WO2009111990A1 - Multicast controlling method, device and system - Google Patents

Multicast controlling method, device and system Download PDF

Info

Publication number
WO2009111990A1
WO2009111990A1 PCT/CN2009/070776 CN2009070776W WO2009111990A1 WO 2009111990 A1 WO2009111990 A1 WO 2009111990A1 CN 2009070776 W CN2009070776 W CN 2009070776W WO 2009111990 A1 WO2009111990 A1 WO 2009111990A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
multicast
multicast group
label
edge node
Prior art date
Application number
PCT/CN2009/070776
Other languages
French (fr)
Chinese (zh)
Inventor
黄勇
查敏
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2009111990A1 publication Critical patent/WO2009111990A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/16Multipoint routing

Definitions

  • the IPTV service basically has two forms of video on demand (VoD, Video on Demand) and broadcast television (BTV, Broadcast TV), in which VoD transmits television programs through unicast.
  • the BTV service specifically refers to a television program transmitted by broadcasting or multicast on an IP network.
  • the video service occupies a lot of network bandwidth resources. When a large number of video programs are flooded on the network, it will randomly compete for the limited bandwidth of the network. To ensure the quality of services, the network needs to manage the resources occupied by the network. According to the idle status of the bandwidth resources, whether to allow a certain service to occupy network bandwidth or release network bandwidth is called resource admission and release control.
  • the existing technology implements the resource admission and release control of the multicast service in the core network.
  • the existing multicast protocol technology is to establish a multicast tree for the media stream corresponding to each multicast group.
  • the multicast routers that perform multicast routing management will consume processing resources by a large number of multicast routing protocol packets, and the storage space will be occupied by a large number of multicast state records.
  • Network capacity is difficult to expand due to these limitations. This problem is called scalability.
  • the process of multicast transmission does not have a bandwidth guarantee mechanism.
  • the technical problem to be solved by the embodiments of the present invention is to provide a multicast control method, device, and system, which implements unified management of multicast bandwidth resources, and responds to an edge node requesting to join or leave in time.
  • a request from a multicast group dynamically allocates bandwidth resources.
  • An embodiment of the present invention provides a method for multicast control, including the following steps: receiving a request for joining a multicast group by a first edge node, or a request for leaving a multicast group by a first edge node; when receiving a join request Using the unicast route to send a lookup packet to the downstream node to find the nearest join copy point corresponding to the first edge node; or when receiving the leave request, use the unicast route to find the nearest leave copy point corresponding to the first edge node
  • the search packet carries the label identifier of the first edge node, the multicast label of the multicast group, and the label identifier of the egress node of the multicast group; when the recently added copy point is found, Requesting to reserve bandwidth resources for the recently added replication point to the first edge node, and after receiving the resource acknowledgement response signal, modifying the multicast label of the multicast group to join the first edge node to the multicast Group; or when the recently left copy point is found, the request releases the bandwidth resource between
  • an embodiment of the present invention further provides a network node, including:
  • a receiving unit configured to receive a join request of the first edge node to the multicast group, or a leave request of the first edge node to the multicast group; and receive a resource reservation acknowledgement response or a resource release acknowledgement response;
  • a searching unit configured to: according to the received join request to the multicast group, find the nearest join point of the first edge node on the forwarding link of the media stream of the multicast group; or according to the received pair of multicast groups The leaving request, on the forwarding link of the media stream of the multicast group, find the nearest leaving copy point of the first edge node;
  • a sending unit configured to send a bandwidth resource reservation request between the found nearest joining copy point to the first edge node, or send a bandwidth resource release request between the found nearest leaving copy point to the first edge node;
  • a processing unit configured to: after receiving the resource reservation confirmation response, modify the multicast label of the multicast group to join the first edge node to the multicast group; or after receiving the resource release confirmation response, modify the A multicast label of a multicast group to delete the first edge node from the multicast group.
  • an embodiment of the present invention further provides a network node, including:
  • a receiving unit configured to receive a recent join of the first edge node that finds the request to join the multicast group
  • the lookup message of the point, or the lookup message of the nearest leaving copy point of the first edge node that requests to leave the multicast group, the lookup message includes the label identifier of the first edge node and the current receiving multicast group. a label identifier of a network edge node of the media stream, and a multicast label of the multicast group;
  • a determining unit configured to calculate, according to the received search message, a current multicast label by using a unicast route, and determine whether it is a recently added copy point or a recently left copy point;
  • the sending unit is configured to: when the judgment result is yes, send the node identifier of the node to the ingress node of the multicast group; when the judgment result is no, the device forwards the search packet to the downstream node according to the unicast route, and the search packet is searched.
  • the multicast label in the modification is changed to the current multicast label.
  • the embodiment of the present invention further provides a system for multicast control, including:
  • first edge node a first edge node, a second edge node, an intermediate node, and a network resource management node
  • the first edge node is configured to send a join request to the multicast group, or send a leave request to the multicast group;
  • the second edge node is configured to receive a join request of the first edge node to the multicast group, or receive a leave request of the first edge node to the multicast group, according to the received first edge node joining the multicast group And sending, on the forwarding link of the media stream of the multicast group, a lookup packet for finding a recently added copy point; or, according to the received first edge node leaving the multicast group, the media stream in the multicast group On the forwarding link, the device searches for a lookup packet that has recently left the replication point; the lookup packet carries the label identifier of the first edge node, the multicast label of the multicast group, and all the exit nodes of the multicast group.
  • the network resource management node after finding the nearest join copy point, requesting the network resource management node to reserve a bandwidth resource between the recently added replication point and the first edge node, and receiving the resource confirmation response sent by the network resource management node
  • the multicast label of the multicast group is modified to join the first edge node to the multicast group; or when the nearest copy point is found,
  • the network resource management node requests to release the bandwidth resource between the recently left replication point and the first edge node; and after receiving the network resource management node resource sending the acknowledgement response signal, modify the multicast label of the multicast group to The first edge node is deleted from the multicast group;
  • the intermediate node is configured to receive a lookup packet sent by the second edge node, calculate a current multicast label by using a unicast route according to the lookup packet, and determine whether the intermediate node itself is recently added to the replication point or recently left. Copying the point; if yes, reporting to the second edge node; if otherwise forwarding the lookup packet to the downstream node on the forwarding link of the media stream of the multicast group, changing the multicast label in the lookup message to the Current multicast label;
  • the network resource management node is configured to perform bandwidth resource reservation after receiving a request for reserving bandwidth resources between the recently added replication point and the first edge node sent by the second edge node, and when the reservation is successful Transmitting a resource acknowledgment response to the second edge node; or receiving a request for releasing the bandwidth resource between the recently leaving the replication point and the first edge node sent by the second edge node, releasing the bandwidth resource, and releasing the resource after the resource is released
  • the edge node sends a resource acknowledg
  • the process of uniformly managing multicast transmission can be implemented, especially when an egress node newly joins or quits a multicast group, and the nearest join copy point is reserved in time by searching for the nearest join copy point corresponding to the egress node or leaving the copy point recently.
  • the bandwidth between the egress nodes, or the bandwidth that has recently left the replication point to the egress node, enables efficient and reasonable allocation of bandwidth resources. Increased network bandwidth utilization.
  • FIG. 1 is a schematic diagram of a method for implementing multicast admission control according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of a method for implementing multicast release control according to an embodiment of the present invention
  • FIG. 3 is a network diagram of a method for implementing multicast admission control according to an embodiment of the present invention
  • FIG. 4 is a flowchart of a method for implementing multicast admission control according to an embodiment of the present invention
  • FIG. 5 is a network diagram of a method for implementing multicast release control according to an embodiment of the present invention
  • FIG. 6 is a flowchart of a method for implementing multicast release control according to an embodiment of the present invention
  • FIG. 7 is a system block diagram of a multicast control according to an embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of an entry node according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of an intermediate node according to an embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of a network resource management node according to an embodiment of the present invention.
  • the embodiments of the present invention provide a method for implementing multicast resource admission and release control, which uses a Multi-Protocol Label Switching (MPLS) label to identify a multicast packet to be reached. All network egresses, a packet with an MPLS multicast label can identify the egress that the packet is currently pointing to at any node in the network.
  • the multicast label changes with the process of forwarding, when an interface from a node Packet group sent by other interfaces of this node when sent to a specific exit The broadcast tag no longer contains the identity of the particular exit, thus automatically avoiding forwarding loops.
  • MPLS Multi-Protocol Label Switching
  • a media stream having a multicast group ⁇ g, S> is transmitted according to a forwarding path of a current arrow connection, where the multicast group ⁇ 8
  • the entry node of > is the node
  • the egress node is the nodes (8) and (11).
  • a new egress node (7) needs to join the multicast group, it must find the node with the shortest path on the existing media stream forwarding link. Thereby establishing a new multicast forwarding path.
  • the recently added replication point is a node in the network domain and meets the following three conditions: a) The node has received the media stream of the multicast group when a new egress node requests to join the multicast group; b) the node is located on the current unicast forwarding path from the network ingress node corresponding to the multicast group to the new egress node joining the multicast group;
  • the unicast forwarding path of the node to the new egress node is the shortest.
  • a media stream having a multicast group ⁇ g, s> is transmitted according to a forwarding path of a current arrow connection, where the multicast group ⁇ 8
  • the entry node of > is node (1)
  • the egress node is node (7), (8), (9), (11).
  • the shortest multicast forwarding branch node releases the bandwidth resource between the multicast forwarding branch node and the egress node (8).
  • the recently left replication point is a multicast forwarding branch node in the network domain and meets the following three conditions:
  • the node has received the media stream of the multicast group when the egress node requests to leave the multicast group; b) the node is located from the network ingress node corresponding to the multicast group to the egress node of the leaving multicast group On the current unicast forwarding path;
  • FIG. 3 it is a schematic diagram of a network for implementing a multicast resource admission control method according to an embodiment of the present invention.
  • a label identifier In the orthogonal label-based network domain, in order to identify each edge node and map the identifier to the label, a label identifier needs to be defined for each edge node, for example, defining labels for edge nodes eight, B, C, and D.
  • the identifiers are 0x00001, 0x00002, 0x00004, 0x00008, and the label identifiers of all edge nodes are orthogonal, that is, the result of any two label identifiers is 0.
  • the media stream of a multicast group ⁇ g, s> is sequentially transmitted from the path of the node B, Y, Z, D, wherein the ingress node of the multicast group ⁇ g, s> is the node B, and the egress node is the node D;
  • the multicast label is the label of each egress node of the multicast group. That is, the multicast label of the multicast group ⁇ g, s> is the label ID 0x00008 of the egress node D.
  • the request to join the multicast group ⁇ 8> arrives at the exit node A.
  • the egress node A sends a request for joining the multicast group with the own tag identifier to the network resource manager RM, and the RM centrally stores all the ingress node information corresponding to the multicast group ⁇ 8>.
  • the RM forwards the request of the egress node A to the multicast group to the ingress node B.
  • the request of the egress node A to join the multicast group ⁇ g, s> is forwarded to the ingress node B through the network resource manager RM, which is formally a stream represented by a quintuple whose destination address is a multicast address.
  • the request to join the multicast group may also be received by the RM from the other RM through the inter-domain interface, and then forwarded to the ingress node B; or sent directly by the node A to the ingress node B through unicast.
  • the ingress node B After receiving the request to join the multicast group ⁇ g, s>, the ingress node B sends a unicast route to send a packet to the nearest join point, where the lookup packet carries the egress node requesting to join the multicast group.
  • the tag identifier of A is 0x00001 and the tag identifier of the egress node that has received the multicast group ⁇ 8> media stream, that is, the tag identifier 0x00008 of the current egress node D, and the multicast tag 0x00008 of the multicast group ⁇ 8>.
  • the ingress node B sends a lookup message for finding the most recently added copy point, and sends it to the downstream node to let all the nodes that have received the media stream perceive the processing. Therefore, the information carried in the search packet is also multicast.
  • the tag of the ingress node B of ⁇ g,s> is 0x00002.
  • the downstream intermediate node Y receives the lookup packet, and determines whether it is the closest join copy point to the exit node A. Specifically, the intermediate node Y calculates the reachability of all the egress nodes by using the unicast route on each of the egress interfaces, and calculates the label of each egress interface label. The sum of all egress node label identifiers that are reachable on the outbound interface, and the label mask of the outbound interface. Since the lookup packet carries a multicast label, the intermediate node Y can know that the egress node A passes through the intermediate node Z according to the current unicast route.
  • the intermediate node Y associates the label mask 0x00009 with the multicast label 0x00008 in the lookup 4, and obtains the exit label of the intermediate node Y to the egress node A as 0x00008, and the egress label is obtained after exchanging labels through the intermediate node Y.
  • Current multicast label Because the intermediate node Z is forwarding the media stream to the egress node D on the unicast forwarding path of the egress node A, the calculated current multicast label in the direction of the egress node A is not zero, indicating that the intermediate node Y to the egress node The unicast forwarding path of A is not the shortest.
  • the intermediate node Y then forwards the lookup packet to the downstream intermediate node Z.
  • the multicast label in the lookup packet is the current multicast label 0x00008.
  • the intermediate node Y determines the direction of the egress node as the node D according to the multicast label in the lookup packet, and calculates the label mask of the outbound interface in the direction of the node D.
  • the egress interface label is the egress node (whether or not the egress node The sum of the tag identifiers in the multicast group, that is, the sum of the tag identifiers of the egress nodes A and D 0x00009.
  • the intermediate node Z After receiving the lookup message, the intermediate node Z also determines whether it is the closest join point to the exit node A. Specifically, the intermediate node Z calculates the label mask of the egress interface of the egress node A by using a unicast route to be 0x00001, and the label mask 0x00001 is matched with the received current multicast label 0x00008 to obtain the intermediate node Z to the egress.
  • the egress label of node A is zero.
  • the egress label is the current multicast label obtained after the intermediate node Z exchanges labels.
  • the current multicast label to the egress node A is zero, which indicates that the intermediate node Z to the egress node A No media stream is forwarded on the unicast forwarding path, thereby determining that the intermediate node Z is the closest join copy point.
  • the intermediate node Z sends a unicast response signal to the ingress node B according to the label identifier of the ingress node B, or other identifiers, and reports that it is the most recently added copy point of the current search.
  • the ingress node B After receiving the unicast response signal, the ingress node B sends a resource admission request to the network resource manager RM, requesting to reserve the multicast bandwidth resource from the intermediate node Z to the egress node A.
  • the intermediate node Z can also directly report to the network resource manager RM that it is the most recently added replication point of the current lookup, and sends a resource admission request, requesting the RM reservation from the intermediate node Z to the exit node A. Multicast bandwidth resources.
  • the RM completes resource reservation and gives a resource confirmation response signal.
  • the ingress node B After receiving the resource confirmation response signal, the ingress node B changes the multicast label corresponding to the multicast group ⁇ &8>, and also uses the newly added egress node A as an egress node of the multicast group ⁇ 8> media stream.
  • the multicast group media stream is forwarded from the intermediate node Z to the egress node A.
  • FIG. 5 it is a schematic diagram of a network for implementing a multicast resource release control method according to an embodiment of the present invention.
  • an identifier needs to be defined for each edge node, for example, a label is defined for the edge nodes A, B, C, and D in advance.
  • the identifiers are 0x00001, 0x00002, 0x00004, 0x00008, and the label identifiers of different edge nodes are orthogonal, that is, the result of any two label identifiers is 0.
  • the media stream with a multicast group ⁇ g, s> is sequentially transmitted from the path of the node B, Y, Z, D and the path from the node: 8, ⁇ , ⁇ , ⁇ , where the entry of the multicast group ⁇ 8>
  • the node is the node B
  • the egress node is the node D and the A.
  • the multicast label of the multicast group is the sum of the identifiers of the egress node A and the D tag, that is, the multicast tag of the multicast group ⁇ 8> is 0x00009.
  • the method for implementing multicast release control is as follows:
  • the request to leave the multicast group ⁇ g, s> reaches the exit node A.
  • the egress node A sends a request to leave the multicast group with the own tag identifier to the network resource manager RM, where the RM stores all the ingress node information corresponding to the multicast group.
  • the network resource manager after finding the ingress node B corresponding to the multicast group ⁇ g, s>, sends the request of the egress node A to leave the multicast group to the ingress node B.
  • the request of the egress node A to leave the multicast group ⁇ g, s> is forwarded to the ingress node B through the network resource manager RM, which may be a stream represented by a quintuple whose destination address is a multicast address.
  • the leaving multicast request may also be received by the RM from the other RM through the inter-domain interface, and then forwarded to the ingress node B; or sent by the egress node A to the ingress node B by unicast.
  • the ingress node B After receiving the request to leave the multicast group ⁇ g, s>, the ingress node B sends a packet to the nearest leaving replication point by using the unicast route, where the lookup packet carries the label identifier of the egress node A. 0x00001 and the label ID 0x00008 of the egress node D that has received the multicast group ⁇ g, s> media stream, and the multicast label 0x00009 of the multicast group ⁇ g, s>.
  • the ingress node B sends a lookup message for the nearest leaving of the replication point, and sends it to the downstream node, so that all the nodes that have received the media stream are aware of the processing, so the information carried in the search packet also corresponds to the multicast group ⁇ 8 >
  • the tag identifier of the ingress node B is not limited to the multicast group ⁇ 8 >.
  • the downstream intermediate node Y receives the lookup packet, and determines whether it is the nearest leaving replication point to the egress node A. Specifically, the lookup packet carries the multicast label, and the intermediate node Y learns that the egress node A passes through the intermediate node Z according to the current unicast route, and the label mask of the outbound interface in the Z direction of the intermediate node is 0x00009.
  • the label mask is the sum of the IDs of the egress node A and the D tag.
  • the intermediate node Y calculates the reachability of all egress nodes through unicast routes on each outbound interface, and calculates the label mask of each egress interface.
  • the intermediate node Y determines the direction of the egress node as the node D according to the multicast label, and calculates the label mask of the outbound interface in the direction of the node D.
  • the outbound interface label is the egress node (whether the egress node is in the multicast group or not) Irrelevant)
  • the sum of the tag identifiers that is, the sum of the tag identifiers of the exit nodes A and D 0x00009.
  • the label mask 0x00009 is matched with the multicast label 0x00009 carried in the search text, and the exit label of the intermediate node Y to the exit node A is 0x00009.
  • the egress label is the current multicast label obtained after the label is exchanged by the intermediate node Y. Because the path to the egress node A also has the multicast forwarding branch node Z forwarding the media stream to the egress node D, the current group is calculated.
  • the broadcast tag is the sum of the two exit node A and D tag identifiers, which can determine that the intermediate node Y is not the closest to the copy point.
  • the intermediate node Y forwards the lookup packet to the downstream intermediate node Z.
  • the multicast label carried in the lookup packet is the current multicast label 0x00009.
  • the intermediate node Z is a multicast forwarding branch node. After receiving the lookup packet, it determines whether it is the closest leaving replication point to the egress node A. Specifically, the intermediate node Z calculates a label mask of the outbound interface to the exit node A by using a unicast route to be 0x00001, and the label mask is matched with the received current multicast label 0x00009, and the obtained exit label is 0x00001, the egress label 0x00001 is the current multicast label obtained after the intermediate node Z exchanges labels, because the value of the current multicast label in the direction from the intermediate node Z to the egress node A is equal to the label identification value of the egress node A.
  • no other multicast forwarding branch nodes are forwarding the media stream between the intermediate node Z and the egress node A, thereby determining that the intermediate node Z is the closest leaving the replication point.
  • the intermediate node Z sends a unicast response signal to the ingress node B according to the label identifier of the ingress node B, and reports that it is the nearest leaving copy point of the current lookup.
  • the ingress node B After receiving the unicast response signal, the ingress node B sends the resource to the network resource manager RM.
  • the release request requests to release the multicast bandwidth resource from the intermediate node to the egress node.
  • the intermediate node Z can also directly report to the network resource manager RM that it is the nearest leaving replication point of the current lookup, and send a resource admission request, requesting the RM to release the multicast bandwidth resource from the intermediate node Z to the egress node A.
  • the network resource manager sends a resource confirmation response signal after the RM completes the resource release.
  • the ingress node B After receiving the resource confirmation response signal, the ingress node B changes the multicast sending label of the corresponding multicast group ⁇ &8>, and deletes the egress node A from the multicast group ⁇ 8>. Thus, the intermediate node Z no longer forwards the media stream of the multicast group ⁇ g, s> to the egress node A.
  • the network resource management node can be used to uniformly manage the multicast transmission process, especially when an egress node newly joins or exits the multicast group, and finds the nearest join copy point corresponding to the egress node or recently leaves the replication point, and reserves in time.
  • the bandwidth of the replication point to the egress node is recently added, or the bandwidth between the egress node and the egress node is released in time to realize dynamic and reasonable allocation of bandwidth resources. Increased network bandwidth utilization.
  • the multicast forwarding process of each node does not depend on the multicast routing protocol, but identifies a network domain edge node, and uses the forwarding tree for a multicast group in the topology of the network domain from the portal.
  • the forwarding topology of the node to the egress node is described.
  • all multicast groups with the same forwarding topology in the network domain perform no separate multicast transmission in the network domain, thereby realizing multicast convergence.
  • the identifier of the network edge node to be reached is uniformly placed in a multicast label and carried by the multicast packet traversing the network domain.
  • the node can implement multicast forwarding by using the multicast label information. Based on this technology, the intermediate nodes do not need to maintain the forwarding state of each multicast group, which reduces the number of management packets.
  • the embodiment of the present invention further provides a multicast control system, where the network domain includes: a network resource management node, an ingress node, an intermediate node, and an egress node.
  • the ingress node, the intermediate node, and the egress node may all be routers in hardware, but the functions implemented are different.
  • the egress node is configured to send a request to join the multicast group or leave the multicast group. Specifically, the egress node analyzes the requirement of the media stream that the user receives the multicast group, and sends a join request to the multicast group by using the sending unit. , or a request to leave a multicast group;
  • the ingress node is configured to send a lookup packet to the downstream node of the multicast group according to the received request that the egress node joins the multicast group, and find the latest join with the shortest forwarding path between the egress node. Copying a point; or sending a lookup packet to a downstream node of the multicast group according to the received request that the egress node leaves the multicast group, and searching for the closest leaving copy point with the shortest forwarding path between the egress node;
  • the packet carries the multicast label of the multicast group and all the egress node label identifiers of the multicast group, and the label identifier of the egress node that requests to join or leave; after finding the nearest join point, the request is for the egress node to Reserving resources between recently joined replication points, or requesting to release resources between the egress node and the nearest leaving replication point after finding the nearest leaving replication point; and tampering with the group after receiving the resource reservation confirmation response
  • the ingress node includes:
  • a receiving unit configured to receive a request for joining the multicast group by the egress node, or a request for leaving the multicast group by the egress node;
  • a searching unit configured to: according to the received join request of the egress node to the multicast group, search for the shortest recent join replication with the egress node on the forwarding link of the media stream of the multicast group Or: according to the received leaving request of the egress group to the multicast group, on the forwarding link of the media stream of the multicast group, searching for the closest leaving copy point with the shortest forwarding path between the egress node;
  • a sending unit configured to send the discovered bandwidth resource reservation request between the recently added replication point to the egress node, or send the discovered bandwidth resource release request between the recently left replication point to the egress node To the network resource management node;
  • the processing unit is further configured to: after receiving the resource reservation confirmation response, modify the multicast label of the multicast group to join the egress group of the request to the multicast group to join the multicast group; or receive the resource After the acknowledgment response is released, the multicast label of the multicast group is modified, and the egress node that leaves the multicast group is deleted from the multicast group.
  • the search unit described above includes:
  • a packet generating unit configured to generate a lookup packet, where the lookup packet includes a label identifier of the egress node, a label identifier of all network edge nodes that currently receive the media stream of the multicast group, and a group of the multicast group Broadcast label; may also include the label identifier of the entry node itself;
  • a message sending unit configured to send the lookup message to a node on the multicast group media stream forwarding link;
  • the response receiving unit is configured to receive the node identifier of the most recently joined replication point, or the node identifier of the most recently left replication point.
  • the intermediate node in the system is configured to calculate a current multicast label by using a unicast route according to the received search packet, and determine whether the intermediate node itself joins the recently copied point or recently leaves the copy point; if yes, the The ingress node of the multicast group reports that if the downstream node on the media stream forwarding link of the multicast group forwards the lookup packet, the multicast label in the lookup packet is modified to the current multicast label.
  • the intermediate node includes:
  • a receiving unit configured to receive a lookup message for finding a recently added copy point or a recently left copy point, where the lookup message includes a label identifier of an egress node that requests to join the multicast group and a network that currently receives the multicast group media stream a label identifier of the edge node, or a label identifier of the egress node requesting to leave the multicast group and a label identifier of the network edge node currently receiving the multicast group media stream;
  • a determining unit configured to calculate, according to the received lookup message, a current multicast label by using a unicast route, and determine whether it is the recently added copy point or the recently left copy point;
  • a sending unit configured to: when the judgment result is yes, send the node identifier of the node to the ingress node of the multicast group; if the judgment result is no, the received search packet is forwarded according to the unicast route to the The media stream of the multicast group forwards the downstream node on the link.
  • the determining unit described above includes:
  • a multicast label switching unit configured to: associate a label mask of an outgoing interface of the egress node with the multicast label of the multicast group, and obtain an exchanged multicast label; And calculating, according to the label identifier of the network edge node of the media stream that is currently receiving the multicast group in the packet, using the unicast route, calculating a label mask of the outbound interface on the forwarding path of the egress node, and the specific calculation method See method embodiment.
  • a comparison processing unit configured to determine whether the current multicast label is zero, if yes, determine that it is the closest join point; or determine whether the current multicast label is equal to the label identifier of the egress node requesting to leave the multicast group, and if yes It is determined that it is the most recent departure point.
  • the network resource management node RM includes:
  • a receiving unit configured to receive a bandwidth resource reservation request sent by the ingress node to the egress node, or receive a bandwidth resource release request between the egress node and the egress node sent by the ingress node; Receiving, by the egress node, a request for joining a multicast group, or a request for leaving the multicast group by the egress node; a processing unit, configured to reserve, according to the received bandwidth resource reservation request, a bandwidth resource that is recently added to a replication point to the egress node; or release the latest according to the received bandwidth resource release request Leaving the copy point to the bandwidth resource between the egress nodes.
  • the network resource management node further includes:
  • a forwarding unit configured to search for an ingress node corresponding to the multicast group according to the received join request or leave request, and send the received join request of the egress node to the multicast group or the egress node to the multicast group Leaving the request, forwarding to the ingress node;
  • a bandwidth allocation confirmation unit configured to send a resource reservation confirmation response to the ingress node after the processing unit reserves the bandwidth resource between the most recently added replication point and the egress node; or in the processing unit After releasing the bandwidth resource that has recently left the replication point to the egress node, a resource release confirmation response is sent to the ingress node.
  • the network resource management node, the ingress node, the intermediate node, and the egress node are all connected by a link or connected by a wireless network, and the network resource management node may be a gateway. , or the interconnected border point gateway, or server, centrally stores information about all ingress nodes corresponding to the multicast group ⁇ g, S, such as the ingress node identifier.
  • the system has a simple structure.
  • the multicast forwarding process of each node does not depend on the multicast routing protocol. That is, the intermediate node does not need to maintain the state of each group. This reduces the number of management packets and prevents network congestion.
  • the system manages the multicast transmission process through the network resource management node. In particular, when an egress node joins or quits a multicast group, the multicast label is modified in time to dynamically allocate bandwidth resources.
  • an embodiment of the present invention further provides an ingress node 80, including:
  • the receiving unit 801 is configured to receive a request for joining the multicast group by the egress node, or a request for leaving the multicast group by the egress node, and is also used for receiving a resource reservation confirmation response or a resource release confirmation response sent by the network resource management node.
  • the searching unit 802 is configured to: according to the received join request of the egress group to the multicast group, search for a shortest nearest join copy point with the egress node on the forwarding link of the corresponding multicast group media stream; or And according to the received leaving request of the egress group to the multicast group, searching for the nearest leaving copy point with the shortest forwarding path between the egress node and the forwarding link of the corresponding multicast group media stream;
  • the sending unit 803 is configured to send the discovered bandwidth resource reservation request between the recently added replication point to the egress node, or send the discovered bandwidth resource between the recently left replication point and the egress node. Release a request to the network resource management node;
  • the processing unit 804 is configured to modify the multicast label of the multicast group according to the received resource reservation confirmation response, to join the egress node to the multicast group, or release an acknowledgment response according to the received resource Modifying the multicast label of the multicast group to remove the egress node from the multicast group.
  • the search unit 802 described above includes:
  • the message generating unit 8021 is configured to generate a lookup message, where the lookup message includes a label identifier of the egress node, a label identifier of a network edge node that currently receives the multicast group media stream, and a multicast group multicast group.
  • the tag may also include the tag identifier of the ingress node 80 itself.
  • a message sending unit 8022 configured to send the lookup message to a downstream node of a media stream forwarding link of the multicast group
  • the response receiving unit 8023 is configured to receive the node identifier of the most recently joined copy point, or the node identifier of the closest leaving the copy point.
  • the embodiment of the present invention further provides an intermediate node 90, including: a receiving unit 901, configured to receive a lookup packet for finding a recently added copy point or a recently left copy point, where the lookup message includes a request.
  • a receiving unit 901 configured to receive a lookup packet for finding a recently added copy point or a recently left copy point, where the lookup message includes a request.
  • the label of the egress node that joins and leaves the multicast group the label of the network edge node that receives the media stream of the multicast group, and the multicast label of the multicast group.
  • the search packet may also include the multicast group.
  • the tag identifier of the ingress node 80 is an intermediate node 90, including: a receiving unit 901, configured to receive a lookup packet for finding a recently added copy point or a recently left copy point, where the lookup message includes a request.
  • the label of the egress node that joins and leaves the multicast group the label of the network edge node that
  • the determining unit 902 is configured to calculate, according to the received search message, a current multicast label by using a unicast route, and determine whether it is the recently added copy point or the recently left copy point; the sending unit 903 When the judgment result is yes, the node identifier of the multicast group is sent to the ingress node of the multicast group; when the judgment result is no, the downstream node on the link is forwarded to the media stream of the multicast group according to the unicast route. Forward the search packet and change the multicast label in the lookup packet to the current multicast label.
  • the determining unit 902 includes:
  • the multicast label switching unit 9021 is configured to calculate, according to the label identifier of the network edge node of the media stream that currently receives the multicast group in the packet, use the unicast route to calculate the forwarding path to the egress node.
  • the label mask of the interface is matched with the multicast label of the multicast group to obtain an exchanged multicast label, that is, the current multicast label. For the specific calculation process, refer to the method embodiment.
  • the comparison processing unit 9022 is configured to determine whether the exchanged multicast label value (current multicast label) is zero, and if it is zero, determine that it is the latest join copy point; or determine the exchange Whether the multicast tag value is equal to the tag identifier value of the egress node, and if equal, it is determined that it is the most recently left copy point.
  • the network nodes shown in Figures 8 and 9 can be the same type of network device, such as a router, so the network device can have the functions of an ingress node, an egress node, and an intermediate node at the same time, and can be set to be functionally selectable, so that the operator can For its own needs, there is no same device in the network, such as a router, which is set as an edge node or an intermediate node according to the location of the device.
  • a router which is set as an edge node or an intermediate node according to the location of the device.
  • the embodiment of the present invention further provides a network resource management node 100, including:
  • the receiving unit 101 is configured to receive the join request of the egress node to the multicast group, or the egress request of the egress node to the multicast group; of course, the receiving unit is further configured to receive the latest join copy point sent by the ingress node to a bandwidth resource reservation request between the egress nodes, or a bandwidth resource release request between the egress node and the egress node;
  • the forwarding unit 102 is configured to forward the received join request of the egress node to the multicast group or the exit request of the egress node to the multicast group to the ingress node;
  • the processing unit 103 is configured to reserve, according to the received bandwidth resource reservation request, a bandwidth resource that is recently added to the replication point to the egress node, or release the content according to the received bandwidth resource release request. Recently leaving the replication point to the bandwidth resource between the egress nodes;
  • the bandwidth allocation confirmation unit 104 is configured to reserve, according to the processing unit, the bandwidth resource between the recently added replication point and the egress node, and send a resource reservation confirmation response to the ingress node; or according to the processing unit Release the bandwidth resource between the nearest leaving replication point to the egress node, and send a resource release confirmation response to the ingress node.
  • a system and a network node device for implementing multicast resource admission control or release control can use the network resource management node to uniformly manage the multicast transmission process, especially when an egress node newly joins or exits.
  • the bandwidth of the most recently added copy point to the egress node is reserved in time by searching for the nearest join copy point corresponding to the egress node or the nearest leave copy point, or releasing the recently left copy point to the The bandwidth between the egress nodes enables dynamic and reasonable allocation of bandwidth resources. Increased network bandwidth utilization.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

A multicast controlling method, device and system, the method including steps: receiving an joining request of the first edge node to a multicasting group, or a leaving request of the first edge node to a multicasting group; when receiving the joining request, searching a latest joining copy node corresponding to the first edge node; when receiving the leaving request, searching a latest leaving copy node corresponding to the first edge node; allocating bandwidth resource between the found latest joining copy node and the first edge node, or releasing bandwidth resource between the found latest joining copy node and the first edge node.

Description

一种组播控制的方法、 设备和系统  Method, device and system for multicast control
本申请要求于 2008 年 3 月 13 日提交中国专利局、 申请号为 200810026778.4、 发明名称为 "一种组播控制的方法、 设备和系统" 的中国专利 申请的优先权, 其全部内容通过引用结合在本申请中。 技术领域 本发明涉及通信技术领域, 尤其涉及一种组播控制的方法、 设备和系统。 背景技术 This application claims priority to Chinese Patent Application No. 200810026778.4, entitled "Method, Apparatus and System for Multicast Control", filed on March 13, 2008, the entire disclosure of which is incorporated by reference. In this application. The present invention relates to the field of communications technologies, and in particular, to a method, device, and system for multicast control. Background technique
目前, 互联网上提供的 TV类业务称为 IPTV, IPTV业务基本地主要有视频 点播(VoD, Video on Demand )和广播电视 ( BTV, Broadcast TV ) 两种形式, 其中 VoD通过单播传送电视节目, BTV业务特指在 IP网络上通过广播或组播 传送的电视节目。 视频业务占用网络带宽资源很多, 当大量的视频节目充斥在 网络上时, 会无规则地争夺网络有限的带宽。 为保证业务的质量, 需要网络对 业务占用资源进行统一管理, 根据带宽资源的空闲状况来决定是否允许某次业 务占用网络带宽或者释放网络带宽, 这个称为资源接纳与释放控制。 现有技术在核心网实施对组播业务的资源接纳与释放控制, 遇到的问题在 于现有的组播协议技术, 是为每一个组播组对应的媒体流建立一棵组播树, 在 核心网络域, 存在巨量的媒体流时, 进行组播路由管理的组播路由器会被大量 的组播路由协议报文消耗掉处理资源, 存贮空间也会被大量的组播状态记录占 据, 网络容量受到这些限制而变得扩容困难, 这个问题称为可扩展性问题。 另外, 现有技术方案中组播传送的过程没有带宽保证机制, 对一个新加入 该组播组的出口节点容易造成等待分配带宽资源的时间过长, 媒体流不能顺畅 传递, 或者对一个要离开组播组的出口节点不能及时释放带宽资源, 容易引起 网络拥塞。 发明内容 基于上述现有技术的不足, 本发明实施例要解决的技术问题在于提供一种 组播控制的方法、 设备和系统, 实现组播带宽资源的统一管理, 及时响应边缘 节点要求加入或者离开组播组的请求, 动态分配带宽资源。 At present, the TV-type service provided on the Internet is called IPTV. The IPTV service basically has two forms of video on demand (VoD, Video on Demand) and broadcast television (BTV, Broadcast TV), in which VoD transmits television programs through unicast. The BTV service specifically refers to a television program transmitted by broadcasting or multicast on an IP network. The video service occupies a lot of network bandwidth resources. When a large number of video programs are flooded on the network, it will randomly compete for the limited bandwidth of the network. To ensure the quality of services, the network needs to manage the resources occupied by the network. According to the idle status of the bandwidth resources, whether to allow a certain service to occupy network bandwidth or release network bandwidth is called resource admission and release control. The existing technology implements the resource admission and release control of the multicast service in the core network. The existing multicast protocol technology is to establish a multicast tree for the media stream corresponding to each multicast group. In the core network domain, when there is a huge amount of media streams, the multicast routers that perform multicast routing management will consume processing resources by a large number of multicast routing protocol packets, and the storage space will be occupied by a large number of multicast state records. Network capacity is difficult to expand due to these limitations. This problem is called scalability. In addition, in the prior art solution, the process of multicast transmission does not have a bandwidth guarantee mechanism. For an egress node that newly joins the multicast group, it is easy to wait for the bandwidth resource to be allocated for a long time, the media stream cannot be smoothly transmitted, or one is to leave. The egress node of a multicast group cannot release bandwidth resources in time, which may cause network congestion. SUMMARY OF THE INVENTION Based on the above deficiencies of the prior art, the technical problem to be solved by the embodiments of the present invention is to provide a multicast control method, device, and system, which implements unified management of multicast bandwidth resources, and responds to an edge node requesting to join or leave in time. A request from a multicast group dynamically allocates bandwidth resources.
本发明实施例提供了一种组播控制的方法, 包括下列步骤: 接收第一边缘 节点对组播组的加入请求、 或第一边缘节点对组播组的离开请求; 当接收到加 入请求时, 利用单播路由向下游节点发送查找报文, 查找与第一边缘节点对应 的最近加入复制点; 或当接收到离开请求时, 利用单播路由查找与第一边缘节 点对应的最近离开复制点; 该查找报文中携带有所述第一边缘节点的标签标识, 所述组播组的组播标签和所述组播组的出口节点的标签标识; 当找到所述最近 加入复制点后, 请求为该最近加入复制点到第一边缘节点之间预留带宽资源, 并在接收到资源确认响应信号后, 修改该组播组的组播标签, 以将该第一边缘 节点加入该组播组; 或者当找到该最近离开复制点后, 请求为该最近离开复制 点到第一边缘节点之间释放带宽资源, 并在接收到资源确认响应信号后, 修改 该组播组的组播标签, 以将该第一边缘节点从该组播组中删除。  An embodiment of the present invention provides a method for multicast control, including the following steps: receiving a request for joining a multicast group by a first edge node, or a request for leaving a multicast group by a first edge node; when receiving a join request Using the unicast route to send a lookup packet to the downstream node to find the nearest join copy point corresponding to the first edge node; or when receiving the leave request, use the unicast route to find the nearest leave copy point corresponding to the first edge node The search packet carries the label identifier of the first edge node, the multicast label of the multicast group, and the label identifier of the egress node of the multicast group; when the recently added copy point is found, Requesting to reserve bandwidth resources for the recently added replication point to the first edge node, and after receiving the resource acknowledgement response signal, modifying the multicast label of the multicast group to join the first edge node to the multicast Group; or when the recently left copy point is found, the request releases the bandwidth resource between the recently left copy point and the first edge node, and is in the connection After receiving the resource confirmation response signal, the multicast label of the multicast group is modified to delete the first edge node from the multicast group.
相应地, 本发明实施例还提供了一种网络节点, 包括:  Correspondingly, an embodiment of the present invention further provides a network node, including:
接收单元, 用于接收第一边缘节点对组播组的加入请求, 或第一边缘节点 对组播组的离开请求; 以及接收资源预留确认响应或资源释放确认响应;  a receiving unit, configured to receive a join request of the first edge node to the multicast group, or a leave request of the first edge node to the multicast group; and receive a resource reservation acknowledgement response or a resource release acknowledgement response;
查找单元, 用于根据接收的对组播组的加入请求, 在该组播组的媒体流的 转发链路上, 查找到第一边缘节点的最近加入复制点; 或者根据接收的对组播 组的离开请求, 在该组播组的媒体流的转发链路上, 查找到第一边缘节点的最 近离开复制点;  a searching unit, configured to: according to the received join request to the multicast group, find the nearest join point of the first edge node on the forwarding link of the media stream of the multicast group; or according to the received pair of multicast groups The leaving request, on the forwarding link of the media stream of the multicast group, find the nearest leaving copy point of the first edge node;
发送单元, 用于发送找到的最近加入复制点到第一边缘节点之间的带宽资 源预留请求, 或发送找到的最近离开复制点到第一边缘节点之间的带宽资源释 放请求;  a sending unit, configured to send a bandwidth resource reservation request between the found nearest joining copy point to the first edge node, or send a bandwidth resource release request between the found nearest leaving copy point to the first edge node;
处理单元,用于在接收到资源预留确认响应后,修改该组播组的组播标签, 以将该第一边缘节点加入该组播组; 或者在接收到资源释放确认响应后, 修改 该组播组的组播标签, 以将该第一边缘节点从该组播组中删除。  a processing unit, configured to: after receiving the resource reservation confirmation response, modify the multicast label of the multicast group to join the first edge node to the multicast group; or after receiving the resource release confirmation response, modify the A multicast label of a multicast group to delete the first edge node from the multicast group.
相应地, 本发明实施例还提供了一种网络节点, 包括:  Correspondingly, an embodiment of the present invention further provides a network node, including:
接收单元, 用于接收查找到请求加入组播组的第一边缘节点的最近加入复 制点的查找报文, 或接收查找到请求离开组播组的第一边缘节点的最近离开复 制点的查找报文, 该查找报文包括第一边缘节点的标签标识和当前接收组播组 的媒体流的网络边缘节点的标签标识, 以及该组播组的组播标签; a receiving unit, configured to receive a recent join of the first edge node that finds the request to join the multicast group The lookup message of the point, or the lookup message of the nearest leaving copy point of the first edge node that requests to leave the multicast group, the lookup message includes the label identifier of the first edge node and the current receiving multicast group. a label identifier of a network edge node of the media stream, and a multicast label of the multicast group;
判断单元, 用于根据接收到的查找报文, 利用单播路由, 计算当前组播标 签, 并判断自身是否为最近加入复制点或最近离开复制点;  a determining unit, configured to calculate, according to the received search message, a current multicast label by using a unicast route, and determine whether it is a recently added copy point or a recently left copy point;
发送单元, 用于在判断结果为是时, 将自身的节点标识发送给该组播组的 入口节点; 在判断结果为否时, 根据单播路由向下游节点转发查找报文, 将查 找报文中的组播标签修改为当前组播标签。  The sending unit is configured to: when the judgment result is yes, send the node identifier of the node to the ingress node of the multicast group; when the judgment result is no, the device forwards the search packet to the downstream node according to the unicast route, and the search packet is searched. The multicast label in the modification is changed to the current multicast label.
当然, 本发明实施例还提供了一种组播控制的系统, 包括:  Certainly, the embodiment of the present invention further provides a system for multicast control, including:
第一边缘节点、 第二边缘节点、 中间节点和网络资源管理节点;  a first edge node, a second edge node, an intermediate node, and a network resource management node;
所述第一边缘节点, 用于发送对组播组的加入请求, 或发送对组播组的离 开请求;  The first edge node is configured to send a join request to the multicast group, or send a leave request to the multicast group;
所述第二边缘节点, 用于接收第一边缘节点对组播组的加入请求, 或接收 第一边缘节点对组播组的离开请求, 根据接收的第一边缘节点对组播组的加入 请求, 在该组播组的媒体流的转发链路上, 发送查找最近加入复制点的查找报 文; 或者根据接收的第一边缘节点对组播组的离开请求, 在该组播组的媒体流 的转发链路上, 发送查找最近离开复制点的查找报文; 该查找报文中携带有第 一边缘节点的标签标识, 该组播组的组播标签和该组播组的所有出口节点的标 签标识; 当找到最近加入复制点后, 向网络资源管理节点请求为该最近加入复 制点到第一边缘节点之间预留带宽资源, 并在接收到所述网络资源管理节点发 送的资源确认响应信号后, 修改该组播组的组播标签, 以将第一边缘节点加入 该组播组; 或者当找到最近离开复制点后, 向网络资源管理节点请求为该最近 离开复制点到第一边缘节点之间释放带宽资源; 并在接收到网络资源管理节点 资源发送确认响应信号后, 修改该组播组的组播标签, 以将所述第一边缘节点 从该组播组中删除;  The second edge node is configured to receive a join request of the first edge node to the multicast group, or receive a leave request of the first edge node to the multicast group, according to the received first edge node joining the multicast group And sending, on the forwarding link of the media stream of the multicast group, a lookup packet for finding a recently added copy point; or, according to the received first edge node leaving the multicast group, the media stream in the multicast group On the forwarding link, the device searches for a lookup packet that has recently left the replication point; the lookup packet carries the label identifier of the first edge node, the multicast label of the multicast group, and all the exit nodes of the multicast group. a label identifier; after finding the nearest join copy point, requesting the network resource management node to reserve a bandwidth resource between the recently added replication point and the first edge node, and receiving the resource confirmation response sent by the network resource management node After the signal is modified, the multicast label of the multicast group is modified to join the first edge node to the multicast group; or when the nearest copy point is found, The network resource management node requests to release the bandwidth resource between the recently left replication point and the first edge node; and after receiving the network resource management node resource sending the acknowledgement response signal, modify the multicast label of the multicast group to The first edge node is deleted from the multicast group;
所述中间节点, 用于接收第二边缘节点发送的查找报文, 根据该查找报文, 利用单播路由, 计算当前组播标签; 并判断该中间节点自身是否为最近加入复 制点或最近离开复制点; 如果是则向第二边缘节点报告; 如果否则向在该组播 组的媒体流的转发链路上的下游节点转发查找报文, 将查找报文中的组播标签 爹改为该当前组播标签; 所述网络资源管理节点, 用于在接收到第二边缘节点发送的为最近加入复 制点到第一边缘节点之间预留带宽资源的请求后, 进行带宽资源预留, 并在预 留成功时向第二边缘节点发送资源确认响应; 或者在接收第二边缘节点发送的 为最近离开复制点到第一边缘节点之间释放带宽资源的请求, 进行带宽资源释 放, 并在资源释放后向第二边缘节点发送资源确认响应。 The intermediate node is configured to receive a lookup packet sent by the second edge node, calculate a current multicast label by using a unicast route according to the lookup packet, and determine whether the intermediate node itself is recently added to the replication point or recently left. Copying the point; if yes, reporting to the second edge node; if otherwise forwarding the lookup packet to the downstream node on the forwarding link of the media stream of the multicast group, changing the multicast label in the lookup message to the Current multicast label; The network resource management node is configured to perform bandwidth resource reservation after receiving a request for reserving bandwidth resources between the recently added replication point and the first edge node sent by the second edge node, and when the reservation is successful Transmitting a resource acknowledgment response to the second edge node; or receiving a request for releasing the bandwidth resource between the recently leaving the replication point and the first edge node sent by the second edge node, releasing the bandwidth resource, and releasing the resource after the resource is released The edge node sends a resource acknowledgment response.
实施本发明的实施例, 具有如下有益效果:  Embodiments of the present invention have the following beneficial effects:
能够实现统一管理组播传送的过程, 尤其是在有出口节点新加入或退出组 播组时, 通过查找所述出口节点对应的最近加入复制点或者最近离开复制点, 及时预留最近加入复制点到所述出口节点之间的带宽, 或者释放最近离开复制 点到所述出口节点之间的带宽, 实现有效合理地分配带宽资源。 提高了网络带 宽利用率。 附图说明  The process of uniformly managing multicast transmission can be implemented, especially when an egress node newly joins or quits a multicast group, and the nearest join copy point is reserved in time by searching for the nearest join copy point corresponding to the egress node or leaving the copy point recently. The bandwidth between the egress nodes, or the bandwidth that has recently left the replication point to the egress node, enables efficient and reasonable allocation of bandwidth resources. Increased network bandwidth utilization. DRAWINGS
图 1是本发明实施例一种实现组播接纳控制的方法原理示意图;  1 is a schematic diagram of a method for implementing multicast admission control according to an embodiment of the present invention;
图 2是本发明实施例一种实现组播释放控制的方法原理示意图;  2 is a schematic diagram of a method for implementing multicast release control according to an embodiment of the present invention;
图 3是本发明实施例一种实现组播接纳控制的方法的网络示意图; 图 4是本发明实施例一种实现组播接纳控制的方法流程图;  3 is a network diagram of a method for implementing multicast admission control according to an embodiment of the present invention; FIG. 4 is a flowchart of a method for implementing multicast admission control according to an embodiment of the present invention;
图 5是本发明实施例一种实现组播释放控制的方法的网络示意图; 图 6是本发明实施例一种实现组播释放控制的方法流程图;  5 is a network diagram of a method for implementing multicast release control according to an embodiment of the present invention; FIG. 6 is a flowchart of a method for implementing multicast release control according to an embodiment of the present invention;
图 7是本发明实施例一种组播控制的系统框图;  7 is a system block diagram of a multicast control according to an embodiment of the present invention;
图 8是本发明实施例一种入口节点的构成示意图;  FIG. 8 is a schematic structural diagram of an entry node according to an embodiment of the present invention; FIG.
图 9是本发明实施例一种中间节点的构成示意图;  9 is a schematic structural diagram of an intermediate node according to an embodiment of the present invention;
图 10是本发明实施例一种网络资源管理节点的构成示意图。 具体实施方式 本发明实施例提供了一种实现组播资源接纳与释放控制的方法, 是利用多 协议标签交换(MPLS, Multi-Protocol Label Switching )组播标签来标识组播才艮 文需要到达的所有网络出口, 一个带有 MPLS组播标签的报文就可以在网络中 任何一个节点标识出报文当前所指向的出口, 组播标签随着转发的过程而变化, 当从一个节点的一个接口发送到特定出口时, 此节点的其它接口发送的报文组 播标签不会再包含该特定出口的标识, 从而自动避免转发环路。 FIG. 10 is a schematic structural diagram of a network resource management node according to an embodiment of the present invention. The embodiments of the present invention provide a method for implementing multicast resource admission and release control, which uses a Multi-Protocol Label Switching (MPLS) label to identify a multicast packet to be reached. All network egresses, a packet with an MPLS multicast label can identify the egress that the packet is currently pointing to at any node in the network. The multicast label changes with the process of forwarding, when an interface from a node Packet group sent by other interfaces of this node when sent to a specific exit The broadcast tag no longer contains the identity of the particular exit, thus automatically avoiding forwarding loops.
首先对本发明实施例的实现组播接纳控制的方法, 在原理思路上进行说明。 参见图 1, 在一个由节点 (1 )到 (11 )组成网络域中, 有一个组播组 <g,S>的媒 体流按当前箭头连线的转发路径传递, 其中该组播组< 8>的入口节点是节点 First, the method for implementing multicast admission control in the embodiment of the present invention is described in the principle. Referring to FIG. 1, in a network domain composed of nodes (1) to (11), a media stream having a multicast group <g, S> is transmitted according to a forwarding path of a current arrow connection, where the multicast group < 8 The entry node of > is the node
(1 ), 出口节点是节点 (8)和(11 ), 当一个新出口节点 (7)需要加入该组播 组时, 必须找到和现有媒体流转发链路上, 路径最短的节点连接, 从而建立一 条新的组播转发路径。 (1), the egress node is the nodes (8) and (11). When a new egress node (7) needs to join the multicast group, it must find the node with the shortest path on the existing media stream forwarding link. Thereby establishing a new multicast forwarding path.
在图 1中, 所有当前已收到媒体流的节点中, 只有节点 (4)和新出口节点 (7) 的单播转发路径最短, 从而判断节点 (4)是最近加入复制点。  In Figure 1, among all the nodes that have received the media stream, only the node (4) and the new egress node (7) have the shortest unicast forwarding path, so that the node (4) is the closest join point.
所述最近加入复制点, 是网络域中的一个节点, 并同时符合以下三个条件: a) 该节点在有新出口节点请求加入组播组时已收到所述组播组的媒体流; b) 该节点位于从组播组对应的网络入口节点到所述加入组播组的新出口节 点的当前单播转发路径上;  The recently added replication point is a node in the network domain and meets the following three conditions: a) The node has received the media stream of the multicast group when a new egress node requests to join the multicast group; b) the node is located on the current unicast forwarding path from the network ingress node corresponding to the multicast group to the new egress node joining the multicast group;
c) 该节点到所述新出口节点的单播转发路径是最短的。  c) The unicast forwarding path of the node to the new egress node is the shortest.
接着对本发明实施例的实现组播释放控制的方法, 在原理思路上进行说明。 参见图 2, 在一个由节点 ( 1 )到 ( 11 )组成网络域中, 有一个组播组 <g,s>的媒 体流按当前箭头连线的转发路径传递, 其中该组播组< 8>的入口节点是节点 (1 ), 出口节点是节点 (7)、 (8)、 (9)、 (11 ), 当一个出口节点 (8)需要离开 该组播组时, 必须找到和现有媒体流转发链路上, 路径最短的组播转发分支节 点, 从而释放该组播转发分支节点到所述出口节点 (8)之间的带宽资源。  The method for implementing multicast release control in the embodiment of the present invention is described in the principle. Referring to FIG. 2, in a network domain composed of nodes (1) to (11), a media stream having a multicast group <g, s> is transmitted according to a forwarding path of a current arrow connection, where the multicast group < 8 The entry node of > is node (1), and the egress node is node (7), (8), (9), (11). When an egress node (8) needs to leave the multicast group, it must find and existing On the media stream forwarding link, the shortest multicast forwarding branch node releases the bandwidth resource between the multicast forwarding branch node and the egress node (8).
在图 2中, 所有当前已收到媒体流的组播转发分支节点中, 只有节点 (4) 到出口节点 (8) 的单播转发路径最短, 从而判断组播转发节点 (4)是最近离 开复制点。 虽然节点 (4)到出口节点 (8)之间还有节点 (5), 但节点 (5)不 是当前已收到媒体流的组播转发分支节点, 即转发媒体流的枢纽节点 (在一个 组播组中转发媒体流到不止一个下游节点上的节点), 所以节点 (5) 不是最近 离开复制点。  In Figure 2, among all the multicast forwarding branch nodes that have received the media stream, only the unicast forwarding path from the node (4) to the egress node (8) is the shortest, so that the multicast forwarding node (4) is recently left. Copy the point. Although there is a node (5) between the node (4) and the egress node (8), the node (5) is not a multicast forwarding branch node that has received the media stream currently, that is, a hub node that forwards the media stream (in a group) The broadcast group forwards the media stream to the node on more than one downstream node), so the node (5) is not the closest to the replication point.
所述最近离开复制点, 是网络域中的一个组播转发分支节点, 并同时符合 以下三个条件:  The recently left replication point is a multicast forwarding branch node in the network domain and meets the following three conditions:
a) 该节点在有出口节点请求离开组播组时已收到所述组播组的媒体流; b) 该节点位于从组播组对应的网络入口节点到所述离开组播组的出口节点 的当前单播转发路径上; a) the node has received the media stream of the multicast group when the egress node requests to leave the multicast group; b) the node is located from the network ingress node corresponding to the multicast group to the egress node of the leaving multicast group On the current unicast forwarding path;
C) 该节点到所述离开组播组的出口节点的组播转发路径上没有其它组播转 发分支节点。  C) There is no other multicast forwarding branch node on the multicast forwarding path of the node to the egress node leaving the multicast group.
如图 3所示, 为本发明实施例提供的一种实现组播资源接纳控制方法的网 络示意图。  As shown in FIG. 3, it is a schematic diagram of a network for implementing a multicast resource admission control method according to an embodiment of the present invention.
在基于正交标签的网络域中, 为标识出每一个边缘节点, 并把标识映射到 标签中, 需要为每一个边缘节点定义一个标签标识, 例如为边缘节点八、 B、 C、 D定义标签标识分别为 0x00001、 0x00002、 0x00004、 0x00008, 所有边缘节点 的标签标识正交, 即任意两个标签标识相与的结果都为 0。 有一个组播组 <g,s> 的媒体流依次从节点 B、 Y、 Z、 D的路径传递, 其中该组播组 <g,s>的入口节点 是节点 B, 出口节点是节点 D; 组播标签为组播组的各出口节点标签标识相加, 即该组播组 <g,s>的组播标签为出口节点 D的标签标识 0x00008。  In the orthogonal label-based network domain, in order to identify each edge node and map the identifier to the label, a label identifier needs to be defined for each edge node, for example, defining labels for edge nodes eight, B, C, and D. The identifiers are 0x00001, 0x00002, 0x00004, 0x00008, and the label identifiers of all edge nodes are orthogonal, that is, the result of any two label identifiers is 0. The media stream of a multicast group <g, s> is sequentially transmitted from the path of the node B, Y, Z, D, wherein the ingress node of the multicast group <g, s> is the node B, and the egress node is the node D; The multicast label is the label of each egress node of the multicast group. That is, the multicast label of the multicast group <g, s> is the label ID 0x00008 of the egress node D.
结合图 4所示, 当一个新出口节点 A需要加入该组播组时, 实现组播接纳 控制的方法步骤如下:  As shown in Figure 4, when a new egress node A needs to join the multicast group, the method for implementing multicast admission control is as follows:
101、 加入组播组 < 8>的请求到达出口节点 A。  101. The request to join the multicast group <8> arrives at the exit node A.
102、 出口节点 A将携带有自己标签标识的加入组播组的请求发给网络资 源管理器 RM, RM集中保存有对应于组播组 < 8>的所有入口节点信息。  102. The egress node A sends a request for joining the multicast group with the own tag identifier to the network resource manager RM, and the RM centrally stores all the ingress node information corresponding to the multicast group <8>.
103、 网络资源管理器 RM查找到对应于组播组 <g,s>的入口节点 B后, 将 出口节点 A加入组播组的请求转发给入口节点 B。  103. After the network resource manager finds the ingress node B corresponding to the multicast group <g, s>, the RM forwards the request of the egress node A to the multicast group to the ingress node B.
出口节点 A加入组播组 <g,s>的请求通过网络资源管理器 RM转发给入口 节点 B, 这在形式上是一个目的地址为组播地址的五元组表示的流。 当然, 该加 入组播组的请求还可以由 RM通过域间接口从其它 RM收到, 再转发给入口节 点 B; 或者由节点 A直接通过单播发送给入口节点 B。  The request of the egress node A to join the multicast group <g, s> is forwarded to the ingress node B through the network resource manager RM, which is formally a stream represented by a quintuple whose destination address is a multicast address. Of course, the request to join the multicast group may also be received by the RM from the other RM through the inter-domain interface, and then forwarded to the ingress node B; or sent directly by the node A to the ingress node B through unicast.
104、 入口节点 B 收到这个加入组播组 <g,s>的请求后, 利用单播路由发送 一个查找最近加入复制点的报文, 该查找报文中携带请求加入组播组的出口节 点 A的标签标识 0x00001和当前已接收到该组播组 < 8>媒体流的出口节点的标 签标识, 即当前出口节点 D的标签标识 0x00008 , 以及组播组< 8>的组播标签 0x00008。  After receiving the request to join the multicast group <g, s>, the ingress node B sends a unicast route to send a packet to the nearest join point, where the lookup packet carries the egress node requesting to join the multicast group. The tag identifier of A is 0x00001 and the tag identifier of the egress node that has received the multicast group <8> media stream, that is, the tag identifier 0x00008 of the current egress node D, and the multicast tag 0x00008 of the multicast group <8>.
入口节点 B发送一个寻找最近加入复制点的查找报文, 发送给下游节点, 让所有已收到媒体流的节点感知处理, 所以该查找报文中还携带的信息有组播 <g,s>的入口节点 B的标签标识 0x00002。 The ingress node B sends a lookup message for finding the most recently added copy point, and sends it to the downstream node to let all the nodes that have received the media stream perceive the processing. Therefore, the information carried in the search packet is also multicast. The tag of the ingress node B of <g,s> is 0x00002.
105、 下游中间节点 Y收到所述查找报文, 判断自己是否是到出口节点 A 的最近加入复制点。 具体地, 中间节点 Y在其每个出接口上通过单播路由计算 到所有出口节点的可达性, 计算各个出接口标签掩码。 一个出接口上单播可达 的所有出口节点标签标识之和, 组成该出接口的标签掩码。 由于查找报文中携 带有组播标签, 中间节点 Y根据当前单播路由可得知到出口节点 A要经过中间 节点 Z。中间节点 Y将该标签掩码 0x00009与查找 4艮文中携带组播标签 0x00008 相与, 得到中间节点 Y到出口节点 A的出口标签为 0x00008, 该出口标签即是 经过中间节点 Y交换标签后得到的当前组播标签。 因为到出口节点 A的单播转 发路径上还有中间节点 Z在转发媒体流给出口节点 D, 所以计算得到的到出口 节点 A方向的当前组播标签不为零,说明中间节点 Y到出口节点 A的单播转发 路径不是最短的。 这就能判断中间节点 Y不是最近加入复制点。 中间节点 Y因 此继续将该查找报文转发给下游中间节点 Z,该查找报文中的组播标签为当前组 播标签 0x00008。 其中, 中间节点 Y根据查找报文中的组播标签确定其出口节 点方向为节点 D, 并计算节点 D方向出接口的标签掩码, 该出接口标签掩码为 出口节点(与该出口节点是否在该组播组中无关)标签标识之和, 即出口节点 A 与 D的标签标识之和 0x00009。  105. The downstream intermediate node Y receives the lookup packet, and determines whether it is the closest join copy point to the exit node A. Specifically, the intermediate node Y calculates the reachability of all the egress nodes by using the unicast route on each of the egress interfaces, and calculates the label of each egress interface label. The sum of all egress node label identifiers that are reachable on the outbound interface, and the label mask of the outbound interface. Since the lookup packet carries a multicast label, the intermediate node Y can know that the egress node A passes through the intermediate node Z according to the current unicast route. The intermediate node Y associates the label mask 0x00009 with the multicast label 0x00008 in the lookup 4, and obtains the exit label of the intermediate node Y to the egress node A as 0x00008, and the egress label is obtained after exchanging labels through the intermediate node Y. Current multicast label. Because the intermediate node Z is forwarding the media stream to the egress node D on the unicast forwarding path of the egress node A, the calculated current multicast label in the direction of the egress node A is not zero, indicating that the intermediate node Y to the egress node The unicast forwarding path of A is not the shortest. This can be judged that the intermediate node Y is not recently added to the copy point. The intermediate node Y then forwards the lookup packet to the downstream intermediate node Z. The multicast label in the lookup packet is the current multicast label 0x00008. The intermediate node Y determines the direction of the egress node as the node D according to the multicast label in the lookup packet, and calculates the label mask of the outbound interface in the direction of the node D. The egress interface label is the egress node (whether or not the egress node The sum of the tag identifiers in the multicast group, that is, the sum of the tag identifiers of the egress nodes A and D 0x00009.
106、 中间节点 Z收到该查找报文后,也判断自己是否是到出口节点 A的最 近加入复制点。 具体地, 中间节点 Z通过单播路由计算出到出口节点 A出接口 的标签掩码为 0x00001 , 将该标签掩码 0x00001 与收到的上述当前组播标签 0x00008相与, 得到中间节点 Z到出口节点 A的出口标签为零, 该出口标签即 是经过中间节点 Z交换标签后得到的当前组播标签, 因为到出口节点 A方向的 当前组播标签为零, 这说明中间节点 Z到出口节点 A的单播转发路径上没有媒 体流在转发, 从而确定中间节点 Z是最近加入复制点。 中间节点 Z根据入口节 点 B的标签标识, 或者其它标识, 发送一个单播响应信号返回给入口节点 B , 报告自己是本次查找的最近加入复制点。  106. After receiving the lookup message, the intermediate node Z also determines whether it is the closest join point to the exit node A. Specifically, the intermediate node Z calculates the label mask of the egress interface of the egress node A by using a unicast route to be 0x00001, and the label mask 0x00001 is matched with the received current multicast label 0x00008 to obtain the intermediate node Z to the egress. The egress label of node A is zero. The egress label is the current multicast label obtained after the intermediate node Z exchanges labels. The current multicast label to the egress node A is zero, which indicates that the intermediate node Z to the egress node A No media stream is forwarded on the unicast forwarding path, thereby determining that the intermediate node Z is the closest join copy point. The intermediate node Z sends a unicast response signal to the ingress node B according to the label identifier of the ingress node B, or other identifiers, and reports that it is the most recently added copy point of the current search.
107、 入口节点 B收到单播响应信号后, 向网络资源管理器 RM发送资源 接纳请求, 请求预留从中间节点 Z到出口节点 A的组播带宽资源。  107. After receiving the unicast response signal, the ingress node B sends a resource admission request to the network resource manager RM, requesting to reserve the multicast bandwidth resource from the intermediate node Z to the egress node A.
中间节点 Z也可以直接向网络资源管理器 RM报告自己是本次查找的最近 加入复制点, 并发送资源接纳请求, 请求 RM预留从中间节点 Z到出口节点 A 的组播带宽资源。 The intermediate node Z can also directly report to the network resource manager RM that it is the most recently added replication point of the current lookup, and sends a resource admission request, requesting the RM reservation from the intermediate node Z to the exit node A. Multicast bandwidth resources.
108、 网络资源管理器 RM完成资源预留, 给出资源确认响应信号。  108. Network Resource Manager The RM completes resource reservation and gives a resource confirmation response signal.
109、 入口节点 B收到资源确认响应信号后, 更改对应组播组<&8>的组播 标签, 将新加入的所述出口节点 A也作为组播组 < 8>媒体流的一个出口节点。 这样组播组媒体流就会从中间节点 Z转发到出口节点 A上去。  109. After receiving the resource confirmation response signal, the ingress node B changes the multicast label corresponding to the multicast group <&8>, and also uses the newly added egress node A as an egress node of the multicast group <8> media stream. Thus, the multicast group media stream is forwarded from the intermediate node Z to the egress node A.
相应的, 如图 5所示, 为本发明实施例还提供的一种实现组播资源释放控 制方法的网络示意图。  Correspondingly, as shown in FIG. 5, it is a schematic diagram of a network for implementing a multicast resource release control method according to an embodiment of the present invention.
在基于正交标签的核心网中, 为标识出每一个边缘节点, 并把标识映射到 标签中, 需要为每一个边缘节点定义一个标识, 例如预先为边缘节点 A、 B、 C、 D定义标签标识分别为 0x00001、 0x00002、 0x00004、 0x00008, 不同边缘节点 的标签标识正交, 即任意两个标签标识相与的结果都为 0。 有一个组播组 <g,s> 的媒体流依次从节点 B、 Y、 Z、 D的路径和从节点:8、 Υ、 Ζ、 Α的路径传递, 其中该组播组< 8>的入口节点是节点 B , 出口节点是节点 D和 A, 那么该组播 组的组播标签为出口节点 A和 D标签标识的和, 即组播组< 8>的组播标签为 0x00009。  In the orthogonal label-based core network, in order to identify each edge node and map the identifier to the label, an identifier needs to be defined for each edge node, for example, a label is defined for the edge nodes A, B, C, and D in advance. The identifiers are 0x00001, 0x00002, 0x00004, 0x00008, and the label identifiers of different edge nodes are orthogonal, that is, the result of any two label identifiers is 0. The media stream with a multicast group <g, s> is sequentially transmitted from the path of the node B, Y, Z, D and the path from the node: 8, Υ, Ζ, Α, where the entry of the multicast group <8> The node is the node B, and the egress node is the node D and the A. The multicast label of the multicast group is the sum of the identifiers of the egress node A and the D tag, that is, the multicast tag of the multicast group <8> is 0x00009.
再结合图 6所示, 当一个出口节点 A需要离开该组播组时, 实现组播释放 控制的方法步骤如下:  Referring to FIG. 6, when an egress node A needs to leave the multicast group, the method for implementing multicast release control is as follows:
201、 离开组播组 <g,s>的请求到达出口节点 A。  201. The request to leave the multicast group <g, s> reaches the exit node A.
202、 出口节点 A将携带有自己标签标识的离开组播组的请求发给网络资 源管理器 RM, RM集中保存有对应于组播组 < 8>的所有入口节点信息。  202. The egress node A sends a request to leave the multicast group with the own tag identifier to the network resource manager RM, where the RM stores all the ingress node information corresponding to the multicast group.
203、 网络资源管理器 RM查找到对应于组播组 <g,s>的入口节点 B后, 将 出口节点 A的离开组播组的请求发给入口节点 B。  203. The network resource manager, after finding the ingress node B corresponding to the multicast group <g, s>, sends the request of the egress node A to leave the multicast group to the ingress node B.
出口节点 A的离开组播组 <g,s>的请求通过网络资源管理器 RM转发给入 口节点 B,这在形式上可以是一个目的地址为组播地址的五元组表示的流。当然, 该离开组播请求还可以由 RM通过域间接口从其它 RM收到, 再转发给入口节 点 B; 或者由出口节点 A直接通过单播发送给入口节点 B。  The request of the egress node A to leave the multicast group <g, s> is forwarded to the ingress node B through the network resource manager RM, which may be a stream represented by a quintuple whose destination address is a multicast address. Of course, the leaving multicast request may also be received by the RM from the other RM through the inter-domain interface, and then forwarded to the ingress node B; or sent by the egress node A to the ingress node B by unicast.
204、 入口节点 B收到这个离开组播组 <g,s>的请求后, 利用单播路由发送 一个查找最近离开复制点的报文, 该查找报文中携带所述出口节点 A的标签标 识 0x00001 和当前已接收到该组播组 <g,s>媒体流的出口节点 D 的标签标识 0x00008, 以及组播组 <g,s>的组播标签 0x00009。 入口节点 B发送一个寻找最近离开复制点的查找报文, 发送给下游节点, 让所有已收到媒体流的节点感知处理, 所以该查找报文中还携带的信息有对应 于组播组< 8>的入口节点 B的标签标识。 After receiving the request to leave the multicast group <g, s>, the ingress node B sends a packet to the nearest leaving replication point by using the unicast route, where the lookup packet carries the label identifier of the egress node A. 0x00001 and the label ID 0x00008 of the egress node D that has received the multicast group <g, s> media stream, and the multicast label 0x00009 of the multicast group <g, s>. The ingress node B sends a lookup message for the nearest leaving of the replication point, and sends it to the downstream node, so that all the nodes that have received the media stream are aware of the processing, so the information carried in the search packet also corresponds to the multicast group < 8 > The tag identifier of the ingress node B.
205、 下游中间节点 Y收到所述查找报文, 判断自己是否是到出口节点 A 的最近离开复制点。 具体地, 由于查找报文中携带有组播标签, 中间节点 Y根 据当前单播路由可得知到出口节点 A要经过中间节点 Z, 而到中间节点 Z方向 出接口的标签掩码为 0x00009, 该标签掩码即为出口节点 A与 D标签标识的和 值。 其中, 中间节点 Y在其每个出接口上通过单播路由计算到所有出口节点的 可达性, 计算各个出接口的标签掩码。 一个出接口上单播可达的所有出口节点 标签标识之和, 组成该出接口的标签掩码。 中间节点 Y根据该组播标签确定其 出口节点方向为节点 D, 并计算节点 D方向出接口的标签掩码, 该出接口标签 掩码为出口节点 (与该出口节点是否在该组播组中无关)标签标识之和, 即出 口节点 A与 D的标签标识之和 0x00009。  205. The downstream intermediate node Y receives the lookup packet, and determines whether it is the nearest leaving replication point to the egress node A. Specifically, the lookup packet carries the multicast label, and the intermediate node Y learns that the egress node A passes through the intermediate node Z according to the current unicast route, and the label mask of the outbound interface in the Z direction of the intermediate node is 0x00009. The label mask is the sum of the IDs of the egress node A and the D tag. The intermediate node Y calculates the reachability of all egress nodes through unicast routes on each outbound interface, and calculates the label mask of each egress interface. The sum of the label IDs of all egress nodes that are reachable on the outbound interface, and the label mask of the egress interface. The intermediate node Y determines the direction of the egress node as the node D according to the multicast label, and calculates the label mask of the outbound interface in the direction of the node D. The outbound interface label is the egress node (whether the egress node is in the multicast group or not) Irrelevant) The sum of the tag identifiers, that is, the sum of the tag identifiers of the exit nodes A and D 0x00009.
再将该标签掩码 0x00009与查找 文中携带的组播标签 0x00009相与, 得 到中间节点 Y到出口节点 A的出口标签为 0x00009。 该出口标签即是经过中间 节点 Y交换标签后得到的当前组播标签, 因为到出口节点 A的路径上还有组播 转发分支节点 Z在转发媒体流给出口节点 D, 所以计算得到的当前组播标签为 两个出口节点 A与 D标签标识的和值,这就能判断中间节点 Y不是最近离开复 制点。 中间节点 Y将查找报文转发给下游中间节点 Z, 该查找报文中携带的组 播标签为当前组播标签 0x00009。  Then, the label mask 0x00009 is matched with the multicast label 0x00009 carried in the search text, and the exit label of the intermediate node Y to the exit node A is 0x00009. The egress label is the current multicast label obtained after the label is exchanged by the intermediate node Y. Because the path to the egress node A also has the multicast forwarding branch node Z forwarding the media stream to the egress node D, the current group is calculated. The broadcast tag is the sum of the two exit node A and D tag identifiers, which can determine that the intermediate node Y is not the closest to the copy point. The intermediate node Y forwards the lookup packet to the downstream intermediate node Z. The multicast label carried in the lookup packet is the current multicast label 0x00009.
206、 中间节点 Z是组播转发分支节点, 它收到查找报文后, 判断自己是 否是到出口节点 A的最近离开复制点。 具体地, 中间节点 Z通过单播路由计算 出到出口节口 A的出接口的标签掩码为 0x00001 , 将该标签掩码与收到的上述 当前组播标签 0x00009相与, 得到的出口标签为 0x00001, 该出口标签 0x00001 即是经过中间节点 Z交换标签后得到的当前组播标签, 因为中间节点 Z到出口 节点 A方向的当前组播标签的值等于出口节点 A的标签标识值。 这说明在中间 节点 Z和出口节点 A之间没有其它组播转发分支节点在转发媒体流, 从而确定 中间节点 Z是最近离开复制点。 中间节点 Z根据入口节点 B的标签标识, 发送 一个单播响应信号返回给入口节点 B, 报告自己是本次查找的最近离开复制点。  206. The intermediate node Z is a multicast forwarding branch node. After receiving the lookup packet, it determines whether it is the closest leaving replication point to the egress node A. Specifically, the intermediate node Z calculates a label mask of the outbound interface to the exit node A by using a unicast route to be 0x00001, and the label mask is matched with the received current multicast label 0x00009, and the obtained exit label is 0x00001, the egress label 0x00001 is the current multicast label obtained after the intermediate node Z exchanges labels, because the value of the current multicast label in the direction from the intermediate node Z to the egress node A is equal to the label identification value of the egress node A. This means that no other multicast forwarding branch nodes are forwarding the media stream between the intermediate node Z and the egress node A, thereby determining that the intermediate node Z is the closest leaving the replication point. The intermediate node Z sends a unicast response signal to the ingress node B according to the label identifier of the ingress node B, and reports that it is the nearest leaving copy point of the current lookup.
207、 入口节点 B收到单播响应信号后, 向网络资源管理器 RM发送资源 释放请求, 请求释放从中间节点 Ζ到出口节点 Α的组播带宽资源。 207. After receiving the unicast response signal, the ingress node B sends the resource to the network resource manager RM. The release request requests to release the multicast bandwidth resource from the intermediate node to the egress node.
中间节点 Z也可以直接向网络资源管理器 RM报告自己是本次查找的最近 离开复制点, 并发送资源接纳请求, 请求 RM释放从中间节点 Z到出口节点 A 的组播带宽资源。  The intermediate node Z can also directly report to the network resource manager RM that it is the nearest leaving replication point of the current lookup, and send a resource admission request, requesting the RM to release the multicast bandwidth resource from the intermediate node Z to the egress node A.
208、 网络资源管理器 RM完成资源释放后, 给出资源确认响应信号。 208. The network resource manager sends a resource confirmation response signal after the RM completes the resource release.
209、 入口节点 B收到资源确认响应信号后, 更改对应组播组<&8>的组播 发送标签, 将所述出口节点 A从组播组<^8>中删除。 这样中间节点 Z就不再 向出口节点 A转发组播组 <g,s>的媒体流。 209. After receiving the resource confirmation response signal, the ingress node B changes the multicast sending label of the corresponding multicast group <&8>, and deletes the egress node A from the multicast group <^8>. Thus, the intermediate node Z no longer forwards the media stream of the multicast group <g, s> to the egress node A.
釆用本发明实施例提供的一种实现组播资源接纳控制的方法或者一种实现 组播资源释放控制的方法。 能利用网络资源管理节点统一管理组播传送的过程, 尤其是在有出口节点新加入或退出组播组时, 通过查找所述出口节点对应的最 近加入复制点或者最近离开复制点, 及时预留最近加入复制点到所述出口节点 之间的带宽, 或者及时释放最近离开复制点到所述出口节点之间的带宽, 实现 动态合理地分配带宽资源。 提高了网络带宽利用率。  A method for implementing multicast resource admission control or a method for implementing multicast resource release control provided by the embodiment of the present invention. The network resource management node can be used to uniformly manage the multicast transmission process, especially when an egress node newly joins or exits the multicast group, and finds the nearest join copy point corresponding to the egress node or recently leaves the replication point, and reserves in time. The bandwidth of the replication point to the egress node is recently added, or the bandwidth between the egress node and the egress node is released in time to realize dynamic and reasonable allocation of bandwidth resources. Increased network bandwidth utilization.
此外, 本实施例中各节点的组播转发过程不依赖组播路由协议, 而是将一 个网络域边缘节点进行标识, 把针对一个组播组的转发树在该网络域的拓朴用 从入口节点到出口节点的转发拓朴来描述。 这样, 所有在该网络域具有相同转 发拓朴的组播组在网络域执行无分别的组播传送, 从而实现了组播会聚。 对要 到达的网络边缘节点的标识统一放在一个组播标签中由穿越该网络域的组播报 文携带, 节点利用组播标签信息即可实现组播转发。 基于这种技术, 中间节点 不需要维护每个组播组的转发状态, 减少了大量的管理报文交互传递。  In addition, in this embodiment, the multicast forwarding process of each node does not depend on the multicast routing protocol, but identifies a network domain edge node, and uses the forwarding tree for a multicast group in the topology of the network domain from the portal. The forwarding topology of the node to the egress node is described. In this way, all multicast groups with the same forwarding topology in the network domain perform no separate multicast transmission in the network domain, thereby realizing multicast convergence. The identifier of the network edge node to be reached is uniformly placed in a multicast label and carried by the multicast packet traversing the network domain. The node can implement multicast forwarding by using the multicast label information. Based on this technology, the intermediate nodes do not need to maintain the forwarding state of each multicast group, which reduces the number of management packets.
为实现以上组播控制的方法, 如图 7 所示, 本发明实施例还提供了一种组 播控制的系统, 在网络域中包括: 网络资源管理节点、 入口节点、 中间节点和 出口节点。 所述入口节点、 中间节点和出口节点在硬件上都可以是路由器, 只 是实现的功能不同。  To implement the above multicast control method, as shown in FIG. 7, the embodiment of the present invention further provides a multicast control system, where the network domain includes: a network resource management node, an ingress node, an intermediate node, and an egress node. The ingress node, the intermediate node, and the egress node may all be routers in hardware, but the functions implemented are different.
其中, 出口节点用于发送加入组播组或离开组播组的请求; 具体的, 所述 出口节点分析用户接收组播组的媒体流的需求情况, 通过发送单元发送对组播 组的加入请求, 或对组播组的离开请求;  The egress node is configured to send a request to join the multicast group or leave the multicast group. Specifically, the egress node analyzes the requirement of the media stream that the user receives the multicast group, and sends a join request to the multicast group by using the sending unit. , or a request to leave a multicast group;
所述入口节点, 用于根据接收到的出口节点加入组播组的请求, 向该组播 组的下游节点发送查找报文, 查找与该出口节点之间转发路径最短的最近加入 复制点; 或者根据接收到的出口节点离开组播组的请求, 向该组播组的下游节 点发送查找报文, 查找与该出口节点之间转发路径最短的最近离开复制点; 该 查找 4艮文中携带有该组播组的组播标签和该组播组的所有出口节点标签标识, 以及请求加入或离开的出口节点的标签标识; 在查找到最近加入复制点后, 请 求为该出口节点到最近加入复制点之间预留资源, 或在查找到最近离开复制点 后, 请求为该出口节点到最近离开复制点之间释放资源; 并在收到资源预留确 认响应后, 爹改该组播组的组播标签, 以将该请求加入组播组的出口节点加入 到该组播组; 或者在收到资源释放确认响应后, 修改该组播组的组播标签, 以 将该请求离开组播组的出口节点从该组播组中删除。 The ingress node is configured to send a lookup packet to the downstream node of the multicast group according to the received request that the egress node joins the multicast group, and find the latest join with the shortest forwarding path between the egress node. Copying a point; or sending a lookup packet to a downstream node of the multicast group according to the received request that the egress node leaves the multicast group, and searching for the closest leaving copy point with the shortest forwarding path between the egress node; The packet carries the multicast label of the multicast group and all the egress node label identifiers of the multicast group, and the label identifier of the egress node that requests to join or leave; after finding the nearest join point, the request is for the egress node to Reserving resources between recently joined replication points, or requesting to release resources between the egress node and the nearest leaving replication point after finding the nearest leaving replication point; and tampering with the group after receiving the resource reservation confirmation response The multicast label of the multicast group is added to the egress group of the multicast group to join the multicast group; or after receiving the resource release confirmation response, modify the multicast label of the multicast group to leave the request The egress node of the multicast group is deleted from the multicast group.
具体的, 所述入口节点, 包括:  Specifically, the ingress node includes:
接收单元, 用于接收所述出口节点对组播组的加入请求, 或所述出口节点 对组播组的离开请求;  a receiving unit, configured to receive a request for joining the multicast group by the egress node, or a request for leaving the multicast group by the egress node;
查找单元, 用于根据接收的所述出口节点对组播组的加入请求, 在所述组 播组的媒体流的转发链路上, 查找与所述出口节点之间转发路径最短的最近加 入复制点; 或者根据接收的所述出口节点对组播组的离开请求, 在所述组播组 的媒体流的转发链路上, 查找与所述出口节点之间转发路径最短的最近离开复 制点;  a searching unit, configured to: according to the received join request of the egress node to the multicast group, search for the shortest recent join replication with the egress node on the forwarding link of the media stream of the multicast group Or: according to the received leaving request of the egress group to the multicast group, on the forwarding link of the media stream of the multicast group, searching for the closest leaving copy point with the shortest forwarding path between the egress node;
发送单元, 用于发送所述查找到的最近加入复制点到所述出口节点之间的 带宽资源预留请求, 或发送所述查找到的最近离开复制点到出口节点之间的带 宽资源释放请求至所述网络资源管理节点;  a sending unit, configured to send the discovered bandwidth resource reservation request between the recently added replication point to the egress node, or send the discovered bandwidth resource release request between the recently left replication point to the egress node To the network resource management node;
还包括处理单元, 用于在收到资源预留确认响应后, 修改该组播组的组播 标签, 以将该请求加入组播组的出口节点加入到该组播组; 或者在收到资源释 放确认响应后, 修改该组播组的组播标签, 以将该请求离开组播组的出口节点 从该组播组中删除。  The processing unit is further configured to: after receiving the resource reservation confirmation response, modify the multicast label of the multicast group to join the egress group of the request to the multicast group to join the multicast group; or receive the resource After the acknowledgment response is released, the multicast label of the multicast group is modified, and the egress node that leaves the multicast group is deleted from the multicast group.
其中以上所述查找单元包括:  The search unit described above includes:
报文生成单元, 用于生成查找报文, 所述查找报文包括所述出口节点的标 签标识、 当前接收所述组播组的媒体流的所有网络边缘节点的标签标识和组播 组的组播标签; 还可以包括该入口节点自身的标签标识;  a packet generating unit, configured to generate a lookup packet, where the lookup packet includes a label identifier of the egress node, a label identifier of all network edge nodes that currently receive the media stream of the multicast group, and a group of the multicast group Broadcast label; may also include the label identifier of the entry node itself;
报文发送单元, 用于发送所述查找报文至所述组播组媒体流转发链路上的 节点; 响应接收单元, 用于接收所述最近加入复制点的节点标识, 或所述最近离 开复制点的节点标识。 a message sending unit, configured to send the lookup message to a node on the multicast group media stream forwarding link; The response receiving unit is configured to receive the node identifier of the most recently joined replication point, or the node identifier of the most recently left replication point.
系统中所述中间节点, 用于根据接收的查找报文, 利用单播路由, 计算当 前组播标签; 并判断该中间节点自身是否为最近加入复制点或最近离开复制点; 如果是则向该组播组的入口节点报告, 如果否则向该组播组的媒体流转发链路 上的下游节点转发查找报文, 将查找报文中的组播标签修改为当前组播标签。  The intermediate node in the system is configured to calculate a current multicast label by using a unicast route according to the received search packet, and determine whether the intermediate node itself joins the recently copied point or recently leaves the copy point; if yes, the The ingress node of the multicast group reports that if the downstream node on the media stream forwarding link of the multicast group forwards the lookup packet, the multicast label in the lookup packet is modified to the current multicast label.
所述中间节点包括:  The intermediate node includes:
接收单元, 用于接收查找最近加入复制点或最近离开复制点的查找报文, 所述查找报文包括请求加入组播组的出口节点的标签标识和当前接收所述组播 组媒体流的网络边缘节点的标签标识, 或包括请求离开组播组的出口节点的标 签标识和当前接收所述组播组媒体流的网络边缘节点的标签标识;  a receiving unit, configured to receive a lookup message for finding a recently added copy point or a recently left copy point, where the lookup message includes a label identifier of an egress node that requests to join the multicast group and a network that currently receives the multicast group media stream a label identifier of the edge node, or a label identifier of the egress node requesting to leave the multicast group and a label identifier of the network edge node currently receiving the multicast group media stream;
判断单元, 用于根据接收到的查找报文, 利用单播路由, 计算当前组播标 签, 并判断自身是否为所述最近加入复制点或所述最近离开复制点;  a determining unit, configured to calculate, according to the received lookup message, a current multicast label by using a unicast route, and determine whether it is the recently added copy point or the recently left copy point;
发送单元, 用于在判断结果为是时, 将自身的节点标识发送至该组播组的 入口节点; 如果判断结果为否, 则根据单播路由将所述接收到的查找报文转发 给该组播组的媒体流转发链路上的下游节点。  a sending unit, configured to: when the judgment result is yes, send the node identifier of the node to the ingress node of the multicast group; if the judgment result is no, the received search packet is forwarded according to the unicast route to the The media stream of the multicast group forwards the downstream node on the link.
上述所述判断单元包括:  The determining unit described above includes:
组播标签交换单元, 用于将自身到所述出口节点的转发路径上出接口的标 签掩码, 和所述组播组的组播标签相与, 得到一个交换后的组播标签; 具体的, 根据查找报文中的当前接收所述组播组的媒体流的网络边缘节点的标签标识, 利用单播路由, 计算到所述出口节点的转发路径上出接口的标签掩码, 具体计 算方法参见方法实施例。  a multicast label switching unit, configured to: associate a label mask of an outgoing interface of the egress node with the multicast label of the multicast group, and obtain an exchanged multicast label; And calculating, according to the label identifier of the network edge node of the media stream that is currently receiving the multicast group in the packet, using the unicast route, calculating a label mask of the outbound interface on the forwarding path of the egress node, and the specific calculation method See method embodiment.
比较处理单元, 用于判断当前组播标签是否为零, 如果是则判定自身是最 近加入复制点; 或判断当前组播标签是否等于请求离开该组播组的出口节点的 标签标识 , 如果是则判定自身是所述最近离开复制点。  a comparison processing unit, configured to determine whether the current multicast label is zero, if yes, determine that it is the closest join point; or determine whether the current multicast label is equal to the label identifier of the egress node requesting to leave the multicast group, and if yes It is determined that it is the most recent departure point.
所述网络资源管理节点 RM包括:  The network resource management node RM includes:
接收单元, 用于接收入口节点发送的最近加入复制点到出口节点之间的带 宽资源预留请求, 或接收入口节点发送的最近离开复制点到出口节点之间的带 宽资源释放请求; 还用于接收所述出口节点对组播组的加入请求, 或者所述出 口节点对组播组的离开请求; 处理单元, 用于根据所述接收的带宽资源预留请求, 预留所述最近加入复 制点到所述出口节点之间的带宽资源; 或者根据所述接收的带宽资源释放请求, 释放所述最近离开复制点到所述出口节点之间的带宽资源。 a receiving unit, configured to receive a bandwidth resource reservation request sent by the ingress node to the egress node, or receive a bandwidth resource release request between the egress node and the egress node sent by the ingress node; Receiving, by the egress node, a request for joining a multicast group, or a request for leaving the multicast group by the egress node; a processing unit, configured to reserve, according to the received bandwidth resource reservation request, a bandwidth resource that is recently added to a replication point to the egress node; or release the latest according to the received bandwidth resource release request Leaving the copy point to the bandwidth resource between the egress nodes.
所述网络资源管理节点还包括:  The network resource management node further includes:
转发单元, 用于根据收到的加入请求或离开请求, 查找该组播组对应的入 口节点, 并将接收的所述出口节点对组播组的加入请求或所述出口节点对组播 组的离开请求, 转发至所述入口节点;  a forwarding unit, configured to search for an ingress node corresponding to the multicast group according to the received join request or leave request, and send the received join request of the egress node to the multicast group or the egress node to the multicast group Leaving the request, forwarding to the ingress node;
带宽分配确认单元, 用于在所述处理单元预留所述最近加入复制点到所述 出口节点之间的带宽资源后, 发送资源预留确认响应至所述入口节点; 或在所 述处理单元释放所述最近离开复制点到所述出口节点之间的带宽资源后, 发送 资源释放确认响应至所述入口节点。 在以上所述的一种组播控制的系统中, 网 络资源管理节点、 入口节点、 中间节点和出口节点之间, 都有链路连接或者通 过无线网络连接, 所述网络资源管理节点可以为网关, 或互联边界点网关, 或 服务器, 集中保存有对应于组播组 <g,S 々所有入口节点信息, 比如入口节点标 识。  a bandwidth allocation confirmation unit, configured to send a resource reservation confirmation response to the ingress node after the processing unit reserves the bandwidth resource between the most recently added replication point and the egress node; or in the processing unit After releasing the bandwidth resource that has recently left the replication point to the egress node, a resource release confirmation response is sent to the ingress node. In a multicast control system, the network resource management node, the ingress node, the intermediate node, and the egress node are all connected by a link or connected by a wireless network, and the network resource management node may be a gateway. , or the interconnected border point gateway, or server, centrally stores information about all ingress nodes corresponding to the multicast group <g, S, such as the ingress node identifier.
该系统结构简单, 各节点的组播转发过程不依赖组播路由协议, 即不需要 中间节点维护每个组的状态, 这样减少大量的管理报文交互传递, 避免了网络 拥塞。 该系统通过网络资源管理节点统一管理组播传送的过程, 尤其是在有出 口节点新加入或退出组播组时, 及时修改组播标签, 动态分配带宽资源。  The system has a simple structure. The multicast forwarding process of each node does not depend on the multicast routing protocol. That is, the intermediate node does not need to maintain the state of each group. This reduces the number of management packets and prevents network congestion. The system manages the multicast transmission process through the network resource management node. In particular, when an egress node joins or quits a multicast group, the multicast label is modified in time to dynamically allocate bandwidth resources.
为进一步说明实施例的一种组播控制系统, 如图 8 所示, 本发明实施例还 提供了一种入口节点 80 , 包括:  To further illustrate a multicast control system of an embodiment, as shown in FIG. 8, an embodiment of the present invention further provides an ingress node 80, including:
接收单元 801 , 用于接收出口节点对组播组的加入请求, 或出口节点对组播 组的离开请求; 当然还用于接收网络资源管理节点发送的资源预留确认响应或 资源释放确认响应;  The receiving unit 801 is configured to receive a request for joining the multicast group by the egress node, or a request for leaving the multicast group by the egress node, and is also used for receiving a resource reservation confirmation response or a resource release confirmation response sent by the network resource management node.
查找单元 802 , 用于根据接收到的出口节点对组播组的加入请求, 在对应组 播组媒体流的转发链路上, 查找与该出口节点之间转发路径最短的最近加入复 制点; 或者根据接收到的出口节点对组播组的离开请求, 在对应组播组媒体流 的转发链路上, 查找与该出口节点之间转发路径最短的最近离开复制点;  The searching unit 802 is configured to: according to the received join request of the egress group to the multicast group, search for a shortest nearest join copy point with the egress node on the forwarding link of the corresponding multicast group media stream; or And according to the received leaving request of the egress group to the multicast group, searching for the nearest leaving copy point with the shortest forwarding path between the egress node and the forwarding link of the corresponding multicast group media stream;
发送单元 803 ,用于发送查找到的最近加入复制点到所述出口节点之间的带 宽资源预留请求, 或发送查找到的最近离开复制点到出口节点之间的带宽资源 释放请求至所述网络资源管理节点; The sending unit 803 is configured to send the discovered bandwidth resource reservation request between the recently added replication point to the egress node, or send the discovered bandwidth resource between the recently left replication point and the egress node. Release a request to the network resource management node;
处理单元 804 , 用于根据接收的所述资源预留确认响应,修改该组播组的组 播标签, 以将所述出口节点加入所述组播组; 或根据接收的所述资源释放确认 响应, 修改该组播组的组播标签, 以将所述出口节点从所述组播组中删除。  The processing unit 804 is configured to modify the multicast label of the multicast group according to the received resource reservation confirmation response, to join the egress node to the multicast group, or release an acknowledgment response according to the received resource Modifying the multicast label of the multicast group to remove the egress node from the multicast group.
以上所述查找单元 802包括:  The search unit 802 described above includes:
报文生成单元 8021 , 用于生成查找报文, 所述查找报文包括所述出口节点 的标签标识、 当前接收所述组播组媒体流的网络边缘节点的标签标识和组播组 的组播标签; 还可以包括该入口节点 80 自身的标签标识。  The message generating unit 8021 is configured to generate a lookup message, where the lookup message includes a label identifier of the egress node, a label identifier of a network edge node that currently receives the multicast group media stream, and a multicast group multicast group. The tag; may also include the tag identifier of the ingress node 80 itself.
报文发送单元 8022, 用于发送所述查找报文至所述组播组的媒体流转发链 路上的下游节点上;  a message sending unit 8022, configured to send the lookup message to a downstream node of a media stream forwarding link of the multicast group;
响应接收单元 8023 , 用于接收所述最近加入复制点的节点标识, 或所述最 近离开复制点的节点标识。  The response receiving unit 8023 is configured to receive the node identifier of the most recently joined copy point, or the node identifier of the closest leaving the copy point.
如图 9所示, 本发明实施例进一步提供了一种中间节点 90, 包括: 接收单元 901 , 用于接收查找最近加入复制点或最近离开复制点的查找报 文, 所述查找报文包括请求加入 /离开组播组的出口节点的标签标识、 当前接收 所述组播组的媒体流的网络边缘节点的标签标识和组播组的组播标签; 查找报 文中还可以包括该组播组入口节点 80的标签标识。  As shown in FIG. 9, the embodiment of the present invention further provides an intermediate node 90, including: a receiving unit 901, configured to receive a lookup packet for finding a recently added copy point or a recently left copy point, where the lookup message includes a request. The label of the egress node that joins and leaves the multicast group, the label of the network edge node that receives the media stream of the multicast group, and the multicast label of the multicast group. The search packet may also include the multicast group. The tag identifier of the ingress node 80.
判断单元 902 , 用于根据所述接收到的查找报文, 利用单播路由, 计算当前 组播标签, 并判断自身是否为所述最近加入复制点或所述最近离开复制点; 发送单元 903 , 用于在判断结果为是时, 将自身的节点标识发送至该组播组 的入口节点; 在判断结果为否时, 根据单播路由向该组播组的媒体流转发链路 上的下游节点转发查找报文, 将该查找报文中的组播标签修改为当前组播标签。  The determining unit 902 is configured to calculate, according to the received search message, a current multicast label by using a unicast route, and determine whether it is the recently added copy point or the recently left copy point; the sending unit 903 When the judgment result is yes, the node identifier of the multicast group is sent to the ingress node of the multicast group; when the judgment result is no, the downstream node on the link is forwarded to the media stream of the multicast group according to the unicast route. Forward the search packet and change the multicast label in the lookup packet to the current multicast label.
其中所述判断单元 902包括:  The determining unit 902 includes:
组播标签交换单元 9021 , 用于根据查找报文中的当前接收所述组播组的媒 体流的网络边缘节点的标签标识, 利用单播路由, 计算自身到所述出口节点的 转发路径上出接口的标签掩码, 并将该标签掩码和所述组播组的组播标签相与, 得到一个交换后的组播标签, 即当前组播标签; 具体计算过程可以参考方法实 施例;  The multicast label switching unit 9021 is configured to calculate, according to the label identifier of the network edge node of the media stream that currently receives the multicast group in the packet, use the unicast route to calculate the forwarding path to the egress node. The label mask of the interface is matched with the multicast label of the multicast group to obtain an exchanged multicast label, that is, the current multicast label. For the specific calculation process, refer to the method embodiment.
比较处理单元 9022, 用于判断所述交换后的组播标签值(当前组播标签) 是否为零, 如果为零, 则判定自身是所述最近加入复制点; 或判断所述交换后 的组播标签值是否等于所述出口节点的标签标识值, 如果等于, 则判定自身是 所述最近离开复制点。 The comparison processing unit 9022 is configured to determine whether the exchanged multicast label value (current multicast label) is zero, and if it is zero, determine that it is the latest join copy point; or determine the exchange Whether the multicast tag value is equal to the tag identifier value of the egress node, and if equal, it is determined that it is the most recently left copy point.
图 8和 9所示的网络节点可以是同一种网络设备, 如路由器, 因此该网络 设备可以同时具备入口节点、 出口节点和中间节点的功能, 并可设置为功能可 选, 这样运营商可以根据自身需求, 在网络中不设同一种设备, 如路由器, 根 据该设备的位置将其设置为边缘节点或中间节点。  The network nodes shown in Figures 8 and 9 can be the same type of network device, such as a router, so the network device can have the functions of an ingress node, an egress node, and an intermediate node at the same time, and can be set to be functionally selectable, so that the operator can For its own needs, there is no same device in the network, such as a router, which is set as an edge node or an intermediate node according to the location of the device.
如图 10所示, 本发明实施例更进一步提供了一种网络资源管理节点 100, 包括:  As shown in FIG. 10, the embodiment of the present invention further provides a network resource management node 100, including:
接收单元 101 , 用于接收所述出口节点对组播组的加入请求, 或者所述出口 节点对组播组的离开请求; 当然所述接收单元还用于接收入口节点发送的最近 加入复制点到出口节点之间的带宽资源预留请求, 或最近离开复制点到出口节 点之间的带宽资源释放请求;  The receiving unit 101 is configured to receive the join request of the egress node to the multicast group, or the egress request of the egress node to the multicast group; of course, the receiving unit is further configured to receive the latest join copy point sent by the ingress node to a bandwidth resource reservation request between the egress nodes, or a bandwidth resource release request between the egress node and the egress node;
转发单元 102,用于将接收的所述出口节点对组播组的加入请求或所述出口 节点对组播组的离开请求, 转发至所述入口节点;  The forwarding unit 102 is configured to forward the received join request of the egress node to the multicast group or the exit request of the egress node to the multicast group to the ingress node;
处理单元 103 , 用于根据所述接收的带宽资源预留请求,预留所述最近加入 复制点到所述出口节点之间的带宽资源; 或者根据所述接收的带宽资源释放请 求, 释放所述最近离开复制点到所述出口节点之间的带宽资源;  The processing unit 103 is configured to reserve, according to the received bandwidth resource reservation request, a bandwidth resource that is recently added to the replication point to the egress node, or release the content according to the received bandwidth resource release request. Recently leaving the replication point to the bandwidth resource between the egress nodes;
带宽分配确认单元 104,用于根据所述处理单元预留所述最近加入复制点到 所述出口节点之间的带宽资源, 发送资源预留确认响应至所述入口节点; 或根 据所述处理单元释放所述最近离开复制点到所述出口节点之间的带宽资源, 发 送资源释放确认响应至所述入口节点。  The bandwidth allocation confirmation unit 104 is configured to reserve, according to the processing unit, the bandwidth resource between the recently added replication point and the egress node, and send a resource reservation confirmation response to the ingress node; or according to the processing unit Release the bandwidth resource between the nearest leaving replication point to the egress node, and send a resource release confirmation response to the ingress node.
釆用本发明实施例提供的一种实现组播资源接纳控制或者释放控制的系 统、 网络节点设备, 能利用网络资源管理节点统一管理组播传送的过程, 尤其 是在有出口节点新加入或退出组播组时, 通过查找所述出口节点对应的最近加 入复制点或者最近离开复制点, 及时预留最近加入复制点到所述出口节点之间 的带宽, 或者及时释放最近离开复制点到所述出口节点之间的带宽, 实现动态 合理地分配带宽资源。 提高了网络带宽利用率。  A system and a network node device for implementing multicast resource admission control or release control provided by the embodiments of the present invention can use the network resource management node to uniformly manage the multicast transmission process, especially when an egress node newly joins or exits. In the case of a multicast group, the bandwidth of the most recently added copy point to the egress node is reserved in time by searching for the nearest join copy point corresponding to the egress node or the nearest leave copy point, or releasing the recently left copy point to the The bandwidth between the egress nodes enables dynamic and reasonable allocation of bandwidth resources. Increased network bandwidth utilization.
以上所述是本发明的优选实施方式而已, 当然不能以此来限定本发明之权 利范围, 应当指出, 对于本技术领域的普通技术人员来说, 在不脱离本发明原 理的前提下, 还可以做出若干改进, 这些改进和变动也视为本发明的保护范围。  The above is a preferred embodiment of the present invention, and the scope of the present invention is not limited thereto. It should be noted that those skilled in the art can also omit the principles of the present invention without departing from the scope of the present invention. A number of improvements are made which are also considered to be within the scope of the invention.

Claims

权 利 要 求  Rights request
1、 一种组播控制的方法, 其特征在于, 包括下列步骤:  A method for multicast control, comprising the steps of:
接收第一边缘节点对组播组的加入请求, 或所述第一边缘节点对所述组播 组的离开请求;  Receiving a request for joining the multicast group by the first edge node, or a request for leaving the multicast group by the first edge node;
当接收到所述加入请求时, 利用单播路由向下游节点发送查找报文, 以查 找与所述第一边缘节点对应的最近加入复制点, 或当接收到所述离开请求时, 利用单播路由向下游节点发送查找报文, 以查找与所述第一边缘节点对应的最 近离开复制点, 所述的查找报文中携带有所述第一边缘节点的标签标识, 所述 组播组的组播标签和所述组播组的出口节点的标签标识;  When receiving the join request, sending a lookup message to the downstream node by using a unicast route to find a nearest join copy point corresponding to the first edge node, or using the unicast when receiving the leave request The route sends a lookup packet to the downstream node to find the nearest leave replication point corresponding to the first edge node, where the search packet carries the label identifier of the first edge node, where the multicast group a multicast label and a label identifier of an egress node of the multicast group;
当找到所述最近加入复制点后, 请求为所述最近加入复制点到所述第一边 缘节点之间预留带宽资源, 并在接收到资源确认响应信号后, 修改所述组播组 的组播标签, 以将所述第一边缘节点加入该组播组; 或者当找到所述最近离开 复制点后 , 请求为所述最近离开复制点到所述第一边缘节点之间释放带宽资源; 并在接收到资源确认响应信号后, 修改所述组播组的组播标签, 以将所述第一 边缘节点从该组播组中删除。  After finding the recently joined replication point, requesting to reserve bandwidth resources between the recently added replication point and the first edge node, and modifying the group of the multicast group after receiving the resource confirmation response signal Broadcasting a label to join the first edge node to the multicast group; or, after finding the most recently left copy point, requesting to release bandwidth resources between the recently left copy point and the first edge node; After receiving the resource confirmation response signal, modifying the multicast label of the multicast group to delete the first edge node from the multicast group.
2、 根据权利要求 1所述的一种组播控制的方法, 其特征在于, 2. A method of multicast control according to claim 1, wherein:
所述当接收到所述加入请求时, 利用单播路由向下游节点发送查找报文, 查找与所述第一边缘节点对应的最近加入复制点 , 具体包括:  When the request for the join is received, the unicast route is used to send a lookup packet to the downstream node, and the latest join copy point corresponding to the first edge node is searched, which specifically includes:
当接收到所述加入请求时, 利用单播路由发送查找报文至所述组播组的媒 体流转发链路上的节点, 判断所述组播组媒体流转发链路上的节点是否为所述 最近加入复制点;  When receiving the join request, the unicast route is used to send a lookup packet to the node on the media stream forwarding link of the multicast group, and determine whether the node on the multicast group media stream forwarding link is Adding a copy point recently;
所述当接收到所述离开请求时, 利用单播路由向下游节点发送查找报文, 查找与所述第一边缘节点对应的最近离开复制点, 具体包括:  When the detachment request is received, the unicast route is used to send a lookup packet to the downstream node to find the nearest leaving replication point corresponding to the first edge node, which specifically includes:
当接收到所述离开请求时, 利用单播路由发送查找报文至所述组播组媒体 流转发链路上的节点, 判断所述组播组媒体流转发链路上的节点是否为所述最 近离开复制点。  When the leaving request is received, the unicast route is used to send a lookup packet to the node on the multicast group media stream forwarding link, and it is determined whether the node on the multicast group media stream forwarding link is the Recently left the copy point.
3、 根据权利要求 2所述的一种组播控制的方法, 其特征在于, 所述判断所述组播组媒体流转发链路上的节点是否为所述最近加入复制 点, 具体包括: 3. A method of multicast control according to claim 2, wherein Determining whether the node on the forwarding group of the multicast group media stream is the closest join copy point, specifically:
判断所述组播组媒体流转发链路上的节点到所述第一边缘节点的最短转发 路径上是否已有所述组播组媒体流在转发, 如果没有, 则确定所述组播组媒体 流转发链路上的节点为所述最近加入复制点;  Determining whether the multicast group media stream is forwarded on the shortest forwarding path of the node on the multicast group media stream forwarding link to the first edge node, and if not, determining the multicast group media. The node on the flow forwarding link is the recently added copy point;
所述判断所述组播组媒体流转发链路上的节点是否为所述最近离开复制 点, 具体包括:  Determining whether the node on the forwarding group of the multicast group media stream is the closest leaving replication point, specifically:
判断所述组播组媒体流转发链路上的节点到所述第一边缘节点的转发路径 上是否有其它组播转发分支节点, 如果没有, 则确定所述组播组媒体流转发链 路上的节点为所述最近离开复制点。  Determining whether there is another multicast forwarding branch node on the forwarding path of the node on the multicast group media stream forwarding link to the first edge node, and if not, determining that the multicast group media stream is forwarded on the link The node is the most recently left copy point.
4、 根据权利要求 2或 3所述的一种组播控制的方法, 其特征在于, 所述判断所述组播组媒体流转发链路上的节点是否为所述最近加入复制 点, 具体包括: The method of multicast control according to claim 2 or 3, wherein the determining whether the node on the forwarding link of the multicast group media stream is the most recently added copy point includes :
所述组播组媒体流转发链路上的节点根据单播路由和接收到的查找报文中 的组播标签, 计算到所述第一边缘节点的转发路径上出接口的标签掩码, 将该 标签掩码和所述查找报文中携带的组播组的组播标签相与, 得到交换后的组播 标签, 如果所述交换后的组播标签为零, 则确定该组播组媒体流转发链路上的 节点是所述最近加入复制点;  The node on the multicast group media stream forwarding link calculates the label mask of the outbound interface on the forwarding path of the first edge node according to the unicast route and the multicast label in the received lookup packet, and The label mask is matched with the multicast label of the multicast group carried in the search packet, and the exchanged multicast label is obtained. If the exchanged multicast label is zero, the multicast group medium is determined. The node on the stream forwarding link is the most recently added copy point;
所述判断所述组播组媒体流转发链路上的节点是否为所述最近离开复制 点, 具体包括:  Determining whether the node on the forwarding group of the multicast group media stream is the closest leaving replication point, specifically:
所述组播组媒体流转发链路上的节点根据单播路由和接收到的查找报文中 的组播标签, 计算到所述第一边缘节点的转发路径上出接口的标签掩码, 将该 标签掩码和所述查找报文中携带的组播组的组播标签相与, 得到交换后的组播 标签, 如果所述交换后的组播标签值与所述第一边缘节点的标签标识值相等, 则确定所述组播组媒体流转发链路上的节点是所述最近离开复制点。  The node on the multicast group media stream forwarding link calculates the label mask of the outbound interface on the forwarding path of the first edge node according to the unicast route and the multicast label in the received lookup packet, and The label mask is matched with the multicast label of the multicast group carried in the search packet, and the exchanged multicast label is obtained, if the exchanged multicast label value and the label of the first edge node are If the identifier values are equal, it is determined that the node on the multicast group media stream forwarding link is the most recently left replication point.
5、 根据权利要求 1或 2所述的一种组播控制的方法, 其特征在于, 在所述请求为所述最近加入复制点到所述第一边缘节点之间预留带宽资源 之前, 还包括, 接收所述最近加入复制点发送的报告自己是本次查找的最近加 入复制点的信息; The method for multicast control according to claim 1 or 2, wherein before the request reserves bandwidth resources between the recently added copy point and the first edge node, Including, receiving the report sent by the recently added copy point is itself the most recent addition of the current search Information into the copy point;
或在所述请求为所述最近离开复制点到所述第一边缘节点之间释放带宽资 源之前, 还包括, 接收所述最近离开复制点发送的报告自己是本次查找的最近 离开复制点的信息。  Or before the request releases the bandwidth resource between the recently left replication point and the first edge node, the method further includes: receiving the report sent by the nearest leaving replication point, which is the nearest leaving replication point of the current search. information.
6、 根据权利要求 5所述的一种组播控制的方法, 其特征在于, 所述最近加 入复制点满足如下条件: 在所述第一边缘节点请求加入所述组播组时, 所述最 近加入复制点已收到所述组播组媒体流; 所述最近加入复制点位于从所述组播 组入口节点到所述第一边缘节点的当前单播转发路径上; 所述最近加入复制点 到所述第一边缘节点的单播转发路径是最短的; The multicast control method according to claim 5, wherein the recent join replication point satisfies the following condition: when the first edge node requests to join the multicast group, the latest Receiving the replication point has received the multicast group media stream; the recent join replication point is located on a current unicast forwarding path from the multicast group ingress node to the first edge node; The unicast forwarding path to the first edge node is the shortest;
所述最近离开复制点满足如下条件: 在所述第一边缘节点请求离开所述组 播组时, 所述最近离开复制点已收到所述组播组媒体流; 所述最近离开复制点 位于从所述组播组入口节点到所述第一边缘节点的当前单播转发路径上; 所述 最近离开复制点到所述第一边缘节点的组播转发路径上没有其它组播转发分支 节点。  The recent leaving replication point satisfies the following condition: when the first edge node requests to leave the multicast group, the most recent leaving replication point has received the multicast group media stream; From the multicast group ingress node to the current unicast forwarding path of the first edge node; there is no other multicast forwarding branch node on the multicast forwarding path of the nearest leaving replication point to the first edge node.
7、 一种网络节点, 其特征在于, 包括: 7. A network node, comprising:
接收单元, 用于接收第一边缘节点对组播组的加入请求, 或第一边缘节点 对组播组的离开请求; 以及接收资源预留确认响应或资源释放确认响应;  a receiving unit, configured to receive a join request of the first edge node to the multicast group, or a leave request of the first edge node to the multicast group; and receive a resource reservation acknowledgement response or a resource release acknowledgement response;
查找单元, 用于根据接收的所述对组播组的加入请求, 在所述组播组的媒 体流的转发链路上, 查找到所述第一边缘节点之间转发路径最短的最近加入复 制点; 或者根据接收的所述对组播组的离开请求, 在所述组播组的媒体流的转 发链路上, 查找到所述第一边缘节点之间转发路径最短的最近离开复制点; 发送单元, 用于发送查找到的最近加入复制点到所述第一边缘节点之间的 带宽资源预留请求, 或发送查找到的最近离开复制点到所述第一边缘节点之间 的带宽资源释放请求;  a locating unit, configured to find, according to the received join request for the multicast group, the latest join replication with the shortest forwarding path between the first edge nodes on the forwarding link of the media stream of the multicast group Pointing; or, according to the received leaving request for the multicast group, finding the closest leaving copy point with the shortest forwarding path between the first edge nodes on the forwarding link of the media stream of the multicast group; a sending unit, configured to send a found bandwidth resource reservation request between the recently added replication point to the first edge node, or send the discovered bandwidth resource between the recently left replication point and the first edge node Release request
处理单元, 用于在接收到资源预留确认响应后, 修改该组播组的组播标签, 以将所述第一边缘节点加入该组播组; 或者在接收到资源释放确认响应后, 修 改该组播组的组播标签, 以将所述第一边缘节点从该组播组中删除。 8、 根据权利要求 7所述的一种网络节点, 其特征在于, a processing unit, configured to: after receiving the resource reservation confirmation response, modify the multicast label of the multicast group to join the first edge node to the multicast group; or after receiving the resource release confirmation response, modify A multicast label of the multicast group to remove the first edge node from the multicast group. 8. A network node according to claim 7, wherein:
所述查找单元包括:  The searching unit includes:
报文生成单元, 用于生成查找报文, 所述查找报文包括所述第一边缘节点 的标签标识和当前接收所述组播组媒体流的网络边缘节点的标签标识, 以及所 述组播组的组播标签;  a message generating unit, configured to generate a lookup message, where the lookup message includes a label identifier of the first edge node, and a label identifier of a network edge node that currently receives the multicast group media stream, and the multicast Group multicast label;
报文发送单元, 用于发送所述查找报文至所述组播组的媒体流转发链路上 的下游节点上;  a message sending unit, configured to send the lookup message to a downstream node on a media stream forwarding link of the multicast group;
响应接收单元, 用于接收所述最近加入复制点的节点标识, 或所述最近离 开复制点的节点标识。  The response receiving unit is configured to receive the node identifier of the most recently joined replication point, or the node identifier of the most recently separated replication point.
9、 根据权利要求 7或 8所述的网络节点, 其特征在于, 9. A network node according to claim 7 or 8, characterized in that
所述最近加入复制点满足如下条件: 在所述第一边缘节点请求加入所述组 播组时, 所述最近加入复制点已收到所述组播组媒体流; 所述最近加入复制点 位于从所述组播组入口节点到所述第一边缘节点的当前单播转发路径上; 所述 最近加入复制点到所述第一边缘节点的单播转发路径是最短的;  The recent join replication point satisfies the following condition: when the first edge node requests to join the multicast group, the latest join replication point has received the multicast group media stream; From the multicast group ingress node to the current unicast forwarding path of the first edge node; the unicast forwarding path of the most recently added replication point to the first edge node is the shortest;
所述最近离开复制点满足如下条件: 在所述第一边缘节点请求离开所述组 播组时, 所述最近离开复制点已收到所述组播组媒体流; 所述最近离开复制点 位于从所述组播组入口节点到所述第一边缘节点的当前单播转发路径上; 所述 最近离开复制点到所述第一边缘节点的组播转发路径上没有其它组播转发分支 节点。  The recent leaving replication point satisfies the following condition: when the first edge node requests to leave the multicast group, the most recent leaving replication point has received the multicast group media stream; From the multicast group ingress node to the current unicast forwarding path of the first edge node; there is no other multicast forwarding branch node on the multicast forwarding path of the nearest leaving replication point to the first edge node.
10、 一种网络节点, 其特征在于, 包括: 10. A network node, comprising:
接收单元, 用于接收查找到请求加入组播组的第一边缘节点的最近加入复 制点的查找报文, 或接收查找到请求离开组播组的第一边缘节点的最近离开复 制点的查找报文, 所述查找报文包括所述第一边缘节点的标签标识和当前接收 所述组播组媒体流的网络边缘节点的标签标识, 以及所述组播组的组播标签; 判断单元, 用于根据所述接收到的查找报文, 利用单播路由, 计算当前组 播标签, 并判断自身是否为所述最近加入复制点或所述最近离开复制点;  a receiving unit, configured to receive a lookup message of a recently added copy point of the first edge node that requests to join the multicast group, or receive a lookup report of the nearest leaving copy point of the first edge node that requests to leave the multicast group The search packet includes a label identifier of the first edge node and a label identifier of a network edge node that currently receives the multicast group media stream, and a multicast label of the multicast group; Calculating the current multicast label by using the unicast route according to the received search packet, and determining whether it is the recently added copy point or the recently left copy point;
发送单元, 用于在判断结果为是时, 将自身的节点标识发送给所述组播组 的入口节点; 在判断结果为否时, 根据单播路由向该组播组的媒体流转发链路 上的下游节点转发查找报文, 将该查找报文中的组播标签修改为当前组播标签。 a sending unit, configured to send a node identifier of the node to the ingress node of the multicast group when the judgment result is yes; and when the judgment result is no, forward the link to the media stream of the multicast group according to the unicast route The upstream node forwards the lookup packet and modifies the multicast label in the lookup packet to the current multicast label.
11、 根据权利要求 10所述的一种网络节点, 其特征在于, 11. A network node according to claim 10, characterized in that
所述判断单元包括:  The determining unit includes:
组播标签交换单元, 用于根据所述查找报文中的当前接收所述组播组的媒 体流的网络边缘节点的标签标识, 利用单播路由, 计算到所述第一边缘节点的 转发路径上出接口的标签掩码, 并将该标签掩码和所述组播组的组播标签相与, 得到当前组播标签;  a multicast label switching unit, configured to calculate a forwarding path of the first edge node by using a unicast route according to a label identifier of a network edge node that currently receives the media stream of the multicast group in the search packet Label the label of the outgoing interface, and the label mask is matched with the multicast label of the multicast group to obtain the current multicast label.
比较处理单元, 用于判断所述当前组播标签是否为零, 如果是则判定自身 是所述最近加入复制点; 或判断所述当前组播标签是否等于所述第一边缘节点 的标签标识, 如果是则判定自身是所述最近离开复制点。  a comparison processing unit, configured to determine whether the current multicast label is zero, and if yes, determine that the current multicast label is the closest join copy point; or determine whether the current multicast label is equal to the label identifier of the first edge node, If so, it is determined that it is the most recent leaving copy point.
12根据权利要求 10或 11所述的网络节点, 其特征在于, A network node according to claim 10 or 11, characterized in that
所述最近加入复制点满足如下条件: 在所述第一边缘节点请求加入所述组 播组时, 所述最近加入复制点已收到所述组播组的媒体流; 所述最近加入复制 点位于从所述组播组入口节点到所述第一边缘节点的当前单播转发路径上; 所 述最近加入复制点到所述第一边缘节点的单播转发路径是最短的;  The recent joining of the replication point satisfies the following condition: when the first edge node requests to join the multicast group, the recently joined replication point has received the media stream of the multicast group; Located on the current unicast forwarding path from the multicast group ingress node to the first edge node; the unicast forwarding path of the most recently added replication point to the first edge node is the shortest;
所述最近离开复制点满足如下条件: 在所述第一边缘节点请求离开所述组 播组时, 所述最近离开复制点已收到所述组播组的媒体流; 所述最近离开复制 点位于从所述组播组入口节点到所述第一边缘节点的当前单播转发路径上; 所 述最近离开复制点到所述第一边缘节点的组播转发路径上没有其它组播转发分 支节点。  The recent leaving replication point satisfies the following condition: when the first edge node requests to leave the multicast group, the most recent leaving replication point has received the media stream of the multicast group; Located on the current unicast forwarding path from the multicast group ingress node to the first edge node; there is no other multicast forwarding branch node on the multicast forwarding path of the nearest leaving replication point to the first edge node .
13、 一种组播控制的系统, 其特征在于, 包括: 第一边缘节点、 第二边缘 节点、 中间节点和网络资源管理节点; A multicast control system, comprising: a first edge node, a second edge node, an intermediate node, and a network resource management node;
所述第一边缘节点, 用于发送对组播组的加入请求, 或发送对组播组的离 开请求;  The first edge node is configured to send a join request to the multicast group, or send a leave request to the multicast group;
所述第二边缘节点, 用于接收所述第一边缘节点对组播组的加入请求, 或 接收所述第一边缘节点对组播组的离开请求, 根据接收的所述第一边缘节点对 组播组的加入请求, 在所述组播组的媒体流的转发链路上, 发送查找最近加入 复制点的查找报文; 或者根据接收的所述第一边缘节点对组播组的离开请求, 在所述组播组的媒体流的转发链路上, 发送查找最近离开复制点的查找报文; 所述的查找报文中携带有所述第一边缘节点的标签标识, 所述组播组的组播标 签和所述组播组的所有出口节点的标签标识; 当找到所述最近加入复制点后, 向所述网络资源管理节点请求为该最近加入复制点到第一边缘节点之间预留带 宽资源, 并在接收到所述网络资源管理节点发送的资源确认响应信号后, 修改 所述组播组的组播标签, 以将所述第一边缘节点加入该组播组; 或者当找到所 述最近离开复制点后, 向所述网络资源管理节点请求为该最近离开复制点到第 一边缘节点之间释放带宽资源; 并在接收到所述网络资源管理节点资源发送确 认响应信号后, 修改所述组播组的组播标签, 以将所述第一边缘节点从该组播 组中删除; The second edge node is configured to receive a request for joining the multicast group by the first edge node, or receive a request for leaving the multicast group by the first edge node, according to the received first edge node pair A multicast group join request, on the forwarding link of the media stream of the multicast group, sending a search to join recently And a lookup message of the copy point; or, according to the received leaving request of the first edge node to the multicast group, sending a lookup packet for finding the nearest leaving copy point on the forwarding link of the media stream of the multicast group The search packet carries the label identifier of the first edge node, the multicast label of the multicast group, and the label identifier of all the egress nodes of the multicast group; when the recently added copy is found After the point, requesting, by the network resource management node, to reserve bandwidth resources between the recently added replication point and the first edge node, and after receiving the resource confirmation response signal sent by the network resource management node, modify the a multicast label of the multicast group to join the first edge node to the multicast group; or, after finding the most recently left copy point, requesting the network resource management node to be the closest leaving copy point to the first And releasing the bandwidth resource between the edge nodes; and after receiving the acknowledgement response signal sent by the network resource management node, modifying the multicast label of the multicast group, so as to The first edge node is deleted from the multicast group;
所述中间节点, 用于接收所述第二边缘节点发送的查找报文, 根据该查找 报文, 利用单播路由, 计算当前组播标签; 并判断该中间节点自身是否为最近 加入复制点或最近离开复制点; 如果是则向所述第二边缘节点报告, 否则向在 所述组播组的媒体流的转发链路上的下游节点转发查找报文, 将查找报文中的 组播标签修改为所述当前组播标签;  The intermediate node is configured to receive a lookup packet sent by the second edge node, calculate a current multicast label by using a unicast route according to the lookup packet, and determine whether the intermediate node itself is newly added to the replication point or Leaving the replication point recently; if yes, reporting to the second edge node, otherwise forwarding the lookup packet to the downstream node on the forwarding link of the media stream of the multicast group, and searching for the multicast label in the packet Modified to the current multicast label;
所述网络资源管理节点, 用于在接收到第二边缘节点发送的为所述最近加 入复制点到第一边缘节点之间预留带宽资源的请求后, 进行带宽资源预留, 并 在预留成功时向所述第二边缘节点发送资源确认响应; 或者在接收第二边缘节 点发送的为所述最近离开复制点到第一边缘节点之间释放带宽资源的请求, 进 行带宽资源释放, 并在资源释放后向所述第二边缘节点发送资源确认响应。  And the network resource management node is configured to perform bandwidth resource reservation after receiving the request for the bandwidth resource reserved by the second edge node to send the copy point to the first edge node, and reserve the bandwidth resource Sending a resource acknowledgment response to the second edge node when successful; or receiving a request for releasing the bandwidth resource between the recently left copy point and the first edge node sent by the second edge node, performing bandwidth resource release, and After the resource is released, a resource confirmation response is sent to the second edge node.
14、 根据权利要求 13所述的系统, 其特征在于: 14. The system of claim 13 wherein:
所述中间节点具体包括:  The intermediate node specifically includes:
接收单元, 用于接收所述第二边缘节点发送的查找报文, 所述查找报文包 括所述第一边缘节点的标签标识和当前接收所述组播组媒体流的网络边缘节点 的标签标识, 以及所述组播组的组播标签; 组播标签交换单元, 用于根据所述 查找报文中的当前接收所述组播组媒体流的网络边缘节点的标签标识, 利用单 播路由, 计算到所述第一边缘节点的转发路径上出接口的标签掩码, 并将该标 签掩码和所述组播组的组播标签相与, 得到当前组播标签; 比较处理单元, 用于判断所述当前组播标签是否为零, 如果是则判定自身 是所述最近加入复制点; 或判断所述当前的组播标签值是否等于所述第一边缘 节点的标签标识 , 如果是则判定自身是所述最近离开复制点。 a receiving unit, configured to receive a lookup packet sent by the second edge node, where the lookup packet includes a label identifier of the first edge node and a label identifier of a network edge node that currently receives the multicast group media stream And the multicast label of the multicast group, where the multicast label switching unit is configured to use the unicast route according to the label identifier of the network edge node that currently receives the multicast group media stream in the lookup packet, Calculating a label mask of the outbound interface on the forwarding path of the first edge node, and matching the label mask with the multicast label of the multicast group to obtain a current multicast label; a comparison processing unit, configured to determine whether the current multicast label is zero, if yes, determine that it is the most recently added replication point; or determine whether the current multicast label value is equal to the label of the first edge node The flag, if yes, determines that it is the most recently left copy point.
15、 根据权利要求 13或 14所述的系统, 其特征在于: 15. A system according to claim 13 or claim 14 wherein:
所述最近加入复制点满足如下条件: 在所述第一边缘节点请求加入所述组 播组时, 所述最近加入复制点已收到所述组播组媒体流; 所述最近加入复制点 位于从所述组播组的第二边缘节点到所述第一边缘节点的当前单播转发路径 上; 所述最近加入复制点到所述第一边缘节点的单播转发路径是最短的;  The recent join replication point satisfies the following condition: when the first edge node requests to join the multicast group, the latest join replication point has received the multicast group media stream; From the second edge node of the multicast group to the current unicast forwarding path of the first edge node; the unicast forwarding path of the most recently added replication point to the first edge node is the shortest;
所述最近离开复制点满足如下条件: 在所述第一边缘节点请求离开所述组 播组时, 所述最近离开复制点已收到所述组播组媒体流; 所述最近离开复制点 位于从所述组播组的第二边缘节点到所述第一边缘节点的当前单播转发路径 上; 所述最近离开复制点到所述第一边缘节点的组播转发路径上没有其它组播 转发分支节点。  The recent leaving replication point satisfies the following condition: when the first edge node requests to leave the multicast group, the most recent leaving replication point has received the multicast group media stream; From the second edge node of the multicast group to the current unicast forwarding path of the first edge node; there is no other multicast forwarding on the multicast forwarding path of the nearest leaving copy point to the first edge node Branch node.
PCT/CN2009/070776 2008-03-13 2009-03-13 Multicast controlling method, device and system WO2009111990A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200810026778.4 2008-03-13
CN2008100267784A CN101534203B (en) 2008-03-13 2008-03-13 Method, equipment and system for multicast control

Publications (1)

Publication Number Publication Date
WO2009111990A1 true WO2009111990A1 (en) 2009-09-17

Family

ID=41064769

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/070776 WO2009111990A1 (en) 2008-03-13 2009-03-13 Multicast controlling method, device and system

Country Status (2)

Country Link
CN (1) CN101534203B (en)
WO (1) WO2009111990A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114338569A (en) * 2021-12-31 2022-04-12 锐捷网络股份有限公司 Message copying and forwarding method and device
CN115134287A (en) * 2022-06-30 2022-09-30 中国电信股份有限公司 Multicast state and data management method and system, controller and storage medium

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109743250B (en) * 2018-12-07 2020-09-04 华为技术有限公司 Transmission method of multicast message, first network equipment and second network equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1747446A (en) * 2005-10-21 2006-03-15 清华大学 Application layer group broadcasting method with integrated type and distributed type combination
CN101047444A (en) * 2006-04-28 2007-10-03 华为技术有限公司 Method for optical network unit depart from multicast set in passive optical network system
CN101083546A (en) * 2007-07-18 2007-12-05 中兴通讯股份有限公司 Multicasting management approach of switch equipment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1747446A (en) * 2005-10-21 2006-03-15 清华大学 Application layer group broadcasting method with integrated type and distributed type combination
CN101047444A (en) * 2006-04-28 2007-10-03 华为技术有限公司 Method for optical network unit depart from multicast set in passive optical network system
CN101083546A (en) * 2007-07-18 2007-12-05 中兴通讯股份有限公司 Multicasting management approach of switch equipment

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114338569A (en) * 2021-12-31 2022-04-12 锐捷网络股份有限公司 Message copying and forwarding method and device
CN115134287A (en) * 2022-06-30 2022-09-30 中国电信股份有限公司 Multicast state and data management method and system, controller and storage medium

Also Published As

Publication number Publication date
CN101534203B (en) 2012-11-07
CN101534203A (en) 2009-09-16

Similar Documents

Publication Publication Date Title
US9860163B2 (en) MPLS traffic engineering for point-to-multipoint label switched paths
JP4612987B2 (en) Multiprotocol label switching based point-to-multipoint traffic routing method
WO2007054032A1 (en) Communication network system and leaf-node network element of the multicasting tree signal transmission method and node network element thereof
CA2289070A1 (en) Multicast switching
WO2009021464A1 (en) Method, device and system for realizing multicast service
WO2010127570A1 (en) Method, device and system for establishing bidirectional point-to-multipoint label switched path
WO2008131651A1 (en) Method, system and device of multicast resource control
WO2009082905A1 (en) Method, system and switch device for dynamically establishing multicast virtual local area network
WO2007059679A1 (en) A method for processing the abnormal multicast service and a network equipment thereof
JP5548696B2 (en) Multicast quality of service module and method
JP4654278B2 (en) Multicast tree assignment method and apparatus
JP4787376B2 (en) Method, apparatus and system for controlling multicast bearer resources
WO2009024054A1 (en) A method, device and system for realizing the management protocol agent for members in a multicast group
WO2009111990A1 (en) Multicast controlling method, device and system
WO2008046336A1 (en) A system and method for realizing distributed access control in multicast service
WO2011144100A2 (en) Service scheduling method and apparatus under multiple broadband network gateways
WO2015014197A1 (en) Method for selecting route in scenario of multicast load, and router
Egilmez et al. Openqos: Openflow controller design and test network for multimedia delivery with quality of service
WO2012103748A1 (en) Multicast control method, routing device and multicast system
WO2013037319A1 (en) Bandwidth resource management method, device and system for label switched path
CA2376690A1 (en) System and method for loop avoidance in multi-protocol label switching
WO2008138236A1 (en) Multicast method and device
US8149852B2 (en) Transmission method, system and router based on a border gateway protocol
US11018886B1 (en) Methods and apparatus for selectively filtering an IP multicast data stream for selected group members of a multicast group
WO2012136062A1 (en) Multicast path switching method and router

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09718878

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09718878

Country of ref document: EP

Kind code of ref document: A1