KR100872176B1 - Method and system for creating data transmission path for providing n:n communication service - Google Patents

Abstract

The present invention provides a many-to-many communication service in a many-to-many communication environment including a plurality of edge domains including a plurality of mobile agents for forwarding multicast data, a core domain connecting the edge domains, and a session manager. A method and system for configuring a transmission path for multicast data, wherein the session manager establishes a session for a many-to-many communication service in a session manager, wherein users of the many-to-many communication service subscribe to the formed session. And control the withdrawal and manage information related to the many-to-many communication service, wherein the plurality of agents perform subscription / leave of the formed session using information received from the session manager, respectively, to relay the data. Configure the data transmission path, and By monitoring and repairing errors in edge and core domains, it is possible to provide many-to-many communication services that can minimize changes to current Internet devices.

Many-to-many communication service, multicast, data transmission path, core domain, edge domain, core multicast agent (CMA), edge multicast agent (EMA), thin line manager (SM) .

Description

METHOOD AND SYSTEM FOR CREATING DATA TRANSMISSION PATH FOR PROVIDING N: N COMMUNICATION SERVICE}

1 is a diagram illustrating a network configuration for providing a many-to-many communication service according to an embodiment of the present invention;

2 is a diagram illustrating a many-to-many communication service operation in a session manager (SM) according to an embodiment of the present invention.

3 is a diagram illustrating a procedure for monitoring a many-to-many communication service according to an embodiment of the present invention;

4 illustrates an operation for a many-to-many communication service in a core multicast agent (CMA) in accordance with an embodiment of the present invention;

5A illustrates a subscription request and approval procedure for a CMA's many-to-many communication service through an SM according to an embodiment of the present invention;

5B illustrates a procedure for inserting a many-to-many communication topology of a CMA according to an embodiment of the present invention;

5C is a diagram illustrating a heartbeat (HB) procedure for ensuring robustness of a CMA according to an embodiment of the present invention;

5d illustrates a communication service withdrawal procedure of a CMA according to an embodiment of the present invention;

5E is a diagram illustrating a termination procedure of a many-to-many communication service by an SM according to an embodiment of the present invention;

6 illustrates edge multicast agent (EMA) operation for a many-to-many communication service according to an embodiment of the present invention;

7A is a diagram illustrating a multi-to-many communication service subscription application procedure of an EMA according to an embodiment of the present invention;

7B illustrates a procedure for joining a many-to-many communication service group of an EMA according to an embodiment of the present invention;

7C is a diagram illustrating a procedure for leaving a many-to-many communication service group of an EMA according to an embodiment of the present invention;

8A illustrates a partitioning problem of a core domain according to an embodiment of the present invention;

8B is a diagram illustrating a partitioning problem of an edge domain according to an embodiment of the present invention;

8C is a diagram illustrating a narrowing ring problem of a core domain according to an embodiment of the present invention;

8D illustrates a loop solving problem in the edge domain according to an embodiment of the present invention;

9A illustrates a method of detecting a reduced ring in a core domain according to an embodiment of the present invention;

9B illustrates a reduction ring resolution method in a core domain according to an embodiment of the present invention.

The present invention relates to a relay transmission multicast technology for providing a many-to-many communication service in an environment in which a plurality of senders and a plurality of receivers are mixed, and particularly, a method and system for configuring a data transmission path for providing a many-to-many communication service. It is about.

The following methods have been used to provide many-to-many communication services.

The most popular method is to configure high performance servers and many members to provide many-to-many communication services. In this method, prior to performing the many-to-many communication service, a plurality of communication service participants connect to a device such as a proxy or a network server. The server then copies the data sent by one member and provides it to multiple recipients. However, this method encounters a limit on the number of concurrent users due to the resource limitation of the server device.

Another approach is to use multiple IP and many-to-many communication services using pure IP multicast, which is a significant cost to upgrade all Internet routers to multicast routers to provide communication environment services. There is a problem that the burden occurs.

Accordingly, an object of the present invention is to provide a method and system for configuring a data transmission path for providing a many-to-many communication service in a unicast-based current Internet environment.

Another object of the present invention is to provide a method and system for configuring a data transmission path for providing an efficient many-to-many group communication service by applying an overlay multicast scheme without changing the current Internet router apparatus.

The method for achieving the above object of the present invention is a session for configuring a many-to-many communication Internet environment formed of a plurality of edge domains including a plurality of agents for forwarding multicast data and a core domain connecting the edge domains. A method of configuring a data transmission path for providing a many-to-many communication service in a manager, the method comprising: forming a new session according to a session establishment request from the outside; Subscribing to the created new session; Periodically checking the status of agents in the new session to monitor session services; And terminating the new session according to a preset time or a request of an administrator.

Another method for achieving the above objects of the present invention is a many-to-many consisting of a plurality of edge domains including a plurality of multicast agents for forwarding multicast data, a core domain connecting the edge domains, and a session manager. In a communication providing system, a method of configuring a data transmission path for providing a many-to-many communication service by a core multicast agent existing in the core domain among the plurality of multicast agents, wherein the session manager attempts to join a newly formed session. process; Subscribing to the session using the information of the multicast agents obtained at the session join attempt; Continuously identifying information about other core multicast agents present in the session; And informing the session manager and the other core multicast agents of the withdrawal when leaving the session.

Another method for achieving the above object of the present invention is a many-to-many consisting of a plurality of edge domains including a plurality of multicast agents for forwarding multicast data, a core domain and a session manager connecting the edge domains. In a communication service providing system, any edge multicast agent existing in the edge domain among the plurality of multicast agents is a data transmission path configuration method for providing a many-to-many communication service, comprising: subscribing to a newly formed session in the session manager. The process of trying; Selecting an optimal core multicast agent existing in the edge domain among the multicast agents by using information received from the session manager when attempting to join the session; Joining the session group by connecting to the selected core multicast agent; And informing the session manager and the other edge multicast agents of the withdrawal when leaving the group of sessions.

On the other hand, the data transmission path configuration system for achieving the objects of the present invention, to establish a session for a many-to-many communication service, to control the joining and withdrawal to the formed session, and to receive information related to the many-to-many communication service Managing session manager; A plurality of edge multicast agents connected in a tree form in a plurality of edge domains to subscribe to the formed session using information received from the session manager, and informing the session manager that the session manager has left the session; And a plurality of core multicast agents connecting the edge domains in a ring form in a core domain, joining the formed session using information received from the session manager, and informing the session manager that the session manager has left when the session is left. It consists of consisting of a data transmission path for providing a many-to-many communication service.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In describing the present invention, if it is determined that detailed descriptions of related known functions or configurations may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

 In an embodiment of the present invention, a network for providing a many-to-many communication service and a many-to-many group communication service is configured as shown in FIG. 1 attached below, where a "node" described in the embodiments of the present invention is multiplied. Meaning a cast agent (hereinafter referred to as a MA), the node and the MA may be mixedly described for convenience of description, but they should be noted that they are the same.

In a communication environment according to an embodiment of the present invention, a plurality of nodes to simultaneously transmit and receive exist in a local multicast network, and each local multicast network is connected to each other by a unicast network. Therefore, multicast traffic in one local multicast network is not delivered to another local multicast network.

Referring to FIG. 1, a network for the many-to-many communication service is divided into a core domain (CD: Codre-Domain) 10 and an edge domain (Edge-domain) 20, and within the core domain 10. Plural core multicast agents (hereinafter referred to as CMAs) 11 are formed, and in the edge domain 20 plural edge multicast agents (hereinafter referred to as EMAs) 21 are formed for each core MA ( 11) are formed in a tree shape. Here, the edge domain 10 refers to the local multicast network, and the CMA 11 refers to a node connecting the localcast network to each other through the unicast network. The network configured by the CMAs 11 is the core domain, and the EMA means nodes located in the edge domain 20 except for the CMA.

The network for the many-to-many communication service may include a session manager (SM) for configuring and managing a many-to-many communication environment. The operation of the session manager will be described with reference to the accompanying drawings.

2 is a diagram illustrating a many-to-many communication service operation in a session manager according to an embodiment of the present invention.

Referring to FIG. 2, in step 101, the session manager SM starts a many-to-many group communication service. Thereafter, in step 102, if a new session establishment request is received from the CP or the manager, and a new session establishment request is received, a new session is formed in step 103. The session establishment request may have a form of session profile information, and the like. The session profile may be used for a specific group service session such as a name of a session to be communicated, characteristics of media used (codec, transmission rate, etc.), and a group address to be used. Means a set of information needed to define.

Then, in step 104, the session manager helps each node (Node A, Node B, Node C) to join the session, and based on node-specific authentication information to control subscription of each specific node. Check if you can join the communication service session. If it is impossible to join the session as a result of the test, the process proceeds to step 108 and ends the session. On the other hand, if the session subscription is granted, in step 105, the session manager notifies the node of the permission of the session subscription and simultaneously provides the session profile as information for subscribing to the session service.

Thereafter, in step 106, the session manager periodically checks the state of the session members, that is, monitors many-to-many communication services in order to increase the reliability of the session service. In this case, the reliability of the session service means how much quality of service a service provider can provide to a service user. Such service monitoring enables the session manager to monitor members constituting the session, that is, the MA, where the monitoring target includes not only whether the members operate but also the communication qualities of the members. Here, the parameters used to measure the communication quality take into account bandwidth, delay, delay jitter, packet loss rate, received bits, and the like.

The operation of the session manager for service monitoring is as shown in FIG. 3, and when the session service monitoring is required for a specific node or the whole, the session manager 30 performs the specific node (MA). Or a status request message (STREQ) to all nodes (MA) 11 or 21. The session manager then receives a status response message (STANS) from the specific node (MA) or all nodes (MA) 11 or 21. In this case, the status response message may send information separately for each MA 11 or 21, and may collect and send information of all session MAs 11 or 21.

After the service monitoring, the session manager checks whether the reservation time has expired (or the manager's request) in step 107. If it is determined that the reservation time has not expired, step 106 is performed again. If the reservation time has expired, the session is terminated at step 108.

Next, an operation for a many-to-many communication service in the CMA will be described with reference to the accompanying drawings.

4 is a diagram illustrating operation for a many-to-many communication service in a core multicast agent (CMA) according to an embodiment of the present invention.

Referring to FIG. 4, in step 201, the CMA attempts to join a session for a many-to-many communication service after starting an operation.

In detail, the CMA may be actively started by setting up a subscription in advance regardless of whether a CP or an ISP starts a session, or may be started manually by a MA in an edge domain applying for a subscription. In this way, the CMA transmits / receives a SM and SUBSREQ / SUBSANS control message with the SM as shown in FIG. 5A. Each CMA is provided with information about the virtual connection from the SM. This is because CMAs are not physically connected to each other, that is, because they are virtually connected, there is no physical information about neighboring nodes, so it is not easy to change the ring structure or immediately change the neighboring nodes. . The information on the virtual connection must be updated in all the MAs in order to change the ring structure or continuously update the information on the neighbor node, and is used to extend or recover the ring structure in the future. Therefore, the MAs can update the MAP information by sending a status request message to the SM every cycle. The information on the virtual connection may consider delay, bandwidth, address, and the like, and is managed in the SM.

If the CMA starts operation in step 201, the CMA performs a group joining procedure. In general, since the MA (or CMA) in the core domain may be operated by the ISP or the CP in advance, the MAs pre-installed by the ISP or the CP do not need a group joining procedure. However, when adding a MA in the core domain or adding a CMA on demand of the edge domain, they must perform a procedure for joining the group.

Accordingly, in step 202, the CMA inserts a many-to-many communication topology by finding an optimal location, ie, between adjacent nodes, by using enough MAP information in the core domain as shown in FIG. 5B. At this time, the newly joining node should never cause problems with the existing ring.

Accordingly, in step 203, the CMA continuously identifies CMAs existing in the same group (RMCP-3 session) while the session continues. For example, as shown in the accompanying FIG. 5C, information about the entire CAMs is obtained through the heartvest (CD_HB (HeartBest)) of the code domain operating along the ring shape in the core domain. Each CMA can continue to identify other CMAs through this process.

Then, in step 204, the CMA determines whether to leave the group that is currently joined. As a result, when the CMA leaves the core domain and leaves the group, in step 205, the CMA performs the group withdrawal procedure. Specifically, the group withdrawal procedure is performed as shown in the attached FIG. 5D between the members MA and the SMs constituting the session in the core domain. And even if the CMA leaves the group, it should not cause problems with the existing ring. In other words, the session must be maintained regardless of which node has left. Therefore, the leaving CMA sends the withdrawal request (LEAVREQ) to both the upstream and downstream streams, and informs the SM of the withdrawal. Inform the CMAs. Once the successful leaving process is completed, the horizontal HeartBeat (hereinafter referred to as HHB) message informs all members (MAs) of the leaving of the leaving CMA. The HHB message is a Heartvest (HB) message exchanged between the CMAs. The HHB message is used to announce the withdrawal of the CMA or to detect an abnormality in the connection forming the core domain.

When the group withdrawal procedure is completed as described above, in step 206, the CMA receives the many-to-many communication service termination request from the SM and terminates the session as shown in FIG. 5E. At this time, the message for requesting termination should be exchanged between the specific member (MA) constituting the session in the core domain and the SM, and in order to ensure delivery, the termination message should be delivered to all members constituting the session.

Meanwhile, an operation for a many-to-many communication service in an EMA will be described with reference to the accompanying drawings.

6 is a diagram illustrating EMA operation for a many-to-many communication service according to an embodiment of the present invention.

The operation of the EMA is similar to that of the CMA, but unlike the CMA having a ring-shaped control structure, the EMA has a tree-type control structure.

Referring to FIG. 6, unlike the CMA, the subscription of the EMA mainly consists of a passive subscription that subscribes to a session only when necessary. In step 301, the EMA attempts to join a session. In this session subscription, a subscription approval must be received from the SM. To this end, the EMAs transmit and receive control messages for the SUBSREQ / SUBSANS with the SM, as shown in FIG. 7A.

Since the EMA is not physically connected, there is no information about neighboring nodes. Therefore, bootstrapping information is required to join a session. In this case, the SM provides the EMA with the list of currently operating CMAs as necessary information for joining the session. Accordingly, in step 302, the EMA receives the CMA list received from the SM.

Then, in step 303, the EMA selects the most optimal CMA from the received CMA list and performs a group joining procedure for connecting to the selected CMA as shown in FIG. 7B. To determine this optimality, delay, bandwidth, address, and the like are considered. Specifically, the CMA provides the MA with a list of MAs in the edge domain in which the MA requesting new subscription will operate. Accordingly, the MA selects the most optimal EMA from the EMA list provided by the CMA and attempts to join.

Thereafter, when the EMA leaves the edge domain in step 304, the EMA performs a many-to-many communication service group leave procedure, as shown in FIG. 7C.

Then, in step 305, the EMA checks whether there is a session termination request from the SM or the outside. If there is a session termination request, the EMA terminates the session in step 306.

On the other hand, in order to provide a many-to-many communication service as described above, a mechanism for recovering a network error is required. These network errors consider largely the partitioning problem of the core network, the partitioning problem of the edge domain network, the shrunken problem of the core domain network, and the loop problem of the edge domain network. do.

First, a partitioning problem recovery mechanism of the core domain network will be described with reference to FIG. 8A.

In the core domain network, core domain partitioning may occur due to link disconnection between node 4 and node 5, node 4 disconnection, and node 5 software disconnection.

In the case of the link disconnection of the node 4 and the node 5 is not connected due to the problem of the network connecting the nodes 4 and 5, this is to establish a new path between the node 4 and node 5 so that the node 4 and node 5 The error can be recovered by connecting to the new node.

In the case of disconnection of the node 4, the application program of the node 4 does not operate normally. In this case, the link connecting the node 4 is normally operated, but since the node 4 does not operate, the ring is disconnected, and the ring connection is completed by removing the node 4 from the connection.

Finally, in case of software disconnection, the node 5 is a case in which the application program of the node 5 does not operate normally. In this case, the link connecting the node 5 is normally operated, but the node is disconnected because the node 5 is not operated. The ring connection is completed by removing the node 5 from the connection.

Secondly, the partitioning problem recovery mechanism of the edge domain network will be described with reference to FIG. 8B.

In the edge domain network, edge domain partitioning may occur due to link disconnection between node 2 and node 3 and node 2 disconnection.

In the case of the link disconnection of the node 2 and the node 3, the connection is disconnected due to a network problem connecting the node 2 and the node 3. In this case, node 3 can recover from the error by selecting a new node, so that node 3 selects a new node (node 4).

In the case of disconnection of the node 2, the application program of the node 2 does not operate normally. In this case, the link between node 2 and node 3 operates normally, but because the application of node 2 does not operate normally and the connection is disconnected, node 3 causes node 3 to select a new parent node (node 4).

Third, with reference to FIG. 8C, the reduced ring problem recovery mechanism of the core domain network will be described.

Shrinkage ring refers to a case where the size of a ring-shaped core domain is significantly reduced. In order to detect and recover such a shrinkage ring, the SM needs to know information about CMAs in real time, and all the MAs The premise of being able to share information about MAs must be met. Here, all MAs report this to the SM when joining or leaving the system, and the SM periodically checks the CDMA list information managed by polling the CDMAs. Update In order to presuppose that all MAs share information about all MAs with each other, all MAs in the core domain can be periodically updated between MAs on CD through HHB exchange with each other.

If the ring shrinks or overlaps, the MAs located on the failed CD receive an HHB of reduced information compared to the previous ring information.

Then, a method of detecting a contraction ring in the core domain is as shown in FIG. 9A. Referring to this, a node that can determine that the ring is contracted becomes a node that first sent the HHB at the time when the ring is contracted. . In this case, it is assumed that the nodes detecting the reduction ring are Node 1 and Node 4. The node 1 and the node 4 suspects a shrink ring when a difference occurs from the topology of the ring. At this time, it is also possible to consider a normal case in which the ring shrinks due to the withdrawal of members constituting the ring.

The reduction ring problem thus detected can be solved through the method as illustrated in FIG. 9B. That is, once the node senses the ring shrinkage, it requests the SM for a list of CDMAs. At this time, the node that detects the shrink ring to prevent the plurality of nodes from entering the shrink ring recovery step at the same time sends an HHB * message to the other node, that is, other MAs in the core domain ring, which means that the HHB generation is temporarily stopped. The SM also suppresses duplicate CDMA list request messages (CDMLREQ) for the same CD.

After that, the node that detects the shrinkage of the ring receives the CDMA List Response Message (CDMALANS) message from the SM, checks whether the received CDMA list is the same as the known list, and if it is the same, determines that it is a normal ring and then normal HHB Perform the action. On the other hand, if the received CDMA list is not the same, the node that detects the shrinkage of the ring has a shrinkage ring problem, and the CDMA needs to recover the ring. At this time, a refresh function is used to simultaneously optimize the ring. This refresh function is a procedure that allows each node to search for and connect to a new peer. At this time, since the ring is intended to manage the RMCP-3 communication environment, it is not necessary to immediately change the data transmission path.

Fourth, the loop problem recovery mechanism of the edge network will be described with reference to FIG. 8D.

The reason for loops in edge networks is that nodes in the edge domain occur as the tree changes due to the change of parent node or insertion / deletion of new node. In this case, each node detects that a loop has occurred through a vertical heartbeat (hereinafter referred to as a VHB), and recovers the detected loop through a MA list included in the VHB. VHB messages are HB messages exchanged between EMAs, unlike HHBs that exchange information between CMAs. The VHB message is generated by each of the CMAs, and the VHB message is delivered along the EMAs to the final leaf (last) EMAs. The main purpose of the VHB message is to detect anomalies on the edge domain.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the scope of the following claims, but also by the equivalents of the claims.

As described above, the present invention can be effectively applied to a group communication application service that needs to consider a plurality of senders and receivers at the same time, such as a network game, and can provide many-to-many multicast services at the application layer rather than introducing a pure multicast infrastructure. Therefore, it is possible to reduce the cost by minimizing the change of the current Internet device, and also to introduce the many-to-many communication service as soon as possible.

Claims (28)

Data transmission path for providing a many-to-many communication service in a session manager constituting a many-to-many communication environment formed of a plurality of edge domains including a plurality of agents for forwarding multicast data and a core domain connecting the edge domains. In the configuration method, Forming a new session according to a session establishment request from the outside; Subscribing to the formed new session; Periodically checking the status of agents in the new session to monitor session services; And And terminating the new session according to a preset time or a request of an administrator. The method of claim 1, Forming a new session according to the session establishment request from the outside, forming the new session by using the session profile information that is a set of information required to define a specific group communication service session received when the session establishment request from the outside A data transmission path configuration method for providing a many-to-many communication service. The method of claim 1, wherein the joining of the formed new session comprises: Checking whether the new session can be joined; And And if it is possible to subscribe to the new session, providing a session join permission and information for subscribing to a session service to a corresponding node. The method of claim 1, wherein the monitoring of the session service comprises: Periodically sending a status request message to each node in the new session; And receiving a status response message including information of its operation and communication quality periodically according to a status request from each node. In a many-to-many communication environment consisting of a plurality of edge domains including a plurality of multicast agents for forwarding multicast data, a core domain connecting the edge domains, and a session manager, the core domain of the plurality of multicast agents. A method for configuring a data transmission path for providing a many-to-many communication service by a core multicast agent present in Attempting to join a newly created session in the session manager; Subscribing to the session using the information of the multicast agents obtained at the session join attempt; Continuously identifying information about other core multicast agents present in the session; And And informing the session manager and the other core multicast agents of the leaving of the session when leaving the session. The method of claim 5, And receiving an end of session request from an external session, informing the session manager and all multicast agents in the session of the end of the session. The method of claim 5, wherein attempting to join a newly created session in the session manager comprises: Transmitting a subscription approval request control message to the session manager; And Receiving a subscription approval response control message including information on a virtual connection from the session manager. The method of claim 5, wherein joining the session comprises: Searching for an optimal location in the core domain network by using information of the multicast agents obtained at the session join attempt; And And inserting a communication topology of the mobile station into the searched optimal location. The method of claim 5, Information about other core multicast agents present in the session is periodically provided by the other core multicast agents through a heart domain (CD_HB) message in the core domain operating along a ring in the core domain. A data transmission path configuration method for providing a many-to-many communication service. In a many-to-many communication environment consisting of a plurality of edge domains including a plurality of multicast agents for forwarding multicast data, a core domain and a session manager connecting the edge domains, among the plurality of multicast agents. A method for configuring a data transmission path for any edge multicast agent present to provide many-to-many communication services, the method comprising: Attempting to join a newly created session in the session manager; Selecting an optimal core multicast agent existing in the edge domain among the multicast agents by using information received from the session manager when attempting to join the session; Joining the session group by connecting to the selected core multicast agent; And And notifying the session manager and the other edge multicast agents of leaving the group when leaving the group of sessions. The method of claim 10, And receiving a session termination request from the outside, terminating the session. The method of claim 10, wherein the attempting to join a newly created session in the session manager comprises: Transmitting a subscription approval request control message to the session manager; And Receiving a subscription grant response control message from the session manager including a list of core multicast agents currently operating in the edge domain. The method of claim 10, wherein joining the group of the session by connecting to the selected core multicast agent comprises: Attempting to connect with the selected core multicast agent; Receiving a list of edge multicast agents in the edge domain in which the connected core multicast agent is to operate; Selecting the most optimal edge multicast agent from the list; And Subscribing to the selected edge multicast agent. The method of claim 6, And periodically monitoring whether an error occurs in the core domain during the many-to-many communication service, and recovering the generated error. The method of claim 14, wherein the recovering of the generated error comprises: Determining the cause of division when the core domain is divided; Connecting with a new multicast agent if the cause of the split is a disconnection between any two multicast agents; And If the cause of the division is a disconnection of any one multicast agent, removing the disconnected multicast agent from the connection, comprising: providing a data transmission path for providing a many-to-many communication service. The method of claim 14, wherein the recovering of the generated error comprises: Detecting a reduced ring of the core domain; Requesting the session manager a list of multicast agents in the core domain; And Optimizing the ring of the core domain by connecting to a new multicast agent by using the list of multicast agents. The method of claim 16, wherein recovering the generated error comprises: Sending a message HHB * to other multicast agents in the ring of the core domain to temporarily suspend the generation of a horizontal heartvest message (HHB) exchanged between core multicast agents. A data transmission path configuration method for providing a many-to-many communication service. The method of claim 16, The detecting of the reduced ring of the core domain may include determining that the ring of the core domain is reduced when the topology of the previous topology does not match that of the currently provided ring. How to configure a data transfer path for data. The method of claim 16, The detecting of the reduced ring of the core domain may include determining that the ring of the core domain is reduced when any multicast agent in the ring of the core domain leaves. To configure data transfer paths for data. The method of claim 10, And periodically monitoring whether an error occurs in the edge domain during the many-to-many communication service, and recovering the generated error. The method of claim 20, wherein recovering the generated error comprises: Determining a cause of division when the edge domain is divided; Connecting to a new multicast agent if the cause of the split is a disconnection of any two multicast agents; And Requesting the disconnected multicast agent select a new parent multicast agent if the cause of the split is a disconnection of any one multicast agent. Configuration method. The method of claim 20, Detecting loop occurrences in the edge domain; And Recovering the sensed loop using the list of multicast agents present in the edge domain. A session manager for establishing a session for a many-to-many communication service, controlling joining and leaving the formed session, and managing information related to the many-to-many communication service; A plurality of edge multicast agents connected in a tree form in a plurality of edge domains to subscribe to the formed session using information received from the session manager, and informing the session manager that the session manager has left the session; And A plurality of core multicast agents that connect the edge domains in a ring form in the core domain, join the formed session using the information received from the session manager, and notify the session manager that the session manager has left when leaving the session. And a data transmission path for providing a many-to-many communication service. The method of claim 23, wherein The edge and core multicast agents, upon receiving an external session termination request, inform the session manager and all multicast agents within the session of the session termination, and errors in the edge domain and core domain during the many-to-many communication service. A system for configuring data transmission paths, characterized by a periodic monitoring of occurrences and recovery of errors that have occurred. The method of claim 24, An error occurring in the edge domain is a partitioning and shrinking ring problem, the partitioning problem is connected to a new multicast agent to recover the error, and the shrinking ringing problem is caused by leaving the corresponding multicast agent having an error. Data transmission path configuration system, characterized in that for recovering. The method of claim 24, The error occurring in the core domain is a problem of splitting and looping in the core domain, and connecting to a new multicast agent to recover the error. The method of claim 23, wherein When the session manager receives a subscription request from the edge and core multicast agents existing in the formed session, the session manager grants the subscription to the formed session using authentication information of the corresponding multicast agent and sends the request to the corresponding multicast agent. Providing information necessary for subscribing to the formed session. The method of claim 27, And the session manager periodically checks the edge and core mobile agents existing in the formed session to update the information for the many-to-many communication service.

Images