JP2008071229A - Network av system and controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a network AV system capable of suppressing the occurrence of content breaks in a streaming-delivered content file. <P>SOLUTION: A control point CP computes predicted load applied to a server SV1 when the server SV1 storing a music file MU11 decodes the music file MU11 and carries out streaming delivery (S104-S106). When the predicted load of the server SV1 is less than reference load (Yes in S107), a delivery server that delivers the music file MU11 is set as the server SV1 (S118). When the predicted load of the server SV1 exceeds the reference load (No in S107), the server whose predicted load is less than the reference load, out of other servers SV2, SV3 is set as a delivery server (S108-S112). Since the server whose predicted load is low delivers the music file, the occurrence of a sound break or the like can be suppressed. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a network AV system and a controller, and more particularly, includes a server that stores a plurality of content files, a renderer that plays back content files that are streamed from the server, and a controller that can be connected to the server and the renderer. The present invention relates to a network AV system and a controller.

  Recently, a network AV system that connects personal computers and AV (Audio Visual) devices in the home and uses them in cooperation has appeared. For example, in a network AV system based on DLNA (Digital Living Network Alliance) guidelines, a controller called a control point has a media server that is a source device for storing content such as video and music, and a playback device (sink device) that plays back the content. Control the media renderer.

  In a network system based on such DLNA guidelines, when reproducing a content file by streaming, the server decodes the content file into a predetermined file format and then delivers the content to the renderer. For example, the server decodes the mp3 format music file into the LPCM format and then delivers the streamed stream to the renderer.

  However, when the server decodes the content file and performs streaming delivery, the load on the server increases, and as a result, there is a problem that video and audio reproduced by the renderer are interrupted.

Japanese Patent Laying-Open No. 2004-102339 (Patent Document 1) discloses a technique in which a transcode server decodes a content file in a content server into a predetermined file format and transmits it to a client terminal. However, if the transcoded server performs streaming delivery of the decoded content file to the client terminal, the content may be interrupted due to an increase in the load on the transcoded server.
JP 2004-102339 A

  An object of the present invention is to provide a network AV system capable of suppressing the occurrence of content interruption in a content file distributed in streaming.

Means for Solving the Problems and Effects of the Invention

  The network AV system according to the present invention includes a plurality of servers connectable to each other, a renderer connectable to the plurality of servers, and a controller connectable to the plurality of servers and the renderer. The server includes content storage means, distribution means, and load information transmission means. The content storage unit stores a plurality of content files. In response to a request from the renderer, the distribution means decodes the selected content file and distributes the content file to the renderer. The load information transmission means transmits load information related to the load on itself to the controller. The controller includes storage means, server identification means, predicted load calculation means, and reproduction instruction means. The storage unit stores the load information transmitted from each server and the distribution load information related to the load applied to each server by decoding and streaming distribution of the content file. The server specifying means specifies a server that stores the selected content file based on the content information. The predicted load calculation means calculates a predicted load applied to the server specified when distributing the selected content file based on the specified server load information and the specified server distribution load information. When the calculated predicted load is less than the reference load, the reproduction instruction means transmits a reproduction instruction to the renderer to receive and reproduce the selected content file from the specified server. The renderer includes request means and playback means. The requesting unit requests the server for streaming distribution of the selected content file based on the reproduction instruction. The reproduction means reproduces the distributed content file.

  In the network AV system according to the present invention, the control point specifies the server that stores the selected content file, and the specified server is applied to the specified server when the selected server decodes and distributes the selected content file. Predict the load. Then, the control point instructs the reproduction of the selected content file when the predicted load is less than the reference load. Since the content file is distributed from the specific server when the predicted load is low, it is possible to suppress the interruption of the content being reproduced.

  Preferably, the predicted load calculation means further calculates a predicted load applied to each server when each server distributes the selected content file based on the load information of each server and the distribution load information of each server. The controller further includes search means and acquisition instruction means. When the predicted load of the identified server exceeds the reference load, the search unit searches for a server having a predicted load less than the reference load among the plurality of servers. When the search unit searches for a server whose predicted load is less than the reference load, the acquisition instruction unit transmits an acquisition instruction to acquire the selected content file from the specified server. The server further comprises acquisition means. The acquisition unit acquires the selected content file from the specified server when receiving the acquisition instruction. When the retrieval unit retrieves the server, the reproduction instruction unit transmits a reproduction instruction to the renderer so as to receive and reproduce the selected content file from the retrieved server.

  Thereby, when the predicted load of the specified server is high, the server with the low predicted load substitutes for the decoding and streaming delivery of the content file on behalf of the specified server. Since the distribution of the content file is executed by a server having a low predicted load, it is possible to suppress the content from being interrupted.

  Preferably, the search means further searches for a server having the smallest predicted load when there are a plurality of servers having the predicted load less than the reference load.

  In this case, the content file in which the server with the smallest predicted load is selected is distributed. Therefore, it is possible to prevent the content from being interrupted.

  Preferably, the load information is composed of time-series load data related to the load applied to the server. The predicted load calculation unit includes a reproduction time calculation unit, a time series data calculation unit, and an addition unit. The reproduction time calculation means calculates the reproduction time of the selected content file. The time-series data calculating means predicts time-series load data until the reproduction of the content file selected from the current time is completed based on the load information and the reproduction time by an extrapolation method. The adding means adds the distribution load information to the predicted time-series load data to obtain a predicted load composed of the time-series predicted load data. The reproduction instruction means transmits a reproduction instruction when all of the time-series predicted load data are less than the reference load.

  In this case, the prediction load applied to the server during the reproduction of the content can be obtained in time series. Therefore, the accuracy of prediction is improved.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

[overall structure]
Referring to FIG. 1, a network AV system 1 according to an embodiment of the present invention includes a control point CP, a plurality of servers SV1 to SV3 (collectively referred to as servers SV), and a renderer RE. These correspond to the DLNA guidelines and are connected to each other based on the UPnP protocol. Although there is one renderer RE shown in FIG. 1, there may be two or more. Also, although there are three server SVs in FIG. 1, the number of server SVs may be two, or four or more.

  The control point CP controls the servers SV1 to SV3 and the renderer RE. The control point CP selects a desired music file among a plurality of music files stored in the servers SV1 to SV3 according to a user operation. Then, the renderer RE is instructed to acquire and play the selected music file from the server SV.

  In response to a request from the renderer RE, the servers SV1 to SV3 decode the selected music file and perform streaming distribution. The renderer RE receives the streamed music file and plays the music. Hereinafter, the configurations of the control point CP, the server SV, and the renderer RE will be described.

[Control point]
Referring to FIG. 2, the control point CP includes a CPU (Central Processing Unit) 11, a memory 12, a communication unit 13 for communicating with the servers SV1 to 3 and the renderer RE, and a hard disk drive (hereinafter referred to as HDD). 14 and a display 15.

  The HDD 14 stores a load information database 141, a content information database 142, and a distribution load database 143.

  In the load information database 141, as shown in FIG. 3, load information concerning each of the servers SV1 to SV3 is registered. Here, the load information is composed of time-series load data applied to each of the servers SV1 to SV3. In the present embodiment, the load data is the CPU usage rate of each of the servers SV1 to SV3. For example, in FIG. 3, the load data (CPU usage rate) of the server SV1 at time t01 is 3%. The load data is transmitted from each server SV to the control point CP every predetermined time.

  In the content information database 142, as shown in FIG. 4, information on the details and locations of a plurality of music files stored in the servers SV1 to SV3 (hereinafter referred to as content information) is registered. Specifically, the content information includes a song ID (MU11, MU21, MU31, etc.) for specifying each song file, song detailed information such as song title and singer name, and URI (location information of song file). Uniform Resource Identifier). The content information is transmitted from each server SV to the control point CP every time the music file stored in each server SV is updated.

  In the distribution load database 143, as shown in FIG. 5, load data (hereinafter referred to as distribution load) when each server SV decodes a music file and performs streaming distribution is registered. The distribution load may be measured in advance by each server SV, or may be registered in advance based on past results.

  The HDD 14 further stores a controller program 144. When the controller program 144 is loaded into the memory 12 and the CPU 11 executes the controller program 144 loaded into the memory 12, the control point CP performs the operation described later.

  The display 15 displays a playlist that lists songs stored in each server SV, and displays the operation status of the renderer RE (song playback, stopped, etc.). The playlist is created based on the content information database 142 and displayed on the display 15. The user refers to the playlist displayed on the display 15 and selects a desired music file.

[server]
Referring to FIG. 6, server SV includes CPU 21, memory 22, communication unit 23 for communicating with control point CP, renderer RE, and the like, and HDD 24.

  The HDD 24 stores a content database 241. A plurality of music files are registered in the content database 241. Further, content information corresponding to a plurality of registered music files is registered in the content database 241. When a new music file is added to or deleted from the content database 241, that is, when the content database 241 is updated, the server SV updates the content information and transmits it to the control point CP.

  The HDD 24 further stores a server program 242. When the server program 242 is loaded into the memory 22 and the CPU 21 executes the server program 242 loaded into the memory 22, the server SV performs an operation described later.

[Renderer]
Referring to FIG. 7, the renderer RE includes a CPU 31, a memory 32 for storing a renderer program, a communication unit 33 for communicating with the outside, and a playback unit 34 for playing a music file streamed from a server. With. When the CPU 31 executes the renderer program stored in the memory 32, the renderer RE executes the operations described later.

[Operation]
The operation of the network AV system 1 having the above configuration will be described.

  Referring to FIG. 8, the control point CP stores load data periodically transmitted from each server in the load information database 141 (1).

  When the user who refers to the playlist displayed on the display 15 selects the desired music file MU11, an instruction to reproduce the music file MU11 is input to the control point CP according to the user operation (2).

  At this time, the control point CP decodes and distributes the music file MU11 from the servers SV1 to SV3 (hereinafter referred to as “distribution”) in order to suppress the interruption of sound during the reproduction of the music file MU11. (Referred to as server) (distribution server determination process: (3)).

  In the distribution server determination process, the control point CP predicts the load applied to each server when each server SV decodes and distributes the music file MU11. If the predicted load of the server SV1 that stores the music file MU11 is small, the server SV1 is determined as the distribution server (CASE1). In this case, the control point CP transmits a reproduction instruction to the renderer RE (4), and the renderer RE requests the server SV1 to distribute the music file MU11 based on the reproduction instruction (5).

  On the other hand, when the predicted load on the server SV1 is large, sound interruption may occur if the server SV1 becomes a distribution server. Therefore, in this case, another server SV (for example, server SV2) having a small predicted load acts as a proxy for decoding and distributing the music file MU11 (CASE2). When another server SV with a small predicted load acts as a proxy, the occurrence of sound interruption can be suppressed.

  When the server SV2 is a distribution server, the server SV2 does not store the music file MU11. Therefore, the control point CP instructs the server SV2 to acquire the music file MU11 (6). When the server SV acquires and stores the music file MU11 based on the acquisition instruction, the server SV2 transmits the location information (URI) of the acquired music file MU11 to the control point CP (7).

  The control point CP instructs the renderer RE to play the music file MU11 based on the transmitted URI (8). The renderer RE acquires the music file MU11 from the server SV2 and reproduces it (9).

  With the above operation, the network AV system 1 suppresses occurrence of sound interruption in the music file being reproduced. Hereinafter, the operation process of the control point CP, the server SV, and the renderer RE will be described in detail.

[Distribution server decision processing]
Referring to FIG. 9, when music file MU11 is selected in response to a user operation and a playback instruction for the selected music file MU11 is input (YES in S101), control point CP refers to content information database 142. Then, the server that stores the content file MU11 is identified as the server SV1 (S102).

  Subsequently, the control point CP obtains a predicted load on the server SV1 when the identified server SV1 streams the content file MU11 (S103 to S106). First, the control point CP obtains the reproduction time Δt1 of the music file MU11 (S103). The reproduction time Δt1 may be determined based on the time stamp in the file, or may be calculated from the data capacity of the music file MU11.

  After obtaining the reproduction time Δt1, the load information database 141 is referred to, and predetermined time from the current time (that is, the time when the reproduction instruction of the selected music file is inputted) t1 among the time series load data of the server SV1. The load data between the time t1−Δt0 retroactive to the period Δt0 and the time t1 is read (S104). Subsequently, based on the load data read in step S104, time-series load data (approximate function) from time t1 to time t1 + Δt1 is calculated by extrapolation (S105). For example, referring to FIG. 10, when time-series load data up to time t1 is indicated by a solid line in FIG. 10, control point CP reads load data C0 between time t1−Δt0 and time t1 in step S104. In step S105, load data (approximation function) C1 represented by a dotted line is calculated by extrapolation.

  After calculating the load data, the control point CP calculates a predicted load composed of time-series predicted load data based on the following equation (1) (S106).

Predicted load data = load data calculated in step S105 + distributed load data (1)
The distribution load data here is distribution load data of the server SV1 and is read from the distribution load database 143. In FIG. 10, an alternate long and short dash line curve C2 obtained by adding the distribution load data to the load data C1 is the predicted load. With the above operation, the predicted load of the server SV1 at the times t1 to t1 + Δt is calculated.

  The control point CP determines whether or not the calculated predicted load is less than the reference load Lref (S107). The reference load Lref is an upper limit of a load at which no sound break occurs when each of the servers SV1 to SV3 reproduces the music file MU. That is, when the load on the server SV1 exceeds the reference load Lref, sound interruption may occur. The reference load Lref is stored in advance in the HDD 14 in the control point CP.

  In step S107, the control point CP determines that the predicted load is less than the reference load Lref when all of the time-series predicted load data between time t1 and time t1 + Δt1 is less than the reference load Lref (YES in S107). . On the other hand, when at least a part of the predicted load data between time t1 and time t1 + Δt exceeds the reference load Lref, it is determined that the predicted load exceeds the reference load Lref (NO in S107).

  As a result of the determination in step S107, as shown in FIG. 10, when the predicted load C2 of the server SV1 is less than the reference load Lref of the server SV1 (YES in S107), the server SV1 distributes the song file MU11 (hereinafter referred to as “server SV1”). (Referred to as a distribution server) (S118). This is because, since the predicted load is lower than the reference load Lref, it is predicted that no sound interruption will occur even if the server SV1 distributes the music file MU11.

  On the other hand, as a result of the determination in step S107, if the predicted load of the server SV1 exceeds the reference load Lref of the server SV1 (NO in S107), the predicted load is calculated for the other servers SV2 and SV3 (S108). If the music file MU11 is distributed by the server SV1, there is a high possibility that a sound interruption will occur. Therefore, another server (SV2 or SV3) with a low predicted load is made to act as a proxy for the decoding and distribution of the music file MU11. Details of step S108 will be described below.

  Referring to FIG. 11, the process of step S108 is executed for each server (SV2 and SV3) excluding server SV1. Hereinafter, although the process of step S108 with respect to server SV2 is demonstrated, the process with respect to server SV3 is also the same.

  First, the control point CP calculates time (hereinafter referred to as download time) Δt3 required to download and distribute the music file MU11 from the server SV1 to the server SV2 (S401). The control point CP stores the inter-server network speed database shown in FIG. The control point CP refers to the inter-server network speed database and reads the speed aa3 (Bytes / s) when the file is distributed from the server SV1 to the server SV2. Subsequently, the data capacity of the music file MU11 is read, and the read data capacity is divided by the speed aa3 (Bytes / s) to calculate the download time Δt3.

  Subsequently, the control point CP refers to the load information database 141 and reads the load data of the server SV2 between time t1−Δt0 and time t1 (S402).

  After obtaining the download time Δt3 and reading the load data, the control point CP calculates the load data between time t1 + Δt3 and time t1 + Δt1 + Δt3 by extrapolation (S403). Compared with step S105, the time zone for obtaining load data is shifted by Δt3. This is because, when the server SV2 distributes the music file MU11, the music file MU11 must be downloaded from the server SV1 in advance. caused by. That is, the time at which the server SV2 distributes the music file MU11 is from the time t1 + Δt3 when the music file MU11 is downloaded to the time t1 + Δt + Δt3. Therefore, the load data in this time zone is calculated by extrapolation.

  After calculating the load data, the control point CP calculates the predicted load based on the equation (1) (S404). At this time, the distribution load data of the server SV2 in the distribution load database is read and substituted into the equation (1).

  With the above operation, the predicted loads of the servers SV2 and SV3 are calculated.

  Returning to FIG. 9, after calculating the predicted loads of the servers SV2 and SV3 in step S108, the control point CP searches for a server having a predicted load less than the reference load Lref (hereinafter referred to as a compatible server) (S109). Specifically, the control point CP reads the reference load Lref stored in the HDD 14 and determines whether all of the predicted load data of the server SV2 is less than the reference load Lref. The control point CP further determines the server SV3 in the same manner. In short, for all servers (SV2, SV3), it is determined whether the calculated predicted load data is less than the reference load Lref.

  As a result of the search in step S109, if there are compatible servers (YES in S109), the control point CP determines whether there are a plurality of compatible servers (S110).

  As a result of the determination in step S110, when there is only one compatible server (NO in S110), for example, when only the server SV2 is a compatible server, the control point CP determines the server SV2 as a distribution server (S118). ).

  On the other hand, if the result of determination in step S110 is that there are a plurality of servers that can be handled (YES in S110), a server with the smallest predicted load is searched from among the plurality of servers that can be handled (S111 and S112). Specifically, the control point CP calculates an average value of predicted load data of each compatible server (S111). For example, when the servers SV2 and SV3 are compatible servers, the average value of the predicted load data of each of the servers SV2 and SV3 is obtained. After obtaining the average value of the predicted load data for all the servers that can be handled, the server that can handle the average value of the predicted load data is determined as the distribution server (S112).

  On the other hand, as a result of the determination in step S109, when the predicted load of the other servers (SV2 and SV3) excluding the server SV1 exceeds the reference load Lref, no matter which of the servers SV1 to SV3 delivers the music file MU11, the sound Cutting may occur. In this case, a server with the lowest possibility of sound interruption is determined as a distribution server by the following method (S114 to S117).

  First, the control point CP determines whether or not all the predicted load data of the servers SV1 to SV3 exceeds the reference load Lref. When all the predicted load data exceeds the reference load Lref (YES in S113), the control point CP calculates the average value of the predicted load data for each server (S116). Then, the server SV having the smallest calculated average value is determined as the distribution server (S117).

  On the other hand, as a result of the determination in step S113, if at least one of the servers SV1 to SV3 has a part of the predicted load data less than the reference load Lref (NO in S113), the approximate function composed of the predicted load data is the reference load. Crosses the Lref line. For example, as shown in FIG. 13, the predicted load data CPSV2 calculated based on the load data CSV2 of the server SV2 and the predicted load data CPSV3 calculated based on the load data CSV3 of the server SV3 are the reference load Lref line. Intersect. Hereinafter, steps S114 and S115 will be described with reference to FIG.

  The control point CP calculates times Δtsv2 and Δtsv3 when the predicted load data CPSV2 and CPSV3 are less than the reference load Lref (SV) (S114). Specifically, a difference value obtained by subtracting the reference load Lref (SV) from each predicted load data is obtained, and times Δtsv2 and Δtsv3 are obtained from the sum of times when the difference values are negative.

  After calculating the times Δtsv2 and Δtsv3, the control point CP determines the server having the maximum calculated time Δtsv as the distribution server. In FIG. 13, since the time Δtsv2 is longer than the time Δtsv3, the control point CP determines the server SV2 as a distribution server (S115).

  As described above, the control point CP determines the distribution server by predicting the load applied to each of the servers SV1 to SV3 when distributing the music file MU11. Therefore, it is possible to suppress the occurrence of sound interruption during reproduction.

[Operation of network AV system after distribution server is determined]
The control point CP that has determined the distribution server transmits a reproduction instruction to the renderer RE. When the distribution server is a server (SV2, SV3) other than the server SV1, the control point CP transmits a reproduction instruction to the renderer after the distribution server acquires the music file MU11.

[When distribution server is server SV1]
14 to 16, the control point CP first determines whether or not the distribution server is a server that stores the selected music file MU11 (S201). The control point CP refers to the content information database 142 to determine whether or not the distribution server stores the music file MU11.

  When the distribution server is the server SV1, the server SV1 has already accumulated the music file MU11 (YES in S201). In this case, the distribution server SV1 does not need to acquire the music file MU11 from the other server SV. Therefore, the control point CP transmits a reproduction instruction including the location (URI11) of the music file MU11 registered in the content information database 142 to the renderer RE (S205).

  When the renderer RE receives the reproduction instruction (YES in S501), the server that requests streaming distribution is determined as the server SV1 based on the URI 11 included in the reproduction instruction, and a streaming distribution request including the URI 11 is sent to the server SV1. Transmit (S502).

  When the server SV11 determines that the streaming distribution request has been received (YES in S301, NO in S302), the server SV11 decodes the music file MU11 and distributes it to the renderer RE (S306). At this time, the renderer RE receives and plays the music file MU11 that has been streamed (S503).

  When the server SV1 completes the streaming distribution (YES in S307), the music file MU11 determines that the music file MU11 has been stored (NO in S308), and sends a streaming distribution completion notification to the control point CP. (S310).

[When distribution server is server SV2 or SV3]
Next, the operation of the network AV system 1 when the distribution server is the server SV2 will be described. In step S201, the control point CP determines that the distribution server (server SV2) does not store the music file MU11 (NO in S201). If the distribution server (SV2) does not acquire the music file MU11 from the server SV1, it cannot stream the music file MU11 to the renderer RE. Therefore, the control point CP transmits an acquisition instruction including the location (URI11) of the music file MU11 to the server SV2 (S202).

  When the server SV2 receives the acquisition instruction (YES in S301 and S302), the server SV2 transmits a download distribution request for the music file MU11 to the server SV1 based on the URI 11 of the music file MU11 included in the acquisition instruction (S303). The server SV2 receives the music file MU11 downloaded from the server SV1 and stores it in the HDD 14 (S304). After completing the storage of the music file MU11, the server SV2 transmits the URI (URI 25) of the music file MU11 based on the storage location of the music file MU11 to the control point CP (S305).

  The control point CP receives the URI 25 of the music file MU11 transmitted from the server SV2, and stores it in the memory 12 (S203). Subsequently, the control point CP transmits a reproduction instruction including the URI 25 stored in the memory 12 to the renderer RE (S204).

  The renderer RE transmits a streaming distribution request including the URI 25 of the music file MU11 to the server SV2 based on the reproduction instruction (S502), and the server SV2 decodes and distributes the music file MU11 based on the URI 25 (S306). When the distribution is completed (S307), the server SV2 determines that the music file MU11 has been acquired from the other server SV1 (YES in S308). At this time, the server SV2 deletes the music file MU11 stored by itself (S309). This is because the music file MU11 is already stored in the server SV1, and therefore it is not necessary to store it further in the server SV2.

  After deleting the music file MU11, the server SV2 transmits a distribution completion notice to the control point CP (S310).

  Even when the distribution server is the server SV3, the same operation as that of the server SV2 is executed.

  As described above, the network AV system according to the present embodiment predicts the load during decoding and distribution of the server SV that stores the selected music file MU, and distributes the music file when the predicted load is small. Further, when the predicted load is large, the other server SV having a small load is made to proxy the decoding and distribution of the music file MU. Therefore, it is possible to suppress the sound interruption of the music file being played.

  In the network AV system according to the present embodiment, the content file is a music file, but it may be a video file or a video and audio file.

  In the present embodiment, the reference load Lref is constant without depending on the server SV. However, the reference load Lref may be different for each server. Further, the reference load Lref may be different for each music file. In this case, the reference load Lref is registered in advance corresponding to each record in the content information database 142, for example.

  In the present embodiment, in steps S111 and S112 in FIG. 9, the server having the smallest predicted load data average value is determined as the distribution server, but the distribution server may be determined by other methods. For example, the control point CP stores a server list in which the servers SV1 to SV3 on the network are listed in the HDD 14 in advance. In the case of YES in step S110, a server listed at the top of the server list among a plurality of servers having a predicted load lower than the reference load Lref may be set as the distribution server. In this case, it is not necessary to calculate the average value of predicted load data. In addition, among the plurality of servers SV, the server SV that is first found that the predicted load is less than the reference load may be determined as the distribution server.

  In steps S105 and S106 and steps S403 and S404, the distribution load is added to the load data calculated by the extrapolation method to obtain the prediction load. However, the prediction load may be calculated by another method. For example, only the most recent load data at time t1 may be read, and the distribution load information may be simply added to the load data to obtain the predicted load.

  The network speed in the inter-server network speed database is registered in advance by measuring the network speed between servers. The network speed between servers may be estimated using ping, and the estimated network speed may be registered in the network speed database between servers.

  While the embodiments of the present invention have been described above, the above-described embodiments are merely examples for carrying out the present invention. Therefore, the present invention is not limited to the above-described embodiment, and can be implemented by appropriately modifying the above-described embodiment without departing from the spirit thereof.

1 is a functional block diagram showing an overall configuration of a network AV system according to an embodiment of the present invention. It is a functional block diagram which shows the structure of the control point shown in FIG. It is a figure which shows the data structure of the load information database stored in the hard-disk drive in FIG. It is a figure which shows the data structure of the content information database stored in the hard-disk drive in FIG. It is a figure which shows the data structure of the distribution load database stored in the hard-disk drive in FIG. It is a functional block diagram which shows the structure of the server shown in FIG. It is a functional block diagram which shows the structure of the renderer shown in FIG. It is a figure for demonstrating the operation | movement outline | summary of the network AV system shown in FIG. It is a flowchart which shows the detail of the delivery server determination process which the control point in FIG. 1 performs. It is a figure which shows transition of the load data of the server for demonstrating step S105 and S106 in FIG. It is a flowchart which shows the detail of step S108 in FIG. It is a figure which shows the data structure of the network speed database between servers stored in the hard-disk drive in FIG. It is another figure which shows transition of the load data of a server different from FIG. It is a flowchart which shows the detail of operation | movement of the control point after performing the delivery server determination process shown in FIG. It is a flowchart which shows the detail of operation | movement of the server in FIG. It is a flowchart which shows the detail of operation | movement of the renderer in FIG.

Explanation of symbols

1 Network AV System 141 Load Information Database 142 Content Information Database 143 Distribution Load Database 241 Content Database CP Control Point RE Renderers SV1 to SV3 Server

Claims (8)

A network AV system comprising a plurality of servers connectable to each other, a renderer connectable to the plurality of servers, and a controller connectable to the plurality of servers and the renderer,
The server
Content storage means for storing a plurality of content files;
In response to a request from the renderer, a delivery means for decoding the selected content file and streaming it to the renderer;
Load information transmitting means for transmitting load information related to the load on itself to the controller;
The controller is
Content information related to details and location of content files stored in each server, load information transmitted from each server, and distribution load information related to load applied to each server by decoding and streaming the content file Storage means for storing;
Server specifying means for specifying a server for storing the selected content file based on the content information;
A predicted load calculating means for calculating a predicted load applied to the specified server when distributing the selected content file based on the load information of the specified server and the distribution load information of the specified server; ,
Replay instruction means for transmitting a replay instruction to the renderer to receive and play the distribution of the selected content file from the identified server when the calculated predicted load is less than a reference load;
The renderer is
Requesting means for requesting the server to perform streaming delivery of the selected content file based on the reproduction instruction;
A network AV system comprising playback means for playing back a distributed content file. The network AV system according to claim 1,
The predicted load calculation means further calculates a predicted load applied to each server when each server distributes the selected content file based on the load information of each server and the distribution load information of each server,
The controller further includes:
When the predicted load of the identified server exceeds the reference load, search means for searching for a server having a predicted load less than the reference load among the plurality of servers;
An acquisition instruction for transmitting an acquisition instruction to acquire the selected content file from the specified server to the searched server when the search unit searches for a server whose predicted load is less than a reference load. Means and
The server further includes:
An acquisition means for acquiring the selected content file from the specified server when receiving the acquisition instruction;
When the search means searches the server, the playback instruction means transmits a playback instruction to the renderer to receive and play the selected content file from the searched server. Network AV system. The network AV system according to claim 2,
The network AV system according to claim 1, wherein the search means further searches for a server having the smallest predicted load when there are a plurality of servers having the predicted load less than a reference load. The network AV system according to claim 1,
The load information is composed of time-series load data related to the load applied to the server,
The predicted load calculating means includes
Playback time calculating means for calculating the playback time of the selected content file;
Time-series data calculation means for predicting time-series load data from the current time until the reproduction of the selected content file is completed based on the load information and the reproduction time by an extrapolation method;
Adding means for adding the distribution load information to the predicted time-series load data to obtain a predicted load composed of time-series predicted load data;
The network AV system, wherein the reproduction instruction means transmits the reproduction instruction when all of the time-series predicted load data are less than the reference load. A plurality of content files are accumulated, a plurality of servers that decode and stream the content file in response to a request, and the server requests streaming delivery of the selected content file, and the streamed content file is played back The controller of the network AV system comprising a renderer and a controller connectable to the plurality of servers and the renderer,
Content information related to details and location of content files stored in each server, load information transmitted from each server that is information related to the load on each server, and decoding and streaming distribution of the content file Storage means for storing distribution load information regarding the load on each server;
Server specifying means for specifying a server for storing the selected content file based on the content information;
A predicted load calculating means for calculating a predicted load applied to the specified server when distributing the selected content file based on the load information of the specified server and the distribution load information of the specified server; ,
When the calculated predicted load is less than a reference load, a playback instruction unit is provided that transmits a playback instruction to the renderer so that the selected content file is distributed from the specified server and played back. Controller. The controller according to claim 6, comprising:
The predicted load calculation means further calculates a predicted load applied to each server when each server distributes the selected content file based on the load information of each server and the distribution load information of each server,
The controller further includes:
When the predicted load of the identified server exceeds the reference load, search means for searching for a server having a predicted load less than the reference load among the plurality of servers;
An acquisition instruction for transmitting an acquisition instruction to acquire the selected content file from the specified server to the searched server when the search unit searches for a server whose predicted load is less than a reference load. Means and
When the search means searches the server, the playback instruction means transmits a playback instruction to the renderer to receive and play the selected content file from the searched server. controller. The controller according to claim 5, wherein
The controller is further configured to search for a server having the smallest predicted load when there are a plurality of servers having the predicted load less than a reference load. The controller program for making a computer implement | achieve the means of any one of Claims 5-7.

Images