information) of the memory server 30 to the dispatcher 50 as a response to the storage / delivery memory server information acquisition request ( S204).

  The dispatcher 50 returns the address information of the memory server 30 that delivers the program of the designated program number to the client 40 as a response to the memory server allocation request (S205).

  When the client 40 acquires the address information of the memory server 30 to which the program is distributed, the client 40 requests the program distribution to the memory server 30 using the address information (S206).

  When the memory server 30 receives a distribution request for a program number from the client 40, the memory server 30 distributes the program having the program number to the client 40 (S207).

(A-2-3) Distribution Method Switching Process FIG. 9 is an explanatory diagram illustrating distribution method switching processing in the memory server 30 according to the first embodiment.

  FIG. 9A is a diagram for explaining data accumulation at the start of a program. When the broadcast of the program starts, the memory server 30 stores the received program data in both the memory 31 and the HDD 32. The program distribution uses data stored in the memory 31. That is, the distribution method at the start of program broadcasting is assumed to be memory distribution.

  Thereafter, the memory server 30 periodically monitors its own distribution status with the reception of a periodic memory server status acquisition request from the memory server manager 10.

  For programs whose distribution number on the memory 31 is lower than the threshold value α, the storage in the memory 31 is stopped and the data stored in the memory 31 is deleted. Furthermore, the program distribution is switched to the HDD distribution as shown in FIG. At this time, the memory 31 stores data of the next program. In this manner, by storing the next program data in the memory 31, it is possible to correspond to a start-over type storage and distribution system in which the number of programs to be distributed is large.

  Further, when the viewing deadline is reached, the memory server 30 deletes the data of the program stored in both the memory 31 and the HDD 32.

(A-3) Effect of First Embodiment As described above, according to the first embodiment, when the number of program distributions decreases with the end of the program, the memory distribution is switched to the HDD distribution, and new program accumulation is performed. Free the memory for use. As a result, a program with a small number of distributions does not occupy the memory, and a memory area in which a program requiring a large amount of subsequent distribution is released can be used. Therefore, the distribution efficiency of the memory server can be increased.

  In addition, according to the first embodiment, as a memory server for storing and distributing programs, a memory server that has sufficient free space in the memory to store programs and has the smallest number of memory distribution is selected. Therefore, the distribution efficiency of the memory server can be increased as much as possible.

(B) Second Embodiment Next, a second embodiment of the distribution system, distribution server management apparatus, distribution server management program, distribution server, and distribution program of the present invention will be described with reference to the drawings.

  The second embodiment is also an embodiment when the present invention is applied to a start-over type storage and delivery system.

  The first embodiment exemplifies the case of listing the memory server selection methods that have sufficient memory capacity for storing program data. However, in some cases, there is no memory server with sufficient capacity. Therefore, in the second embodiment, a method of starting distribution in such a case will be described.

(B-1) Configuration and Operation of Second Embodiment Since the configuration of the second embodiment is the same as that of the first embodiment illustrated in FIGS. 1, 4 and 5, again, FIG. This will be described with reference to FIGS. The operation of the second embodiment is basically the same as that of the first embodiment illustrated in FIGS. 6 and 8.

  FIG. 10 is a flowchart for explaining a method by which the memory server manager 10 selects the memory server 30 that performs program distribution.

  FIG. 10 shows processing when there is not enough free space in the memory and the memory server allocation described in the first embodiment fails.

(A) About S301 In the memory server manager 10, the memory server selection unit 113 creates a memory free capacity transition table of each memory server 30 (S301).

  The memory server selection unit 113 extracts the memory server status information acquired from each of the memory servers 30-1 to 30-N from the storage / delivery management list 20, and based on the memory server status information, each memory server 30-1 A memory free capacity transition table for ˜30-N is generated.

  FIG. 11 is a configuration diagram showing a configuration example of the memory free capacity transition table. The memory free capacity transition table is used to predict the switching time from memory distribution to HDD distribution based on the number of programs distributed during memory distribution, program end time, program viewing time limit, etc. 6 is a table in which (a value obtained by accumulating the amount of memory 31 released) is associated. Note that FIG. 11B illustrates a case where the table in FIG. 11A is represented by a C language structure.

  First, the memory server selection unit 113 estimates the time for the number of distributions of each program being distributed in memory to reach the threshold value α in order to generate the memory free capacity transition table.

  Here, the time for the number of distributions of programs that have not yet been broadcast to reach the threshold value α is obtained as follows.

  It is assumed that after the program ends, the distribution number of the program decreases linearly until the viewing deadline.

The number of distributions at time t is the number of distributions at the end of the program (here, assuming that the number of distributions at the time of creating the transition table is maintained until the program end time) a, program end time b, and program viewing time limit c become that way.

Therefore, the time when the number of distributions is the threshold value α is as shown in Expression (2).

  In addition, it is assumed that a program whose end time has passed linearly decreases until the viewing deadline.

The number of distributions at time t is as follows, assuming the number of distributions a at the time of creating the memory free capacity transition table, the time b at the time of creation of the transition table, and the program viewing time limit c.

Therefore, the time when the number of distributions is the threshold value α is as shown in Expression (4).

The memory server selection unit 113 creates a memory free capacity transition table based on the time when the number of distributions obtained above becomes the threshold value α (hereinafter referred to as distribution method switching time).
(1) The memory server selection unit 113 calculates the distribution method switching time for all programs being distributed in memory, and generates a list in which the distribution method switching time and the memory capacity allocated to the program are paired.
(2) The memory server selection unit 113 arranges the list generated in (1) in ascending order of time.
(3) When the delivery method switching times are the same, the memory server selection unit 113 integrates the memory capacities of those pairs into one (adds the used memory amount).
(4) Accumulate used memory capacity.

(B) About S302 Next, the memory server selection unit 113 creates a list of memory servers in which the free space at each time point in the memory free space transition table exceeds the expected accumulation amount of the program (S302).

  That is, the memory server selection unit 113 compares the memory free capacity transition table with the transition of the memory consumption due to program accumulation, and confirms whether the transition of the memory free capacity can absorb the transition of the memory consumption.

  FIG. 12 is an explanatory diagram for explaining the relationship between the estimated memory free space transition and the storage amount necessary for storing the program. In FIG. 12, a thick solid line indicates the accumulation amount necessary for program accumulation. The thin solid line and dotted line show the transition of the estimated memory free capacity. It can be seen that the accumulation amount necessary for accumulation indicated by a thick solid line increases with time from the beginning of accumulation.

  In FIG. 12, when there is a free capacity as shown by a thin solid line, it always exceeds the accumulation amount necessary for accumulation (thick solid line). Therefore, in this case, the packet accumulation amount can be covered. That is, even if the memory capacity is insufficient at the start of accumulation and delivery, there is no problem because the memory capacity exceeds the thick solid line in this case.

  In addition, when there is a free capacity as indicated by the dotted line, the free memory capacity exceeds the storage capacity necessary for storage, but the memory capacity is insufficient in the middle. Therefore, it is necessary to perform HDD distribution from the middle. If a sufficient number of HDD distributions cannot be secured, a part of the distributions must be interrupted.

The memory server selection unit 113 determines whether or not the transition of the free memory capacity satisfies the following formula in order to create a list of the memory servers 30 that can absorb the transition of the storage amount for storing the program data. .

  The memory server selection unit 113 converts the memory server 30 that satisfies the condition of Expression (5) into a list.

(C) About S303 The memory server selection unit 113 creates the list, and when the list is not empty, that is, when there is a memory server 30 that satisfies the condition of Expression (5), the process proceeds to S304.

  On the other hand, when the list is empty, that is, when there is no memory server 30 that satisfies the condition of Expression (5), the process proceeds to S305.

(D) Regarding S304 When the list of memory servers 30 satisfying the condition of Expression (5) is generated, the memory server selection unit 113 selects the one with the smallest memory distribution number from the memory servers 30 in the list. select.

(E) About S305 This is a case where a list of memory servers 30 that satisfy the condition of Expression (5) cannot be created. That is, this is a case where no free space is expected in the memory 31. In this case, the memory server selection unit 113 selects the memory server 30 with the smallest HDD distribution number. The selected memory server 30 starts program distribution by HDD distribution.

  The method described above is a method of selecting the memory server 30 to which the memory server manager 10 distributes a program. When each memory server 30-1 to 30-N receives the memory server status acquisition request, it must respond to the memory server manager 10 with information necessary for the calculation of the above conditions.

  In S305 of FIG. 10, the memory server 30 selected by the memory server manager 10 does not have sufficient memory capacity, so the program is distributed by switching from memory distribution to HDD distribution.

  The memory server 30 has an insufficient memory capacity and does not have a capacity necessary for storing program data. In order to store in the memory 31 a program that has started to be distributed with insufficient capacity so that the program can be distributed in the memory 31, a reservation is made in the area of the memory 31 to store the data of the program.

  In general, when the memory 31 becomes an empty area, subsequent new program data is accumulated. However, in order to store not the following program but the program that has been initially distributed due to insufficient capacity in the memory 31, it is necessary to reserve the program to be stored in the memory 31. Therefore, when the memory server 30 runs short of memory capacity and switches from memory delivery to HDD delivery, the memory area of the memory 31 is integrated as follows, and the capacity is insufficient in the integrated area. Reserve a program that has just started being delivered.

  FIG. 13 is an explanatory diagram for explaining integration of memory areas of the memory 31. In FIG. 13A, a part of the amount necessary for program accumulation is allocated to the empty area of the memory 31. At this time, the memory 31 delivers the accumulated data already accumulated in the memory.

  In FIG. 13A, when the accumulated data is delivered from the memory 31 to the memory, as shown in FIG. 13B, the area after the memory delivery (empty area) and a part assigned to the program accumulation are shown. Integrate with other areas. Further, the area distributed in the memory is distributed in the HDD.

  In FIG. 13B, when the accumulated data is further delivered from the memory 31, the areas after the delivery of the memory are further integrated as shown in FIG. 13C. Further, the area distributed in the memory is distributed in the HDD.

  By integrating the memory areas as described above, an area having a capacity necessary for program accumulation can be secured, and the memory 31 can be used for accumulating programs started to be delivered with insufficient capacity.

  In addition, the reservation method for accumulating programs that have started distribution with insufficient capacity in the memory 31 is not particularly limited, and various methods can be widely applied. For example, it is possible to apply a method in which a reservation flag is attached to the memory 31 to be used in the future so that the subsequent program cannot be used at the time of allocation of accumulated distribution. Specifically, for example, it can be realized by adding the program to the stored distribution management list 20 as the next item.

(B-2) Effects of Second Embodiment As described above, according to the second embodiment, in addition to the effects of the first embodiment, the memory server manager distributes memory for a program being distributed. By predicting the time for switching from HDD distribution to HDD distribution, it is possible to start accumulation distribution even if there is not enough free space.

(C) Third Embodiment Next, a third embodiment of the distribution system, distribution server management apparatus, distribution server management program, distribution server, and distribution program of the present invention will be described with reference to the drawings.

  Similarly to the first embodiment, the third embodiment is an embodiment in the case where the present invention is applied to a start-over type storage and delivery system.

  In the first embodiment, as an example, the case where one memory server for program distribution is selected has been described. However, there are cases where the required number of distributions cannot be ensured with only one memory server. Therefore, in the third embodiment, a case where a plurality of memory servers perform storage and distribution will be described.

(C-1) Configuration and Operation of Third Embodiment Since the configuration of the third embodiment is the same as that of the first embodiment illustrated in FIGS. 1, 4 and 5, again, FIG. This will be described with reference to FIGS. The operation of the third embodiment is basically the same as that of the first embodiment illustrated in FIGS. 6 and 8.

  The third embodiment differs from the first and second embodiments in that a memory server selection process and a distribution memory server allocation process in the memory server manager 10 (when the client 40 issues a server allocation request). Process).

  For the memory server 30 selection process, the memory server selection unit 113 allocates the memory server 30 until the target number of distributions is reached.

  FIG. 14 is a sequence diagram illustrating processing of the broadcast program storage function in the distribution system 1 according to the third embodiment.

  First, the system administrator terminal 60 makes a program storage / delivery request to the memory server manager 10 (S401). At this time, the system administrator terminal 60 designates the necessary number of distributions and gives the necessary number of distributions to the memory server manager 10.

  The memory server manager 10 registers the program information of the distribution program in the storage distribution management list 20 (S402), and returns a response to the program storage request to the system administrator terminal 60 (S403).

  Similarly to the first embodiment, the memory server manager 10 makes a memory server status acquisition request to all the memory servers 30-1 to 30-N (S404), and each of the memory servers 30-1 to 30- N returns its own memory state and the like to the memory server manager 10 as a memory server state acquisition request response (S405).

  In the memory server manager 10, the memory server selection unit 113 selects the memory server 30 until the required number of distributions from the system administrator terminal 10 is reached (S406).

  As a method for selecting a plurality of memory servers 30, various methods can be widely applied. For example, the following method can be applied.

  For example, the required number of distributions is x. First, as in the first and second embodiments, the memory server selection unit 113 creates a list and selects one memory server 30. At this time, the memory server selection unit 113 acquires the number of distributions that can be allocated by the selected memory server 30 from the stored distribution management list 20. The number of distributions that can be distributed by the memory server selection unit 113 is y.

  Then, the memory server selection unit 113 performs z = xy, and repeatedly selects the memory server 30 from the list until z becomes 0. Note that z = 0 when z <0.

  Thereafter, in S407 to S413, the memory server manager 10 makes an accumulation and distribution start request to all the selected memory servers 30. In addition, each memory server 30 performs an accumulation / delivery start process, and returns a response to the accumulation / delivery start request to the memory server manager 10. The memory server manager 10 updates the accumulated distribution management list 20.

  Although not shown in FIG. 14, when the memory server manager 10 receives a storage / delivery memory server information acquisition request from the dispatcher 50, the memory server manager 10 determines the number of distributions from a plurality of servers in the storage / distribution management list for the designated program. Select one of the least servers.

(C-2) Effects of the Third Embodiment As described above, according to the third embodiment, in addition to the effects of the first and second embodiments, it is necessary for the first and second embodiments. However, it is possible to satisfy the specified number of distributions by specifying a plurality of storage and distribution memory servers.

(D) Fourth Embodiment Next, a fourth embodiment of the distribution system, distribution server management apparatus, distribution server management program, distribution server, and distribution program of the present invention will be described with reference to the drawings.

  In the fourth embodiment, as in the first embodiment, an embodiment in which the present invention is applied to a start-over type storage and delivery system will be described.

  If there is no free space in the memory of any memory server at the program start time, HDD distribution can only be started with a small number of distributions. In such a case, if the storage and distribution of the program is reserved in advance, it does not mean that there is no server for memory distribution immediately before.

  Thus, in the fourth embodiment, a description will be given of an embodiment in which temporary allocation processing of a memory server is performed when program information is registered.

(D-1) Configuration and operation of the fourth embodiment The configuration of the fourth embodiment is the same as that of the first embodiment illustrated in FIGS. This will be described with reference to FIGS. The operation of the fourth embodiment is basically the same as that of the first embodiment illustrated in FIGS. 6 and 8.

  The fourth embodiment differs from the first to third embodiments in the processing of the program information registration unit 111 in the memory server manager 10.

  When the program information registration unit 111 according to the fourth embodiment receives a program accumulation / delivery request from the system administrator terminal 60, the program information registration unit 111 performs the following process and temporarily allocates a memory server that distributes the program.

  First, the memory server manager 10 acquires eye memory server status information from all the memory servers 30-1 to 30 -N, and stores / updates it in the accumulated distribution management list 20.

  Next, the memory server manager 10 creates a memory free capacity transition table for each of the memory servers 30-1 to 30-N. This memory free capacity transition table is created by the method described in the second embodiment. Further, when creating the memory free capacity transition table, the memory free capacity of the memory server 30 for only the program being distributed is taken into consideration.

  The memory server manager 0 corrects the memory free capacity transition table.

  Various methods can be applied to the method of correcting the memory free capacity transition table. In this embodiment, the following method is applied.

(1) fst_table [i] .free_space (fst_table [i] .time <viewing deadline) (2) From the start time of the program requested for program storage and distribution until the viewing deadline, the expected memory usage of the program scheduled to start is fst_table [i] .free_space (fst_table [i] .time ≧ viewing deadline) Subtract from.

  Then, the transition of the memory consumption of the program requested to start the program storage / delivery is compared with the memory free capacity transition table corrected above, and a list of the memory servers 30 that can cover the memory 31 necessary for the program storage / delivery is created. As a method for creating this list, a method similar to the processing in S302 of FIG. 10 of the second embodiment can be applied.

  The memory server 30 with the smallest number of programs to be simultaneously distributed is selected between the program start time and the viewing deadline of the program requested to be stored and distributed, and the selected memory server 30 is selected as the temporary storage and distribution memory server 30. And

  In the above operation, when a program storage / delivery request is received from the system administrator terminal 60 and a distribution program is registered in the storage / distribution management list 20, the memory server manager 10 stores and distributes the program. Make temporary assignments.

  Further, immediately before the program starts, the memory server manager 10 selects the memory server 30 (this is referred to as main allocation), and the method of the main allocation can be the same as the temporary allocation method.

  That is, in consideration of the memory server 30 reserved by temporary allocation, the memory server 30 having the sufficient capacity of the memory 31 and the smallest number of distributions may be selected using the new memory server state information at the time of selection. Good.

(D-2) Effect of Fourth Embodiment As described above, according to the fourth embodiment, provisional allocation is performed at the stage of program registration before selection (main allocation) of the memory server immediately before the program start time. By doing this, you can secure the memory of the memory server.

(E) Other Embodiments In the first to fourth embodiments, the case where the present invention is applied to a start-over type broadcast program storage / distribution system is illustrated. It can be widely applied to systems.

  Various functions of the memory server and the memory server manager of the first to fourth embodiments can be realized by so-called software processing. For example, the hardware configuration of the memory server and the memory server manager is composed of a CPU, RAM, ROM, EPROM, etc., and the CPU reads the processing program stored in the ROM, and the CPU executes the processing program. Realized.

1 ... delivery system, 20 ... stored delivery management list,
DESCRIPTION OF SYMBOLS 10 ... Memory server manager, 11 ... Control part, 111 ... Program information registration part,
112 ... Memory server status acquisition unit, 113 ... Memory server selection unit,
114 ... Accumulation / distribution request unit, 115 ... Program distribution memory server information response unit,
12 ... communication section,
30-1 to 30-N ... Memory server, 31 ... Memory, 32 ... HDD,
33 ... Control unit, 331 ... Program storage / distribution processing unit, 332 ... State management unit,
333: Distribution method switching control unit, 34: Communication unit,
40 ... client, 50 ... dispatcher.

Claims (8)

The first storage means and the second storage means for storing the content to be distributed to the client, and when the distribution request is received from the client, the above-mentioned stored in either the first storage means or the second storage means A plurality of distribution servers for distributing content;
A delivery server management device that selects the delivery server that delivers the content from among the plurality of delivery servers, and notifies the client of the address information of the selected delivery server;
The delivery server management device is
Accumulated distribution status information acquisition means for periodically acquiring accumulated distribution status information indicating the accumulation status and distribution status of the first accumulation means and the second accumulation means in each of the distribution servers;
Content information acquisition means for acquiring content information related to the content from outside;
Information storage means for storing the accumulated distribution status information and the content information of the plurality of distribution servers;
Based on the stored distribution status information of the plurality of distribution servers, the first storage means has a capacity that exceeds the required storage amount of the content included in the content information and has the minimum number of distributions. Server selection means for selecting the distribution server as the one to distribute the content;
Storage distribution request means for giving the content information related to the content to the distribution server selected by the server selection means and requesting storage distribution of the content;
Server information notifying means for notifying the client of the address information of the delivery server selected by the server selecting means when receiving a notification of the delivery destination of the content from the client system.   If the server selection unit does not have the distribution server having the first storage unit having a capacity that exceeds the required storage amount of the content, the second storage unit starts from the distribution from the first storage unit in each distribution server. 2. The time for switching to distribution from the network is predicted, and the distribution server in which the transition of the free capacity of the first storage means exceeds the transition of the storage amount of the content is selected. Distribution system.   3. The distribution system according to claim 1, wherein when the number of distributions of the content is designated, the server selection unit selects a plurality of distribution servers so as to satisfy the number of distributions. 4. .   When acquiring the content information, based on the accumulated distribution information of each of the distribution servers, the capacity of the first accumulation means at the time of the content distribution is predicted, and the distribution server having the necessary accumulation amount of the content, 4. The distribution system according to claim 1, further comprising selection server reservation means for pre-assignment. The first storage means and the second storage means for storing the content to be distributed to the client, and when the distribution request is received from the client, the above-mentioned stored in either the first storage means or the second storage means In the distribution server management apparatus that selects the distribution server that distributes the content from among a plurality of distribution servers that distribute the content, and notifies the client of the address information of the selected distribution server.
Accumulated distribution status information acquisition means for periodically acquiring accumulated distribution status information indicating the accumulation status and distribution status of the first accumulation means and the second accumulation means in each of the distribution servers;
Content information acquisition means for acquiring content information related to the content from outside;
Information storage means for storing the accumulated distribution status information and the content information of the plurality of distribution servers;
Based on the stored distribution status information of the plurality of distribution servers, the first storage means has a capacity that exceeds the required storage amount of the content included in the content information and has the minimum number of distributions. Server selection means for selecting the distribution server as the one to distribute the content;
Storage distribution request means for giving the content information related to the content to the distribution server selected by the server selection means and requesting storage distribution of the content;
Server information notifying means for notifying the client of the address information of the delivery server selected by the server selecting means when receiving a delivery destination notification request for the content from the client. Server management device. The first storage means and the second storage means for storing the content to be distributed to the client, and when the distribution request is received from the client, the above-mentioned stored in either the first storage means or the second storage means In a distribution server management program for selecting the distribution server that distributes the content from among a plurality of distribution servers that distribute the content, and notifying the client of the address information of the selected distribution server,
Computer
Accumulated distribution status information acquisition means for periodically acquiring accumulated distribution status information indicating the accumulation status and distribution status of the first accumulation means and the second accumulation means in each of the distribution servers;
Content information acquisition means for acquiring content information related to the content from outside,
Information storage means for storing the accumulated distribution status information and the content information of the plurality of distribution servers;
Based on the stored distribution status information of the plurality of distribution servers, the first storage means has a capacity that exceeds the required storage amount of the content included in the content information and has the minimum number of distributions. Server selection means for selecting a distribution server as a distribution of the content;
Storage distribution request means for giving the content information related to the content to the distribution server selected by the server selection means and requesting storage distribution of the content;
A distribution server management program for functioning as server information notification means for notifying the client of the address information of the distribution server selected by the server selection means upon receiving a notification of the distribution destination of the content from the client. The first storage means and the second storage means for storing the content to be distributed to the client, and when the distribution request is received from the client, the above-mentioned stored in either the first storage means or the second storage means A distribution server for distributing content,
A delivery switching control means for controlling delivery of the content stored in either the first storage means or the second storage means;
In response to a request from the distribution server management device, the storage distribution state including the storage state and distribution state of the first storage unit and the second storage unit is monitored, and the storage distribution state information is returned to the distribution server management device. A distribution server comprising: a state management unit that performs: The first storage means and the second storage means for storing the content to be distributed to the client, and when the distribution request is received from the client, the above-mentioned stored in either the first storage means or the second storage means A distribution program for distributing content,
Computer
A delivery switching control means for controlling delivery of the content stored in either the first storage means or the second storage means;
In response to a request from the distribution server management device, the storage distribution state including the storage state and distribution state of the first storage unit and the second storage unit is monitored, and the storage distribution state information is returned to the distribution server management device. A distribution program that functions as a state management means.

Images