EP2436168A2

EP2436168A2 - Technique for distributing content to a user

Info

Publication number: EP2436168A2
Application number: EP10728775A
Authority: EP
Inventors: Emile Stephan; Joël LATTMANN
Original assignee: France Telecom SA
Current assignee: Orange SA
Priority date: 2009-05-29
Filing date: 2010-05-18
Publication date: 2012-04-04
Also published as: WO2010136699A3; WO2010136699A2; US20120072604A1

Abstract

The invention relates to a technique for distributing content (C1) to a user (U1, U2), wherein the user transmits to a first content-distribution device (S2), referred to as a secondary device, at least one quality indicator relating to the distribution of a part of said content by a second content-distribution device (S1), referred to as a primary device. When the secondary device detects, from at least the quality indicators relating to at least one content distribution by the primary device, an event capable of affecting the distribution of the content by the primary device to the user, the secondary device orders the user to make a transfer from the primary device to the device so that the secondary device can carry out a distribution of parts of said content that were not distributed by the primary device.

Description

Technique of distributing content to a user

The invention relates to a technique for distributing content to a user from content distribution devices. The invention lies in the fields of telecommunications and more particularly in the field of content distribution.

There are techniques allowing a user of a terminal to obtain content after a content distribution server In order to guarantee the availability of the content, it is possible to obtain it from a plurality of content servers. distribution of content in a communication network or of peers having at least part of the content in a P2P network, for

"Peer-To-Peer" The terminal simultaneously obtains the data from several servers or peers The possible redundancy introduced by these servers or by these peers makes it possible to increase the effective availability of the content It is for example to vary the number of servers or peers involved in the distribution of the content in order to provide a given quality This given quality is sought after but is not guaranteed Furthermore, a measure of the quality, network or as perceived by the user is complex to achieve, a plurality of servers being involved in the distribution of content In addition, a decision is made only in response to low quality measurements The user then feels this decline in quality and is not satisfied

One of the aims of the invention is to remedy the shortcomings / drawbacks of the state of the art and / or to make improvements thereto.

According to a first aspect, the subject of the invention is a method of distributing content to a user, comprising the following steps implemented by a first content distribution device, said secondary device

a step of obtaining from the user at least one quality indicator relating to a distribution of a portion of said content to said user by a second content distribution device, said primary device,

a step of detecting an event that may affect the distribution of the content by the primary device to the user, based on at least one of the quality indicators relating to at least one content distribution by said primary device; a transfer control step of the primary device to the secondary device for the user, for distribution by the secondary device of parts of said content not distributed by the primary device

Thus, the content distribution process makes it possible to detect a drop in quality based on measurements made by at least one user to whom a content is being broadcast from the primary device. This makes it possible to anticipate actions to be taken for guarantee quality of reception to the user The content distribution devices can be content distribution servers They can also be peers in a peer-to-peer network or P2P The user's terminal can be a peer of the P2P network in this case, it can play both roles, terminal and server

The method is remarkable in that the secondary device is a receiver of these measurements and can therefore take over from the primary device for distributing the content

Indeed, in a conventional manner, the device or the peer distributing a content is a receiver of the measurements made by the user's terminal Thanks to the content distribution method according to the invention, bad feelings of the users related to a degradation of the quality are thus avoided, the content distribution method anticipating a future problem and proposing a solution instead of reacting to the problem Thus, it is not necessary in such a case to introduce redundancy by providing a plurality of devices in order to guarantee that the desired quality will be achieved Thus limiting the redundancies required in the communication network Moreover, the processor load of the primary device distributing the content is reduced since the latter is not in charge of the collection of measurements, nor of their analysis II is no longer both judge and party in securing the content distribution process secondary system has available processor resources to perform these tasks The path taken in the network between the user and the secondary device is a priori different from that between the user and the primary device, once the transfer of the session of the primary device to the secondary device performed, the quality of content distribution can only improve In addition, the supervision of the quality of the distribution of content is less complex to implement in such an architecture than in a network of communication where a plurality of servers or peers are involved in the distribution of content Supervision of the quality of the distribution of content is then deterministic and controlled by the network In a network involving a plurality of servers or peers, the supervision then the change of server or peer decision are usually made by the terminal itself L Server and peer changes by a set of terminals can then cause load and link load oscillations, which can prevent provision of the content delivery service while the system load is not yet available. reached

The distribution method is particularly suitable for implementation in a communication network of the CDN type, for "Content Delivery Network"

According to a particular characteristic, the event belongs to the group comprising a decrease in quality affecting the distribution of content by the primary device, a decrease in quality affecting the distribution of content by the primary device to the user, a cut of exchanges with the primary device, an atypia in synthetic data determined for the primary device compared to other content distribution devices Thus, a transfer decision can be made on different criteria, determined either for the user concerned or for a set of users in relation to the primary device.

According to another particular characteristic, the method further comprises a step of obtaining the content and a step of distributing another content prior to the content distribution step.

The secondary device is able to implement steps preliminary to the distribution of the content before or parallel to the sending of the transfer command. In particular, it may have been selected because of a current distribution of this content to other users. Once the user's session with the primary device has been transferred, he can immediately take over the distribution of the content, taking over the distribution from parts not distributed by the primary device. The user then feels in the worst case a short interruption in the distribution of content. It is also possible for the secondary device to insert another content, as the waiting content. According to yet another particular characteristic, the transfer control comprises a notification of said event to the user.

Thus, the information provided to the user is improved.

Correlatively, the subject of the invention is also a method for obtaining content by a user, comprising the following steps implemented by said user: a step of discovering a first content distribution device, said device primary, and a second device for distributing contents, said secondary device, able to distribute said content;

a step of receiving at least a portion of the content originating from the primary device;

a step of transmitting to the secondary device at least one quality indicator relating to the distribution of a portion of said content to said user by the primary device;

a step of receiving a transfer command from the primary device to the secondary device, with a view to distribution by the secondary device of portions of said content not distributed by the primary device. According to a second aspect, the invention relates to a device for distributing content to a user, comprising:

means for obtaining from the user at least one quality indicator relating to a distribution of a portion of said content to said user by another content distribution device, said primary device; means for detecting an event likely to affect the distribution of the content by the primary device to the user on the basis of quality indicators relating to at least one distribution of contents by the primary device,

means for controlling the transfer of the user from the primary device to said device, with a view to a distribution by said device of parts of said content not distributed by the primary device

According to a third aspect, the invention relates to a device for obtaining a content comprising

means for discovering a first content distribution device, said primary device, and a second content distribution device, said secondary device, capable of distributing said content,

means for receiving at least part of the content coming from the primary device,

means for transmitting, to the secondary device, at least one quality indicator relating to the distribution of a portion of said content to said user by the primary device,

means for receiving a transfer command from the primary device to the secondary device, for distribution by the secondary device of parts of said content not distributed by the primary device

According to a fourth aspect, the invention relates to a content distribution system comprising content distribution devices and at least one device for obtaining content as described above.

According to a fifth aspect, the invention relates to a computer program comprising instructions for implementing the method of distributing content as described previously by a content distribution device, when this program is executed by a processor.

According to a sixth aspect, the invention relates to a computer program comprising instructions for implementing the method of obtaining content as described previously by a device for obtaining content, when this program is The invention will be better understood with the aid of the following description of a particular embodiment of the methods of the invention, with reference to the accompanying drawings in which FIG. 1 represents a system for distributing contents in its environment according to a particular embodiment of the invention, FIG. 2 represents a simplified diagram of the steps of the methods implemented for the distribution of contents according to a particular embodiment of the invention, FIG. 3 represents a content distribution device according to a particular embodiment of the invention; FIG. 4 represents a device for obtaining a content according to a particular embodiment of the invention; FIG. an example of content received by a terminal according to a particular embodiment of the invention

A content distribution system 1 in its environment will be described with reference to FIG. 1 according to a particular embodiment.

The content distribution system 1 comprises a plurality of content distribution servers, of which two of them S 1, S 2 are represented in FIG. 1, intended for a plurality of users U1, U2. the same reference refers to both the terminal and the user of this terminal The contents to be distributed are distributed on the servers Sl,

S2 content distribution and content servers S3, S4, S5 The distribution servers S1, S2 and the content servers S3, S4, S5 form part of a content distribution network CDN or "Content Dehvery Network" in English A CDN network is made up of servers connected in a network through an Internet-type communication network or any other type of network, for example a mobile network. These servers cooperate in order to make available to users multimedia content or data. servers cooperate with each other to satisfy queries made by users for access to content or data and distribute content or data back to users Article "A taxonomy and survey of Content Dehvery Networks" Pathan et al., Technical Report, GRIDS-TR-2007-4,

Gπd Computing and Distnbuted Systems Laboratory, The University of Melbourne, Austraha 12

February, 2007, presents a state of the art of this type of network It is understood that only part of the CDN network used in the context of the description of this embodiment is shown in FIG.

FIG. 1 shows indicatively contents C1, C2, C3 distributed over these different servers S1-S5. The contents may be of different types. These may be contents broadcast in real time, multimedia contents that may be obtained on demand, multimedia content likely to be broadcast in streaming or streaming mode in English, news content in RSS format, multimedia books Subsequently, content is divided into blocks elementary content can be distributed to different servers A block number of the content can identify a given block of content An index block "Index Cl" provides information on the location of data blocks on different servers For example for the content Cl, the index block indicates that the blocks of data bl a b20 are available on the servers S1 and S5, the blocks b21 to b50 on the server S3, blocks b51 to b360 on servers S3 and S5, blocks b361 to b480 on server S5, blocks b400 to b480 on servers S1 and S4

The content distribution server S 1 comprises the blocks b1 to b20 and the blocks b400 to t> 480 of the content C1. The content server S3 comprises the blocks b21 to b50 and the blocks b51 to b360 of the content C1, the content C2. content server S4 comprises blocks b400 to b480 of content C1 and content C3 content server S5 comprises blocks b1 to b20, blocks b51 to b360, blocks b361 to b480 of content C1.

It is emphasized here that the distribution of contents, data blocks is given only as an example illustrating the implementation of the methods of selection, distribution of content and obtaining content

FIG. 1 also shows a catalog server P, offering users Ul, U2 different contents Cl, C3 in a catalog C

The various steps of the methods implementing the invention, that is to say the selection process implemented by a catalog server, that of distribution of content implemented by a content distribution server and that of to obtain a content by a terminal according to a particular embodiment of the invention will now be described in connection with Figure 2 It should be noted that not all steps are shown in Figure 2 so as not to make it more complex

In a step Fl of the content obtaining method, the user U1 requires a distribution of the content C1 to the catalog server P by a message Req (Cl) The request Req (Cl) is received by the catalog server P in a step El of the selection process Still in this step E1, the catalog server P determines the servers capable of distributing the content C1. By server capable of distributing the content, we mean a server having good conditions of availability, of charging, or else located near the server. terminal requiring the distribution of the content In a step E2 of the selection method, the catalog server P transmits to the user U1 a response Resp (Cl, Sl, S2) including the address of the server Sl, said primary server, and an address contentUrl of the Cl content in the communication network or URL for "Uniform Resource

Locator "The response Resp (Cl, Sl, S2) also comprises a command for the transmission by the terminal Ul to the server S2, said secondary server, quality indicators relating to the distribution of the content Cl by the primary server Sl It should be noted that the server S2 is part of the content distribution servers able to distribute the content C1 For example, the server S2 may be in the process of distributing the content C1 to another user

The next case is where the quality indicator transmission control comprises a program interpretable by the user's terminal U1, for example in the form of AJAX code, for "Asynchronous JavaScript and XML", comprising software instructions for transmitting to the server S2 measurements or quality indicators relating to the distribution of the content C l to the user Ul by the primary server S l

In a step F2 of the method for obtaining the content C1, the user Ul obtains the response Resp (C, S1, S2) from the catalog server P The user Ul thus discovers two servers S1, S2, able to distribute the contents

In a step G1 of the distribution method, the primary server S 1 obtains the content C1 in the network CDN by consulting the index block of the content C1 and then from the content servers. This step G1 is not more precisely detailed, given that it implements conventional mechanisms of the CDN-type networks Then, the user Ul obtains sets of data blocks of the content C1, this obtaining phase being symbolized in FIG. 2 by a double arrow between the user Ul and the primary server S 1

In a step F3 of the obtaining method, the user Ul determines quality indicators, also called performance indicators, relating to the distribution of a set of data blocks of the content C1. By quality indicator, we mean a measurement performed by the user Ul following the delivery of a set of blocks II may be a measured quality or a quality felt by the user Ul Such quality indicators are for example a number of lost packets , a jitter measurement, a delay, an average opinion score (MOS) for audio or video content, a change in the rate of the link between the primary server S l and the user Ul, an indicator representative of the processor load of the terminal Ul, an indicator representative of a rate of availability of a memory of the terminal UI. Such quality indicators are for example collected for an RTP connection, for "Real-time Transport", and transmitted in messages. RTCP XR, for "RTP Control Protocol Extended Report" They can also be collected for a "streaming" session and transmitted in multimedia session control messages, for example RTSP, for "Real Time Streaming Protocol" In this case, the indicators are usually written in XML blocks in the format "W3C Extended log format"

Still in this step F3, the user U1 sends to the secondary server S2, as indicated in the response Resp (C1, S1, S2), a Meas-Rep measurement report (Cl, S1) comprising an identifier session report, a content identifier C l, an identifier of the primary server S l, a current block number and the determined quality indicators This measurement report can be transmitted using a hypertext message protocol such as http , for "HyperText Transfer Protocol", for example a GET message The session identifier makes it possible to group the different measurement reports relating to a single session The Meas-Rep message (C l, S l) is received by the secondary server S2 in a step G2 It should be noted that the recipient of the measurement reports server is not the server in charge of the distribution of content C l to the user Ul The secondary server S2 is available to receive and process the measurement reports The supervision of the distribution of the content is thus particularly simple to set up

In the embodiment described here, the secondary server S2 aggregates in this step G2 the quality indicators received with quality indicators relating to the distribution of the content.

C1 to the user U l and to quality indicators relating to content distributions by the primary server S 1 As represented in FIG. 2, the secondary server S2 also receives quality indicators Meas-Rep (C3, Sl ) relating to the distribution of the content C3 by the primary server Sl to the user U2 It should be noted that the secondary server S2 also receives session quality indicators between users and other servers than the primary server

S l The secondary server S2 is thus an element of a distributed monitoring system where each server distributes content to clients and monitors part of the sessions of each of the other servers. The secondary server S2 thus has data for determining metadata or summary data representative of the activity of the content distribution service This metadata is for example the number of servers involved in the provision of the service, a value representative of the activity of the service determined from a variation of the number of sessions over time, the relative load of the servers relative to each other, a value representative of a general operation of the service, a value representative of an operation of each server, a general QoS value of the network determined from the rates packet loss, delay and jitter, a QoE value provided by each server from packet loss rates, delays and jitter

A single transmission of a measurement report is shown in FIG. 2 in order not to complicate FIG. 2 It is understood that these measurement reports are transmitted regularly by the terminal U1 and the secondary server S2 repeats the step G2. upon receipt of new measurement reports

From these aggregated quality indicators at different levels, at a user level or at a primary server level, the secondary server S2 detects an event in a step G3 This event can be a decrease in quality affecting the primary server S 1, a decrease in quality affecting the distribution of the content C l to the user Ul, a cut of exchanges with the primary server S l, an atypia in the metadata describing the operation of the primary server S l compared to the other servers As an example of atypia, we can cite a jitter value ten times greater on the session between the user Ul and the primary server S 1 than on the other sessions. Such a jitter value reflects an overload of the primary server. S 1 In this case, the data relating to the sessions spend on average more time in the queues, which makes very variable the delay between the application processing of the data e t their broadcast on the network This list is of course not exhaustive In the case where the event is the detection of a decrease in quality affecting the distribution of the content Cl to the user Ul or the detection of the interruption of the exchanges of the user Ul with the primary server Ul, the secondary server S2 transmits in step G3 to the user U1 a command SW (C1, S1, S2) for transferring his session or his context from the primary server S 1 to the secondary server S2

In the case where the event is the detection of a decrease in quality affecting the primary server Sl, the secondary server S2 transmits in the step G3 to all the users

U1, U2 having transmitted quality indicators relating to the distribution of content by the primary server S l a SW (Cx, Sl, S2) transfer of their respective sessions or contexts from the primary server Sl to the secondary server S2

In the case where the event is the detection of an atypia in the metadata describing the operation of the primary server S 1 relative to the other servers, the secondary server S2 transmits in the step G3 is a command to one of the users, ie a command intended for a group of users, in order to unload the primary server Sl More precisely, for these different cases, the transfer command SW (C1, S1, S2) is a command for redirecting the session or the context that the user Ul established with the primary server S 1 for the distribution of the content C 1 to the secondary server S 2

At the same time, if the secondary server S2 does not have the content C1 available, it obtains this content C1 in a step G4 of the content distribution method similarly to the step G1 of the same method

The transfer command is received by the user U1 in a step F4 of the obtaining method The processing to be performed by the terminal U1 upon receipt of this transfer command may also be included in the AJAX code received during the step F2. of the obtaining method Still in this step F4, the user Ul transfers the session established with the primary server Sl for the distribution of the content C1 to the secondary server S2

The distribution of the content Cl may resume from the secondary server U2 to the user U1 from the current block number contained in the last measurement report transmitted. The distribution of the content C1 may restart immediately or after the broadcasting of a content.

C2 waiting Thus, it is possible to anticipate the occurrence of a degradation of the quality as felt by the user and to switch the session to a secondary server, itself ready to take over The user does not feel the degradation of quality and gets the desired content from possibly two servers At a given moment, only two servers are involved in the distribution of content to a user and the supervision thereof, thus limiting the number of servers The secondary server S2 has processor resources to monitor the quality of the content distribution Cl We also use reports from measurement transmitted by other users to detect a general problem that could affect the primary server S 1

The description has been made in the case of a CDN-type network. The method is also applicable to any Internet-type communication network. More generally, the primary servers Sl and secondary server S2 are content distribution devices. The terminal Ul can, in some cases, play the role of content distribution server

For example, the terminal Ul can be a peer in a "peer-to-peer" network or P2P II plays in this case both the role of the terminal and that of the content distribution server

Still as an example, the primary content distribution device Sl is a server, the secondary content distribution device S2 is a peer of the P2P network

The presence of the catalog server P is optional content catalogs, for example of the "Internet Media Guide" are accessible in many ways They can be distributed online along with other content

In a particular embodiment, the messages are exchanged in the form of "Web services" requests, as specified by the W3C, for "World Wide Web Consortium", in the WSDL specification. the form of calling functions or procedures coded in XML In a particular embodiment, the user Ul discovers the two servers S l,

S2, capable of distributing content, by a network address assignment service such as DHCP, for "Dynamic Host Configuration Protocol" or by a URL resolving service, such as DNS, for "Domain Name System" In this particular embodiment, the catalog server P is not present In a particular embodiment, the executable program can be installed at initialization or downloaded to the terminal Ul In this case, the response Resp (Cl, S 1, S2) comprises an identifier of the content C1, an identifier of the content distribution server S1, as primary server, and an identifier of the content distribution server S2, as a secondary server. It is also possible to provide in the transfer control an identifier of a third content distribution server, not shown in FIG. 1, to which the measurement reports are addressed once the transfer has been made to the secondary server S2, this latter deny becoming in turn primary server

The transfer control may furthermore comprise a notification of the detected event intended for the user Ul. This makes it possible to inform the user of the current transfer and thus to improve his information. In a variant of the embodiment described above, the server S2 stores in step G2 aggregate quality indicators that it has determined in a database, not shown in Figure 1. This database is used by the catalog server P to determine in step El of the method for selecting the content distribution servers S 1 -S 2 capable of distributing the content Cl. It can thus select content distribution servers able to distribute the content Cl according to quality indicators observed in prior content distributions.

The content C1 as received by the terminal Ul is shown schematically in FIG.

5. Blocks b360 to b360 were received from the primary server S1 and then the transfer took place. The terminal Ul then received from the secondary server S2 the content C2, as standby content, then the distribution of the content C1 resumed from the data block b361 to the data block b480. The waiting content C2 is of course optional.

We will now describe a content distribution device 200 according to a particular embodiment in connection with FIG. 3. The content distribution device 200 comprises:

a module 202 for delivering content to users;

a module 206 for obtaining from a user at least one quality indicator relating to a distribution of a portion of said content to said user by another content distribution device, said primary device; a module 204 for detecting an event that may affect the distribution of the content by the primary device to the user on the basis of quality indicators relating to at least one distribution of contents by the primary device;

a user transfer control module 208 from the primary device to said device, for distribution by said device of portions of said content not distributed by the primary device.

The module 206 is further arranged to aggregate the received quality indicators with quality indicators relating to the distribution of content to the user and quality indicators relating to content distributions by the primary device.

The content distribution device 200 is intended to be integrated in content distribution servers S 1, S 2.

In some of the embodiments described above, especially when the user's terminal Ul plays a role of content distribution server, the content distribution device 200 can be integrated into a terminal Ul.

We will now describe a content obtaining device 300 according to a particular embodiment in connection with FIG. 4.

Such a device 300 for obtaining a content comprises: a module 304 for discovering a first content distribution device, called a primary device, and a second content distribution device, called a secondary device, capable of distributing said content;

a module 306 for receiving at least part of the content coming from the primary device;

a module 308 for determining at least one quality indicator relating to the distribution of a portion of said content to said user by the primary device;

a transmitting and receiving module 302, arranged to transmit at least one determined quality indicator and to receive a transfer command from the primary device to the secondary device, for distribution by the secondary device of parts of said content not distributed by the primary device.

In one embodiment, the module 304 is arranged to require a distribution of the content C1 with a catalog server P and to obtain in return a response comprising the address of the primary device Sl, a contentUrl address of the content C l in the network communication, a request for the transmission to the secondary device S2 quality indicators relating to the distribution of the content C1 by the primary device S l.

The device 300 for obtaining a content is intended to be integrated in the user's terminal U1.

The modules 202, 204, 206, 208 of the content distribution device 200 are arranged to implement the steps of the previously described content distribution method executed by the content distribution device. These are preferably software modules comprising software instructions for executing those of the steps of the distribution method described above, implemented by a content distribution device. The invention therefore also relates to: a program for a content distribution device, comprising program instructions intended to control the execution of the steps of the previously described distribution method which are executed by said device, when said program is executed by a processor of it;

a recording medium readable by a content distribution device on which the program for a content distribution device is recorded.

The modules 302, 304, 306, 308 of the content obtaining device 300 are arranged to implement the steps of the previously described content obtaining method performed by the content obtaining device. These are preferably software modules comprising software instructions for executing those of the steps of the method of obtaining described above, implemented by a device. The invention therefore also relates to: a program for a device for obtaining a content, comprising program instructions intended to control the execution of the steps of the previously described content obtaining method which are executed by said device, when said program is executed by a processor of it; - A recording medium readable by a device for obtaining a content on which is recorded the program for content obtaining device.

The software modules can be stored in or transmitted by a data carrier. This may be a hardware storage medium, for example a CD-ROM, a magnetic diskette or a hard disk, or a transmission medium such as an electrical signal, optical or radio, or a telecommunications network.

The invention also relates to a content distribution system 1 comprising content distribution devices and at least one device for obtaining content, as described above.

Claims

A method for distributing a content (C1) to a user (U1, U2), comprising the following steps implemented by a first device (S2) for distributing content, said secondary device

a step of obtaining (G2) from the user at least one quality indicator relating to a distribution of a portion of said content to said user by a second device (S l) for distribution of content, said primary device,

a step of detecting (G3) an event likely to affect the distribution of the content by the primary device to the user, based on at least quality indicators relating to at least one distribution of contents by said device primary

a control step (G3) for transferring the primary device to the secondary device to the user, for distribution by the secondary device of portions of said content not distributed by the primary device;

The distribution method according to claim 1, wherein the event belongs to the group comprising a decrease in quality affecting the distribution of content by the primary device, a quality decrease affecting the distribution of content by the primary device to the user, a cut of exchanges with the primary device, an atypia in data of synthesis determined for the primary device compared to those of other devices of distribution of contents

The method of distribution of claim 1, further comprising a step of obtaining the content (G1, G4) and a step of distributing another content prior to the content distribution step

The method of distribution of claim 1, wherein the transfer control includes a notification of said event to the user.

Method for obtaining a content (C1) by a user (U1, U2), comprising the following steps implemented by said user

a step of discovering (F2) a first content distribution device (S 1), said primary device, and a second content distribution device (S 2), called a secondary device, capable of distributing said content, a step of receiving at least a portion of the content originating from the primary device, - A transmission step (F3) to the secondary device of at least one quality indicator relating to the distribution of a portion of said content to said user by the primary device;

a step of receiving (F4) a transfer command from the primary device to the secondary device, with a view to a distribution by the secondary device of parts of said content not distributed by the primary device.

6. Device (S1, S2, 200) for distributing content to a user, comprising:

means (206) for obtaining from the user at least one quality indicator relating to a distribution of a portion of said content to said user by another content distribution device, said primary device;

means (204) for detecting an event likely to affect the distribution of the content by the primary device to the user on the basis of quality indicators relating to at least one distribution of contents by the primary device; means (208) for controlling the transfer of the user from the primary device to said device, with a view to a distribution by said device of portions of said content not distributed by the primary device.

7. Device (300) for obtaining a content, comprising: - discovery means (304) of a first device (S1) for distribution of contents, said primary device, and a second device (S2) content distribution device, said secondary device, able to distribute said content;

means (306) for receiving at least a portion of the content from the primary device; - transmission means (302) for the secondary device of at least one quality indicator relating to the distribution of a portion of said content to said user by the primary device;

means for receiving (302) a transfer command from the primary device to the secondary device, for distribution by the secondary device of portions of said content not distributed by the primary device.

A content distribution system (1) comprising content distribution devices according to claim 6 and at least one content obtaining device according to claim 7.

9. Computer program comprising instructions for implementing the method of distributing content according to claim 1 by a content distribution device, when this program is executed by a processor.

10. Computer program comprising instructions for implementing the method of distributing content according to claim 5 by a device for obtaining a content, when the program is executed by a processor.