CN101102249B - A method for quickly switching Ethernet loop network multicast link - Google Patents
A method for quickly switching Ethernet loop network multicast link Download PDFInfo
- Publication number
- CN101102249B CN101102249B CN2007101430365A CN200710143036A CN101102249B CN 101102249 B CN101102249 B CN 101102249B CN 2007101430365 A CN2007101430365 A CN 2007101430365A CN 200710143036 A CN200710143036 A CN 200710143036A CN 101102249 B CN101102249 B CN 101102249B
- Authority
- CN
- China
- Prior art keywords
- multicast
- node
- port
- user
- link
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Small-Scale Networks (AREA)
Abstract
The method comprises: setting a group-discarded mode for each multicast group in the Ethernet; under said group-discarded mode, creating a multicast forwarding table for each multicast node; adding the ring-network port of each node into the multicast forwarding table of the node; the ring-network ports of all nodes forms a multicast path of Ethernet; while the users or links in the Ethernet vary, the multicast link is switched at the node relating to the variation. The invention introduces a group-discarded mode under which the multicast table of each node in Ethernet always exists such that when needing to make link switch, the multicast forwarding table doesn't need to be made any more and the notice to main node is also not to be required so as to reduce the delay of switching the multicast links.
Description
Technical field
The present invention relates to the automatic resist technology of ethernet ring network, particularly a kind of ethernet ring network multicast link fast switching method.
Background technology
Ethernet ring network is obtaining application more and more widely as a kind of Metro Ethernet technology.Ethernet ring network has solved problems such as traditional data net protective capability is weak, delay is big, failure recovery time is long; the quick protective feature of 50ms can be provided; improve service quality and the reliability of network; compatible traditional simultaneously Ethernet protocol is that a kind of important techniques that the optimization of metropolitan area broadband access aggregation networks is transformed is selected and solution.
To a large amount of crucial multicast application such as Web TV (IPTV; Internet protocol television) business etc.; ethernet ring network has many natural advantages: compare with traditional multistage direct-connected network; legacy network is on each second line of a couplet port of the branch node of multicast; all need to carry out duplicating of multicast data flow; ethernet ring network then only needs duplicate once on whole looped network, duplicates validity thereby improved multicast data flow greatly, and function such as link protection change is provided.
As shown in Figure 1, S1, S2, S3, S4, five nodes of S5 are formed ethernet ring network, and S1 is a host node.Under the normal condition, block S1 node the right from port, multicast data flow is transmitted along left direction.When the S3 node has the user to add or leaves,, also need set up multicast forward table at the S1 node except setting up the multicast forwarding off-balancesheet at the S3 node.After multicast forward table was set up and finished, as shown in Figure 2, multicast data flow could be passed through the S1 node and be arrived the S3 node, and the user receives multicast data flow at the S3 node, and the time that such mode makes multicast data flow arrive the user has bigger delay.When S2 node and S3 node link break down, as shown in Figure 3, the S1 node open rapidly obstruction from port, ethernet ring network switches multicast fast and spreads and broadcast the path, multicast data flow is transferred to the S3 node from the S1 node via S5, S4 node.But, need set up new multicast forward table at S1 host node and S3 node, after multicast forward table was built up, as shown in Figure 4, the S3 node can receive multicast data flow.
Wherein, unicast service relies on quick refresh address to transmit (FDB, Fluid Dynamic Bearings) and guarantees professional influence less than 50ms; Multicast service relies on because of special igmpinternet (IGMP, Internet Group Management Protocol), the registration of multicast protocol such as igmp snooping just can finish the forwarding of duplicating of multicast data flow, the foundation of multicast forward table just can be finished after needing CPU to handle multicast protocol.At this moment, need to set up multicast forward table again on the S1 node, switch though the ethernet ring network link has been finished fast, multicast forward table may still not set up complete, causes that multicast service postpones in link occurs fault greatly, recovery time is long.Link-recovery need rebulid multicast forward table equally, has increased the delay that multicast link switches greatly.
In the ethernet ring network, the bottleneck that multicast link switches fast is mainly in the foundation of multicast forward table, rather than the detection of ethernet ring network link failure recovers.Though can optimize the flow process of multicast protocol, accelerate the processing speed of multicast related protocol, the processing time is far longer than the control hierarchy of 50ms.
Summary of the invention
In view of this, main purpose of the present invention is to provide a kind of ethernet ring network multicast link fast switch over method, can support real-time reliable to switch fast.
For achieving the above object, technical scheme of the present invention is achieved in that
A kind of ethernet ring network multicast link fast switching method, this method comprises:
Abandon pattern for each multicast group in the ethernet ring network is provided with group, and under this pattern, set up multicast forward table for each multicast node; The looped network port of each node is added the multicast forward table of this node, and the looped network port of all nodes forms the multicast path of ethernet ring network, ethernet ring network just often, block host node from port, the multicast forward table of each node storage self;
When user or Link State changed in the ethernet ring network, multicast link carried out link switchover at the node relevant with variation, is specially:
Described user or Link State change for: have the user to add node;
Described multicast link switches to: add this user port in the multicast forward table of the node that the user adds, multicast data flow is forwarded to user port at this node,
Described user or Link State change for: have the user to leave node;
Described multicast link switches to: this user port of deletion the multicast forward table of the node that leaves from the user, and multicast data flow is discarded in user port at this node,
Described user or Link State change for: ethernet ring network detects link failure;
Described multicast link switches to: multicast data flow is forwarded to destination node from host node from port.
This method further comprises: specifying a node in ethernet ring network is host node, and as from port, the multicast forward table of host node comprises from port with a looped network port of this node.
This method further comprises: during the ethernet ring network link failure recovery, block host node from port, multicast data flow is abandoning from port direction.
When not having the user in the ethernet ring network, multicast data flow is only in the looped network port transmission of each node.
The present invention has introduced group and has abandoned pattern, under this pattern, the multicast forward table of each node of ethernet ring network exists always, all the time comprise the looped network port, when carrying out link switchover, multicast forward table does not need to rebulid on the looped network path, do not need to notify host node yet, thereby reduced because of rebuliding the delay that multicast forward table causes multicast link to switch, multicast group number or registered user's variation no longer influences the time of link switchover, has satisfied the high real-time of multicast service such as IPTV business, the requirement of high reliability.
Description of drawings
Fig. 1 is schematic diagram when ethernet ring network has the user to add under the existing pattern;
Fig. 2 is existing ethernet ring network user receiving group flow diagram;
Fig. 3 is a schematic diagram when having the ethernet ring network link failure now;
Fig. 4 switches back user's receiving group flow diagram for existing ethernet ring network link failure;
Fig. 5 is the flow chart of the method for the invention;
Fig. 6 is the multicast data flow schematic diagram of ethernet ring network of the present invention;
Fig. 7 adds/leaves schematic diagram for ethernet ring network user of the present invention;
Fig. 8 schematic diagram during for ethernet ring network link failure of the present invention;
Fig. 9 switches back user's receiving group flow diagram for ethernet ring network link failure of the present invention.
Embodiment
The present invention abandons (DROP) pattern for each multicast group is provided with group, and sets up empty multicast forward table in each multicast node.Under the DROP pattern, if there is not the user to add, the content of each node multicast forward table is empty, when the multicast data flow of this group arrives each node, because multicast forward table is empty, does not transmit object, then abandons the multicast data flow content.Simultaneously, under the DROP pattern, each node is added in the multicast path of multicast forward table of group, and at himself multicast forward table of each node storage, and multicast path is set to the whole forward modes of unknown multicast bag.More than be provided with and be static foundation, do not need the participation of agreement.
Like this, the multicast forward table of each node exists always, multicast data flow also exists on ethernet ring network all the time, node local multicast user adds or when leaving, multicast link and multicast data flow are switched synchronously, so, just can guarantee that multicast service is to be consistent the switching of multicast data flow and the time of ethernet ring network link switchover.
The concrete steps of the method for the invention are as shown in Figure 5:
Step 501: for each multicast group is provided with the DROP pattern, set up empty multicast forward table at each node, and the looped network port of each node is added the multicast forward table of this node, like this, the looped network port of all nodes forms the multicast path of ethernet ring network.
As shown in Figure 6, in ethernet ring network, appointment S1 is a host node, and the right looped network port of S1 node is from port, ethernet ring network just often, block the S1 node from port, the multicast forward table of each node storage self.When multicast data flow arrived, if there is not the user to add, multicast data flow only can be transmitted on ethernet ring network, formed the ring protection of multicast data flow, can not flow into other any port.
The multicast path of ethernet ring network comprises the looped network port of all nodes, and like this, each node just can in time be transmitted after upgrading the multicast forward table of self or receiving group flows, and needn't wait for that host node sets up multicast forward table.
When user or Link State changed in the ethernet ring network, multicast link carried out link switchover at the node relevant with variation.Concrete, add the user/leave under the situation of node or link failure, respectively execution in step 502 or step 503:
Step 502: the multicast user of certain node adds/leaves on the ethernet ring network, and multicast link switches at this node.
As shown in Figure 7, when having the user to add on the S3 node in the ethernet ring network, the user adds the multicast forward table of S3 node with user port after registration is finished in S3 node this locality, and like this, multicast path not only comprises the looped network port of S3 node, also comprises this user port.When multicast data flow arrived the S3 node, multicast data flow was forwarded on the user port that adds the S3 node immediately; At the node that does not have the user to add, multicast data flow only can be transmitted on the looped network port of node, can not be forwarded to other user access port.
When the S3 node had the user to leave, at this user port of multicast forwarding list deletion of S3 node, the multicast data flow of ethernet ring network when multicast data flow arrives the S3 node, was discarded in user port with multicast data flow because exist all the time.
Step 503: ethernet ring network detects certain node link fault, and multicast link switches at this node.
If the link occurs fault of ethernet ring network; as shown in Figure 8; disconnect between S2 node and the S3 node, ethernet ring network detects link failure, switches fast; the S1 node is decontroled the protection VLAN (VLAN from port; Virtual Local Area Network) forwarding capability because the S1 host node from port all the time in multicast forward table, multicast data flow is transmitted from blocking direction immediately; change by the S1 node through S4 node, S5 node, arrive the destination node S3 node at user place.
As shown in Figure 9, the S3 node receives multicast data flow by the link after switching.
Ethernet ring network detects fault when eliminating, block again the S1 host node from port, multicast data flow abandons in this outlet, prevents loop, multicast data flow exports uninterrupted forwarding and ethernet ring network keeps synchronously switching time at other simultaneously.Link-recovery between S2 node, the S3 node, it is normal that ethernet ring network detects link, and multicast data flow arrives the S3 node from the S1 host node through the S2 node and is forwarded to user port, does not need subscriber group information to rebulid on the S1 host node.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.
Claims (4)
1. ethernet ring network multicast link fast switching method is characterized in that this method comprises:
Abandon pattern for each multicast group in the ethernet ring network is provided with group, and under this pattern, set up multicast forward table for each multicast node; The looped network port of each node is added the multicast forward table of this node, and the looped network port of all nodes forms the multicast path of ethernet ring network, ethernet ring network just often, block host node from port, the multicast forward table of each node storage self;
When user or Link State changed in the ethernet ring network, multicast link carried out link switchover at the node relevant with variation, is specially:
Described user or Link State change for: have the user to add node;
Described multicast link switches to: add this user port in the multicast forward table of the node that the user adds, multicast data flow is forwarded to user port at this node,
Described user or Link State change for: have the user to leave node;
Described multicast link switches to: this user port of deletion the multicast forward table of the node that leaves from the user, and multicast data flow is discarded in user port at this node,
Described user or Link State change for: ethernet ring network detects link failure;
Described multicast link switches to: multicast data flow is forwarded to destination node from host node from port.
2. ethernet ring network multicast link fast switching method according to claim 1, it is characterized in that, this method further comprises: specifying a node in ethernet ring network is host node, and as from port, the multicast forward table of host node comprises from port with a looped network port of this node.
3. ethernet ring network multicast link fast switching method according to claim 1 is characterized in that this method further comprises: during the ethernet ring network link failure recovery, block host node from port, multicast data flow is abandoning from port direction.
4. ethernet ring network multicast link fast switching method according to claim 1 is characterized in that, when not having the user in the ethernet ring network, multicast data flow is only in the looped network port transmission of each node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2007101430365A CN101102249B (en) | 2007-08-21 | 2007-08-21 | A method for quickly switching Ethernet loop network multicast link |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2007101430365A CN101102249B (en) | 2007-08-21 | 2007-08-21 | A method for quickly switching Ethernet loop network multicast link |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101102249A CN101102249A (en) | 2008-01-09 |
| CN101102249B true CN101102249B (en) | 2011-10-26 |
Family
ID=39036352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2007101430365A Expired - Fee Related CN101102249B (en) | 2007-08-21 | 2007-08-21 | A method for quickly switching Ethernet loop network multicast link |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101102249B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101499950B (en) * | 2008-01-30 | 2011-12-28 | 华为技术有限公司 | Service provider backbone transmission ring multicast method, multicast ring network and node equipment |
| CN101304359B (en) * | 2008-05-28 | 2010-12-08 | 杭州华三通信技术有限公司 | Method and apparatus for RRPP looped network link recuperation |
| CN101667963B (en) * | 2009-09-09 | 2015-06-10 | 中兴通讯股份有限公司 | Link switching method and device |
| CN102598586A (en) * | 2011-10-29 | 2012-07-18 | 华为技术有限公司 | Method and devices for dealing multicast |
| CN103107946B (en) * | 2013-01-29 | 2016-08-03 | 华为技术有限公司 | Multicast recovery method and apparatus and include the trunking of this multicast recovery device |
| CN106330645B (en) * | 2015-06-16 | 2020-06-05 | 中兴通讯股份有限公司 | Method and device for fast convergence of Ethernet ring multicast Proxy mode service |
| CN109561023B (en) * | 2017-09-27 | 2022-03-11 | 华为技术有限公司 | Method, device and system for transmitting multicast message |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1424806B1 (en) * | 2002-11-26 | 2005-12-28 | Hitachi, Ltd. | Multicast Router with Translation Function for Protocols according to Any-Source-Multicast and Source-Specific-Multicast |
| CN1852246A (en) * | 2005-11-25 | 2006-10-25 | 华为技术有限公司 | Method for multi-cast broadcasting flow distribution node to receive multi-cast broadcasting flow |
-
2007
- 2007-08-21 CN CN2007101430365A patent/CN101102249B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1424806B1 (en) * | 2002-11-26 | 2005-12-28 | Hitachi, Ltd. | Multicast Router with Translation Function for Protocols according to Any-Source-Multicast and Source-Specific-Multicast |
| CN1852246A (en) * | 2005-11-25 | 2006-10-25 | 华为技术有限公司 | Method for multi-cast broadcasting flow distribution node to receive multi-cast broadcasting flow |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101102249A (en) | 2008-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101102249B (en) | A method for quickly switching Ethernet loop network multicast link | |
| US7826348B2 (en) | Multicast fast reroute | |
| CN101094175B (en) | Method, device, and system for protecting flux of multicast | |
| JP5543462B2 (en) | Improved convergence of multicast traffic | |
| US6654371B1 (en) | Method and apparatus for forwarding multicast data by relaying IGMP group membership | |
| US7684316B2 (en) | Multicast fast reroute for network topologies | |
| EP3340550B1 (en) | Service message multicast method and device | |
| EP1804423A2 (en) | Method for rapidly recovering multicast service and network device | |
| US20070127477A1 (en) | Method for implementing multicast based on multi-service transport platform | |
| US7719959B2 (en) | Achieving super-fast convergence of downstream multicast traffic when forwarding connectivity changes between access and distribution switches | |
| JP2005210631A (en) | Packet transfer device | |
| CN101291240B (en) | Multicast flow transmission method and system | |
| CN101656679A (en) | Method for rapidly converging multicast, router and communication system | |
| CN100477635C (en) | Transmission method and edge apparatus for multicast between fields | |
| WO2009146622A1 (en) | Method, router and system for implementing multicast | |
| CN106375223A (en) | SDN-based data forwarding system and method | |
| CN101702689A (en) | Transmission control method with balanced multicast service data load and access device thereof | |
| CN103339899A (en) | Multicast dual join for ring network topologies | |
| CN1996956A (en) | L3 network device and method for multi-cast cross-VLAN forwarding | |
| WO2012152134A1 (en) | Rapid convergence method for ethernet layer 2 multicast and ethernet system | |
| US11825534B2 (en) | Multicast replication in 5G networks | |
| CN101610200B (en) | Switching method and device of multicast routing | |
| CN103178976A (en) | Method for multicast tree establishment and fault restoration | |
| WO2016131359A1 (en) | Multicast line switching method and device for ring-shaped network | |
| CN101247541A (en) | Method for implementing multimedia multicast service of mobile communications network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20111026 Termination date: 20170821 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |