CN109660441A - The method and device of multicast replication in Overlay network - Google Patents

The method and device of multicast replication in Overlay network Download PDF

Info

Publication number
CN109660441A
CN109660441A CN201811591183.3A CN201811591183A CN109660441A CN 109660441 A CN109660441 A CN 109660441A CN 201811591183 A CN201811591183 A CN 201811591183A CN 109660441 A CN109660441 A CN 109660441A
Authority
CN
China
Prior art keywords
node
multicast
root node
subnet
website
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.)
Granted
Application number
CN201811591183.3A
Other languages
Chinese (zh)
Other versions
CN109660441B (en
Inventor
纪阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Dt Dream Technology Co Ltd
Original Assignee
Hangzhou Dt Dream Technology Co Ltd
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 Hangzhou Dt Dream Technology Co Ltd filed Critical Hangzhou Dt Dream Technology Co Ltd
Priority to CN201811591183.3A priority Critical patent/CN109660441B/en
Publication of CN109660441A publication Critical patent/CN109660441A/en
Application granted granted Critical
Publication of CN109660441B publication Critical patent/CN109660441B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

This application discloses the method and devices of multicast replication in Overlay network, this method determines the type of node first, the type of the node includes root node and non-root node, and the root node includes at least following one kind: including subnet root node, website root node, global root node;According to the type of the node and network topology structure obtained, IP address is issued into each node, and each node is notified to establish direct tunnel according to the IP address received, so that each node passes through the direct tunnel forwarding multicasting message after receiving multicast message.The application can reduce the duplication of the flux of multicast in Overlay network, save the cpu resource of headend, save the network bandwidth inside data center.

Description

The method and device of multicast replication in Overlay network
Technical field
This application involves the method and devices of multicast replication in Overlay network technology more particularly to Overlay network.
Background technique
In cloud computation data center, based on VXLAN (Virtual eXtensible Local Area Network, Virtual extended local area network) technology Overlay networking model under, needing the flow for carrying out multicast forwarding is mainly two layers wide Broadcast flow such as ARP, the messages such as DHCP, each virtual machine positioned at same two layers of domain of Overlay network or non-void at this time The physical host of quasi-ization all may be multicast source and multicast member.In current cloud computation data center, network is usual Will not deploying network devices multicast forwarding function, usually realized a little in such a way that interchanger is to the duplication of the head end of multicast message To the forwarding of multiple spot.
When the range of VXLAN network is very big, there are when multiple data centers especially in Overlay network, it is responsible for group The interchanger that casting text carries out head end duplication just needs to carry out a large amount of head end duplication, causes CPU to occupy excessive, waste of resource, Waste data center inside across pod (position of device, deployed with devices), across the network bandwidth between subnet.
Summary of the invention
The application provides a kind of method and device of multicast replication in Overlay network, can reduce in Overlay network Flux of multicast duplication, save the cpu resource of headend, save the network bandwidth inside data center.
According to the embodiment of the present application in a first aspect, provide a kind of method of multicast replication in Overlay network, run on On SDN controller, the method comprising the steps of:
Determine that the type of node, the type of the node include root node and non-root node, the root node includes at least Following one kind: including subnet root node, website root node, global root node;
According to the type of the node and network topology structure obtained, IP address is issued into each node, and notify each Node establishes direct tunnel according to the IP address received, so that each node passes through the direct tunnel after receiving multicast message Forwarding multicasting message.
According to the second aspect of the embodiment of the present application, a kind of device of multicast replication in Overlay network is provided, is located at On SDN controller, comprising:
Role's decision-making module, for determining that the type of node, the type of the node include root node and non-root node, institute Root node is stated including at least following one kind: including subnet root node, website root node, global root node;
Communication module issues IP address respectively for the type and network topology structure obtained according to the node Node, and notify each node according to the IP address received, so that each node passes through the direct-connected tunnel after receiving multicast message Road forwarding multicasting message.
It is of the invention using SDN controller to the centralized control of the interchanger in network, be arranged on SDN controller each The type of node establishes the tunnel between nodes at different levels according to different type, to established in virtual network comprising multistage The tree of node, multicast message is replicated and is forwarded layer by layer by nodes at different levels, to reduce the group in Overlay network The cpu resource of headend is saved in the duplication for broadcasting flow, saves the network bandwidth inside data center.
Detailed description of the invention
Fig. 1 is the more typical architecture diagram of Overlay network in the embodiment of the present application;
Fig. 2 is the flow chart of the method for multicast replication in Overlay network in the embodiment of the present application;
Fig. 2 a- Fig. 2 d is the path profile that each interchanger propagates multicast message under the network architecture shown in FIG. 1;
Fig. 2 e is the network topological diagram under the network architecture shown in FIG. 1 after tunnel generation;
Fig. 3 a is the architecture diagram of Overlay network in the application application example;
Fig. 3 b is the flow chart of the method for multicast replication in Overlay network in the application application example;
Fig. 4 is the hardware architecture diagram of the device of multicast replication in Overlay network in the embodiment of the present application;
Fig. 5 is the software logic block diagram of the device of multicast replication in Overlay network in the application application example.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended The example of the consistent device and method of some aspects be described in detail in claims, the application.
It is only to be not intended to be limiting the application merely for for the purpose of describing particular embodiments in term used in this application. It is also intended in the application and the "an" of singular used in the attached claims, " described " and "the" including majority Form, unless the context clearly indicates other meaning.It is also understood that term "and/or" used herein refers to and wraps It may be combined containing one or more associated any or all of project listed.
It will be appreciated that though various information, but this may be described using term first, second, third, etc. in the application A little information should not necessarily be limited by these terms.These terms are only used to for same type of information being distinguished from each other out.For example, not departing from In the case where the application range, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as One information.Depending on context, word as used in this " if " can be construed to " ... when " or " when ... When " or " in response to determination ".
Overlay network can be realized based on VXLAN technology.VXLAN is a kind of to carry out two layer message with three layer protocols The technology of encapsulation can be extended double layer network message in three-layer network range.
VXLAN can be applied to inside data center, migrate virtual machine within the scope of the three-layer network interconnected, Without changing IP address and MAC Address, guarantee the continuity of business.VXLAN uses the network identity of 24bit, makes user The mutually isolated virtual network of 16M can be created, breaches the limitation for the 4K isolation network that vlan technology can indicate, this makes It obtains and is provided with sufficient virtual network partitioned resources in the cloud environment of extensive multi-tenant.
VXLAN by the edge of physical network be arranged intelligent entity VTEP (VXLAN Tunnel End Point, The tunnel VXLAN terminal note), realize the isolation of virtual network and physical network.Tunnel is established between VTEP, on physical network The data frame of transfer of virtual network, physical network do not perceive virtual network.
Fig. 1 is a kind of structure chart for more typical VXLAN network that the application is applicable in.It include two VXLAN nets in figure Network, the website of each one data center of VXLAN network on behalf.Each VXLAN network generally includes sequentially connected physics clothes Business device 101, gateway 102 and router 103.
Each virtual machine (virtual machine 1 ... 16) and interchanger (interchanger 1 ... 8) can be deployed on physical server 101.It hands over It changes planes and can be used as the VTEP equipment of VXLAN network with gateway 102, realized virtual machine in virtual network by establishing tunnel Message other virtual machines are sent to by physical network.
Shown in Fig. 1, interchanger 1 and interchanger 2 are located on the same physical server 101, and two interchangers have Phase same network segment, different IP address.It is in this application that network segment is identical in same website, address is different interchanger is referred to as same Interchanger in subnet.
Interchanger (interchanger 1 and interchanger 3) in figure in subnet 1 and the interchanger in subnet 2 (interchanger 3 and exchange Machine 4) it is located in same VXLAN network, the interchanger of two subnets VXLAN range having the same, different network segments.In this Shen Please in, VXLAN range is identical, network segment is different interchanger is known as the interchanger of same website.
Interchanger (interchanger 5,6,7,8) in figure in VXLAN1 (interchanger 1,2,3,4) and VXLAN2 is located at different In VXLAN network, the interchanger VXLAN range in two networks is different.By the interchanger that VXLAN range is different in the application The interchanger of referred to as different websites.
SDN controller 104 is deployed in the application in Overlay network, passes through each website of SDN controller centralized control Interchanger.In one example, user can be pre-configured the switch device of Overlay network and SND controller is established and connected Connect, SDN controller by Openflow Netconf standard come with switch communication, naturally it is also possible to be other standards. SDN controller can establish the tunnel unicast VXLAN between all VTEP equipment, carry the stream of all VXLAN between VTEP Amount.
Fig. 2 is the working principle of SDN controller in the application.Virtual machine is known as multicast source or multicast member below, it will Interchanger is known as node.
S201 determines that the type of node, the type of node include root node and non-root node, and root node includes at least following It is a kind of: including subnet root node, website root node, global root node;
IP address is issued each node, and notify each according to the type of node and network topology structure obtained by S202 Node establishes direct tunnel according to the IP address received, so that each node passes through the direct tunnel after receiving multicast message Forwarding multicasting message.
For S201, in one embodiment, in order to determine the type of node, the available each node of SDN controller Processing capacity parameter and location information;Processing capacity parameter may include whether to support Multicast Replication Capability parameter, multicast multiple Ability parameter processed etc.;Address information includes site parameter belonging to interchanger, subnet parameter, network segment parameter etc..According to node Location information can calculate the connection relationship between each node, reach each multicast member when institute from multicast source so as to selected The node of process.It as an example, can also include virtual platform in the Overlay network of the application (reference can be made to figure Virtual platform in 3a), communication connection channel is established between SDN controller and virtual platform, carries out information exchange.SDN The processing capacity parameter and location information of each node acquired in controller may come from virtual platform.
Virtual platform is for realizing the management to each virtual machine.User can be pre-configured virtual platform and interchanger The physical server at place establishes management connection, by relevant standard message interface communication, controls virtual on physical server The life cycle of machine, creation, deletion, starting, stopping including virtual machine etc., in addition virtual platform can also manage all The hardware resource of physical server include CPU, the information such as memory and location information.
Virtual platform is by the processing capacity parameter (such as CPU ability and number) of each physical server, location information It notifies to give SDN controller, so that SDN controller is able to the Multicast Replication Capability safeguarded and saved one about interchanger, physics The VTEP IP address for the interchanger disposed on the position of server, physical server and the database of network segment, in this application, The database that these information are stored on SDN controller can be known as multicast replication node policy database.In addition, when virtualization , can be by the deploying virtual machine to some physical server when platform disposes virtual machine in Overlay network, and this is virtual Machine is connected to external network by the interchanger on the physical server, and to each Microsoft Loopback Adapter of virtual machine association one A VXLAN, MAC, IP address.Virtual platform can give these information announcements to SDN controller, so that SDN controller be made to know The interchanger of this virtual machine connection.
By the root nodes at different levels root node that duplicated multicast E-Packets to next stage step by step in the application, therefore which selects The strategy that node serves as root node can refer to the processing capacity parameter of each node.Whether processing capacity parameter can be the node With Multicast Replication Capability, the grade of Multicast Replication Capability etc..It in one embodiment, can be with when selecting root node at different levels The highest node of preferred multicast replication capacity, is highest node if there is two or more Multicast Replication Capabilities, then may be used To select one of lesser node of VTEP IP address.
Specifically, it when determining global root node, can be waited being used as in the node of websites all in Overlay network Object is selected, wherein global root node will be configured to by the highest node of Multicast Replication Capability, if there is multiple Multicast Replication Capabilities Identical node then wherein can will be configured to global root node by the lesser node of VTEP IP address.When determining certain website Website root node when, candidate target can will be used as in all nodes in some website, will wherein Multicast Replication Capability be most High node is configured to website root node, then can will wherein VTEP if there is the identical node of multiple Multicast Replication Capabilities The lesser node of IP address is configured to website root node.When determining subnet root node, can will own in the subnet It is used as candidate target in node, wherein subnet root node will be configured to by the highest node of Multicast Replication Capability, if there is multiple The identical node of Multicast Replication Capability then wherein can will be configured to subnet root node by the lesser node of VTEP IP address.
By taking Fig. 1 as an example, when multicast member of the interchanger 3 as interchanger 1, since the two is in different subnets, Therefore it needs to be determined that the website root node of two subnets of the subnet root node of each subnet and connection, it is assumed that exchange 1,2,3,4 Be provided with Multicast Replication Capability, and interchanger 3 is identical with the Multicast Replication Capability of interchanger 4, then when selecting subnet root node, VTEP IP address lesser one be can choose in interchanger 3 and interchanger 4 as 4 place subnet of interchanger 3 and interchanger Subnet root node.
Table 1 lists the processing capacity parameter for the node that a SDN controller is stored and the example of location information.
Table 1
Table 2 is according to a kind of calculated example as root nodes at different levels of table 1.
Table 2
In one embodiment, selecting which node to serve as the strategy of root node can also include that this current node has been served as Root node relevant information, consider the current cpu load situation of the node to take into account when selecting root node.The letter of record Breath may include VXLAN network identity, in the VXLAN network node at corresponding tunnel both ends IP address list.
For S202, since the message in VXLAN network needs to transmit in physical network by tunnel, according to After the mode that S201 is disclosed determines nodes at different levels, need to notify to establish the tunnel established between each node, therefore SDN controller is also It needs to notify to need the node established tunnel and do not set up tunnel to establish tunnel in tunnel, detailed process may is that
SDN controller traverses selected all root nodes and recipient,
If judging that this node is not subnet root node, and subnet root node tunnel is not set up, then the node is notified to build The vertical tunnel with subnet root node;The path between recipient in same subnet is the tunnel that the two arrives subnet root node, is connect Path between receipts person and subnet root node is tunnel between the two.
If judging that this node is subnet root node, and the node is not set up to the tunnel of website root node, then notice should Node establishes the tunnel with website root node;Path between the recipient of different sub-network is that the two arrives respective subnet root node Tunnel and respective subnet root node are to the tunnel of website root node.
If judging that this node is website root node, and the node is not set up to the tunnel of global root node, then notice should Node establishes the tunnel with global root node;Path between the recipient of different websites is that the two arrives respective subnet root The tunnel of node, the tunnel of respective subnet root node to website root node and website root node are to the tunnel of global root node.
After establishing tunnel, two nodes in tunnel are traversed, if VXLAN is not added for corresponding tunnel on the node, These tunnels are added in corresponding VXLAN.
Nodes at different levels in network may include the root node from multicast source to each multicast member, and with subnet root node Non-root node (hereinafter referred to as recipient) in same subnet.According to the location of multicast member difference, multicast forwarding The tunnel number that path includes can be different.Each node can be created as by executing S201 and S201 step with tree phase The network structure to connect, it can be seen that according to the established network structure of institute, subnet root node be responsible for by multicast message according to Number of nodes duplicated multicast message in same subnet, and other nodes being transmitted in same subnet, and/or by multicast message Issue website root node;Website root node is responsible for the quantity copy group by multicast message according to the subnet root node in same website Casting text, and the subnet root node being transmitted in same website, and/or multicast message is issued into global root node;Global root section Point is responsible for multicast message according to the quantity duplicated multicast message of website root node, and is transmitted to website root node.
For example, root node can be subnet root node when multicast source and multicast member are in same subnet, at this point, The message of multicast source needs to issue subnet root node, and the multicast forwarding path passed through, which can be, is transmitted to son by subnet root node Each node in net.
When multicast source and multicast member are in the different sub-network of same website, root node can be subnet root node and station Point root node, at this point, the message of multicast source needs to issue the subnet root node of place subnet 1, the multicast forwarding path passed through It can be and multicast message is transmitted to website root node by the subnet root node of subnet 1, website root node forwards multicast message To the subnet root node of subnet 2 where multicast member.
When multicast source and multicast member are in different websites, root node may include global root node, website root node With subnet root node, at this point, multicast source message need by the subnet root node of place subnet 1 be transmitted to where website 1 station Point root node, the multicast forwarding path passed through, which can be, is transmitted to global root for multicast message by the website root node of website 1 Node, as the website root node of the website 2 where global root node is transmitted to multicast member, then by the website root node of website 2 The subnet root node of subnet 2 where being transmitted to multicast member.The application in such cases can solve to cross in the prior art The problem of network bandwidth between data center is wasted caused by multiple duplication among data.
Still with the data instance of Tables 1 and 2, the network that SDN controller carries out after root node selection and tunnel foundation is opened up It flutters shown in Fig. 2 e, it is seen that can be formed in a network at this time using global root node as the multicast tree of root.
As one embodiment, for the renewal process in tunnel, when there is new multicast member to be added or exit multicast When group, virtual platform notifies the physical server information of IP, VXLAN of the virtual machine, place to give SDN controller.SDN control Device processed obtains the interchanger of this virtual machine connection.
In one example, the case where new multicast member is added, if the direct-connected interchanger of the multicast member is Already existing interchanger in Overlay network then can obtain the new road that multicast member is added according to established tunnel before Diameter;If the direct-connected interchanger of the multicast member is the interchanger being not present in Overlay network, using this interchanger as The recipient of the subnet where it is located, and the interchanger is notified to establish the tunnel in the interchanger and this subnet between subnet root node, To obtain that the path of multicast member is newly added.
The case where for revocation virtual machine, if the direct-connected interchanger of the virtual machine is also connected with other interchangers, Then without updating multicast forwarding tree channel.If the interchanger does not connect the virtual machine in other same websites, carry out Beta pruning processing, i.e., if the interchanger does not reconnect any other virtual machine of the VXLAN, and only one tunnel is located at the station In point, then it will be deleted in the tunnel slave site;The opposite equip. in the tunnel is similarly handled.
Fig. 2 a- Fig. 2 d is that multicast source is virtual machine 2, when multicast member is virtual machine 1,3,4,5,6,7,8 according to S201 and The tunnel that S202 step is established carries out the schematic diagram in the path of multicast forwarding.
Such as Fig. 2 a, since interchanger 1 and interchanger 2 belong to the node in same subnet, the multicast message of virtual machine 2 Reach the node experienced of virtual machine 1 are as follows: interchanger 2 (the direct-connected node of multicast source) --- (the subnet root node of interchanger 1 1);
Such as Fig. 2 b, since interchanger 3 and interchanger 2 belong to the node of different sub-network, the multicast message of virtual machine 2 reaches To the node experienced of virtual machine 3 are as follows: (the subnet root node 1 and website 1 of subnet 1.1.1.0 of interchanger 2 --- interchanger 1 Root node) --- interchanger 3 (the subnet root node 2 of subnet 1.1.2.0);
Such as Fig. 2 c, since interchanger 4 and interchanger 2 belong to the node of different sub-network, the multicast message of virtual machine 2 reaches To the node experienced of virtual machine 4 are as follows: (the subnet root node 1 and website 1 of subnet 1.1.1.0 of interchanger 2 --- interchanger 1 Website root node) --- interchanger 3 (the subnet root node 2 of subnet 1.1.2.0) --- interchanger 4;
Such as Fig. 2 d, since interchanger 5 and interchanger 2 belong to the node of different websites, the multicast message of virtual machine 2 reaches To the node in the website experienced 1 of virtual machine 5: interchanger 2 --- interchanger 1 (the subnet root node of subnet 1.1.1.0, The website root node of website 1, global root node);Node in website 2 is as shown in number 1 in figure: (the station of website 2 of interchanger 6 The subnet root node 3 of point root node and 2.2.2.0) --- interchanger 5 (recipient);
Since interchanger 6 and interchanger 2 belong to the node of different websites, the multicast message of virtual machine 2 reaches virtual The node experienced of machine 6 are as follows: interchanger 2 --- interchanger 1 (the subnet root node of subnet 1.1.1.0 and the website root of website 1 Section, global root node);Node in website 2 is as shown in number 2 in figure: interchanger 6 (the website root node of website 2 and 2.2.2.0 subnet root node 3);
Since interchanger 7 and interchanger 2 belong to the node of different websites, the multicast message of virtual machine 2 reaches virtual The node experienced of machine 7 are as follows: interchanger 2 --- interchanger 1 (the subnet root node of subnet 1.1.1.0 and the website root of website 1 Node, global root node);Node in website 2 is as shown in number 3 in figure: interchanger 6 (the website root node of website 2) --- Interchanger 8 (the subnet root node 4 of 3.3.3.0) --- interchanger 7 (recipient);
Since interchanger 8 and interchanger 2 belong to the node of different websites, the multicast message of virtual machine 2 reaches virtual The node experienced of machine 8 are as follows: (the website root of the subnet root node of subnet 1.1.1.0, website 1 of interchanger 2 --- interchanger 1 Node, global root node);Node in website 2 is as shown in number 4 in figure: interchanger 6 (the website root node of website 2) --- Interchanger 8 (the subnet root node 4 of 3.3.3.0).
Fig. 3 a is the network architecture diagram under the concrete application scene of the application.It include two in the Overlay network A VXLAN network and SDN controller and virtual platform.SDN controller and virtual platform are by managing network to each Physical server where interchanger is managed, wherein SDN controller management interchanger, and virtual platform manages virtual machine.
Using SDN controller to the centralized control of interchanger, strategy letter of the user based on the input multicast replication of SDN controller Breath, when the physical server where interchanger disposes virtual machine, the VXLAN where the virtual machine is come identification switch The range of VXLAN, physical server where the virtual machine identify whether VXLAN crosses over data center, according to exchange The IP address of machine identifies whether the VXLAN crosses over different network segments, and SDN controller obtains each friendship from virtual platform The CPU Multicast Replication Capability changed planes, to integrate the node for determining responsible multicast replication.
Realize that the process of the multicast replication scheme of the application is as follows in this example:
Referring to Fig. 3 b, S301, user is pre-configured network;
A, user is pre-configured the interchanger of Overlay network and SND controller establishes connection, the communication protocols of SDN controller View is Openflow.
B, the physical server where user is pre-configured virtual platform and interchanger establishes management connection, keeps virtualization flat Platform can control the life cycle of virtual machine on physical server by relevant standard message interface communication.
C, communication connection channel is established between SDN controller and virtual platform, for interacting the relevant information of virtual machine.
S302, virtual platform information relevant to SDN controller interaction virtual machine.Virtual platform is by all interchangers The hardware resource of the physical server at place includes the interchanger disposed on Multicast Replication Capability parameter and physical server VTEP IP address and network segment information issue SDN controller.
S303, SDN controller select root nodes at different levels and recipient.
SDN controller is according to the relevant information for being obtained about virtual machine in S302, with reference to each interchanger multicast Replication capacity select multicast source between multicast member root node and recipient.
Assuming that the relevant information of each interchanger stored in the multicast replication node policy database of SDN is as shown in table 1, The root node then selected can be as shown in table 2.
S304, the tunnel connection for notifying each node to establish between nodes at different levels.The IP of two end nodes can be carried in notice Address.The tree that tunnel after foundation is constituted is as shown in Figure 2 e.
After node experienced between selected multicast source and multicast member, all interdependent nodes are traversed, if this section Point is not subnet root node, and does not set up subnet root node tunnel, then the node is notified to establish to subnet root node tunnel;Such as This node of fruit is subnet root node, and does not set up the node to the tunnel of website root node, then the node is notified to establish to website Root node tunnel;If it is this node website root node, and the node is not set up to the tunnel of global root node, then notify the section Point is established to global root node tunnel.
S305: when interchanger 1 is used as multicast source to 2,3,4,5,6,7 transmission group sowing time of interchanger, each node is according to built Vertical tunnel duplication and forwarding multicasting message.It is specific as follows:
Interchanger 1 is used as subnet root node, and there are the tunnel between interchanger 1 and interchanger 2, therefore interchanger 1 is by group 1 part of duplication of casting text is transmitted to interchanger 2;
Since interchanger 1 is global root node, interchanger 1 no longer needed to after completing the multicast replication in this subnet by Message forwarded upstream directly starts to down forward.Since the lower website root node connected of interchanger 1 is interchanger 1 and interchanger 6, therefore need to only replicate 1 part of multicast message and issue interchanger 6;
For website 1, interchanger 1 is used as website root node, and the subnet root node connected is interchanger 1 and interchanger 3, Therefore interchanger 1 replicates 1 part of multicast message again and issues interchanger 3;Interchanger 3 is used as subnet root node, replicates 1 part of multicast message Issue interchanger 4.
For website 2, interchanger 6 is used as website root node, replicates 1 part of multicast message and issue interchanger 8;Interchanger 6 is made For subnet root node, replicates 1 part of multicast message and issue recipient's interchanger 5 in this subnet;Interchanger 8 is used as subnet root section Point replicates 1 part of multicast message and issues recipient's interchanger 7 in this subnet.
Corresponding with the embodiment of method of multicast replication in aforementioned Overlay network, present invention also provides Overlay The embodiment of the device of multicast replication in network.
The embodiment of the device of multicast replication can be applied on SDN controller in the application Overlay network.Device is real Applying example can also be realized by software realization by way of hardware or software and hardware combining.Taking software implementation as an example, make For the device on a logical meaning, being will be corresponding in nonvolatile memory by the processor of SDN controller where it Computer program instructions are read into memory what operation was formed.For hardware view, as shown in figure 4, being the application Overlay A kind of hardware structure diagram of SDN controller where the device of multicast replication in network, in addition to processor shown in Fig. 4, memory, net Except network interface and nonvolatile memory, the SDN controller in embodiment where device is generally according to the SDN controller Actual functional capability, can also include other hardware, this is repeated no more.
Referring to FIG. 5, in Overlay network multicast replication device 500, be located at SDN controller on, comprising:
Role's decision-making module 501, for determining that the type of node, the type of node include root node and non-root node, root Node includes at least following one kind: including subnet root node, website root node, global root node;
IP address is issued each section according to the type of the node and network topology structure obtained by communication module 502 Point, and each node is notified to establish direct tunnel according to the IP address received, so that each node passes through after receiving multicast message The direct tunnel forwarding multicasting message.
As one embodiment, role's decision-making module 501 determines that node type specifically includes:
Multicast Replication Capability and CPU will be supported in the strongest node of Multicast Replication Capability in same subnet or same subnet Ability highest, the smallest node in address are configured to subnet root node;
Multicast Replication Capability and CPU will be supported in the strongest node of Multicast Replication Capability in same website or same website Ability highest, the smallest node in address are configured to website root node;
Multicast Replication Capability will be supported in all websites and branch in the strongest node of Multicast Replication Capability or all websites It holds Multicast Replication Capability and CPU ability highest, the smallest interchanger in address is configured to global root node.
Communication module 502 notifies each node to establish direct tunnel and can specifically include:
If judging that the node is not subnet root node, and subnet root node tunnel is not set up, then the node is notified to build Found the tunnel of the node Yu subnet root node;
If judging that the node is subnet root node, and the node is not set up to the tunnel of website root node, then notice should Node establishes the tunnel of the node Yu website root node;
If judging that the node is website root node, and the node is not set up to the tunnel of global root node, then notice should Node establishes the tunnel of node foundation with global root node.
Communication module 502 can be also used for:
When cancelling multicast source or multicast member, if direct-connected node does not connect the multicast of other same websites Source or multicast member, and only one tunnel is located in the website, then the node is notified to delete the tunnel in the website of place;Such as The peer node that fruit is located at the node of the tunnel other end does not connect the multicast source or multicast member of other same websites, and Only one tunnel is located in the website, then the peer node is notified to delete the tunnel in the website of place.
Communication module 502 can be also used for:
When there is multicast member addition, if the direct-connected node of the multicast member is the node being not present in network, The node is notified to establish the tunnel in the node and this subnet between subnet root node.
Role's decision-making module 501 can be also used for obtaining from virtual platform in the Overlay network from each node Manage ability parameter and location information;Processing capacity parameter includes whether to support Multicast Replication Capability parameter, Multicast Replication Capability ginseng Number;The address information includes site parameter belonging to the interchanger, subnet parameter, network segment parameter;And the storage place Manage ability parameter and location information.
The function of each unit and the realization process of effect are specifically detailed in the above method and correspond to step in above-mentioned apparatus Realization process, details are not described herein.
For device embodiment, since it corresponds essentially to embodiment of the method, so related place is referring to method reality Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, wherein described be used as separation unit The unit of explanation may or may not be physically separated, and component shown as a unit can be or can also be with It is not physical unit, it can it is in one place, or may be distributed over multiple network units.It can be according to actual The purpose for needing to select some or all of the modules therein to realize application scheme.Those of ordinary skill in the art are not paying Out in the case where creative work, it can understand and implement.
The foregoing is merely the preferred embodiments of the application, not to limit the application, all essences in the application Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the application protection.

Claims (10)

1. a kind of method of multicast replication in Overlay network, which is characterized in that the method is applied in Overlay network Connect the node of multicast source;The Overlay network includes at least one website, includes at least one son in each website Net;Each node is attached to form multicast tree with tree in the Overlay network, the multicast tree are as follows: the multicast The root node of tree is global root node, and global root node has direct tunnel with the foundation of the website root node of each website respectively;Often The subnet root node foundation of each subnet included by one website root node and the website has direct tunnel;Each subnet Non-root node foundation in root node and the subnet has direct tunnel;
The described method includes:
Receive the multicast message that multicast source is sent;
Determine the position in the Overlay network of multicast member;
According to position of the multicast member in the Overlay network, multicast forwarding path is determined from the multicast tree, according to The multicast member is sent to according to the multicast forwarding path.
2. the method according to claim 1, wherein the multicast source is in the first subnet of the first website;
The position according to multicast member in the Overlay network, determines multicast forwarding road from the multicast tree Diameter, comprising:
If section of the multicast member in first subnet, from the multicast forwarding path determined in the multicast tree Point is successively are as follows: the subnet root node of the first subnet and the direct-connected node of multicast member, multicast member.
3. the method according to claim 1, wherein the multicast source is in the first subnet of the first website;
The position according to multicast member in the Overlay network, determines multicast forwarding road from the multicast tree Diameter, comprising:
If the multicast forwarding that the multicast member in the second subnet of the second website, is determined from the multicast tree Node on path is successively are as follows: the subnet root node of the first subnet, the website root node of the first website, global root node, second The website root node of website, the subnet root node of the second subnet and the direct-connected node of multicast member, the multicast member.
4. the method according to claim 1, wherein the multicast tree is established in the following way:
SDN controller in the Overlay network determines the type of each node, and if judging that the node is not subnet root Node, and do not set up subnet root node tunnel, then notify the node to establish the tunnel of the node Yu subnet root node;Sentence If the disconnected node is subnet root node, and does not set up the node to the tunnel of website root node, then the node is notified to build Found the tunnel of the node Yu website root node;If judging that the node is website root node, and the node is not set up to entirely The tunnel of office's root node then notifies the node to establish the tunnel of the node foundation and global root node.
5. the method according to claim 1, wherein
The node type of this node is non-root node;
Alternatively,
The node type of this node includes following one or more kinds of combination:
The subnet root node of subnet where this node;
The website root node of website where this node;
Global root node.
6. the device of multicast replication in a kind of Overlay network, which is characterized in that described device is applied in Overlay network Connect the node of multicast source;The Overlay network includes at least one website, includes at least one son in each website Net;Each node is attached in the Overlay network with tree, forms multicast tree;The multicast tree are as follows: multicast tree Root node is global root node, and global root node has direct tunnel with the foundation of the website root node of each website respectively;Each The subnet root node foundation of each subnet included by website root node and the website has direct tunnel;Each subnet root section Point has direct tunnel with the non-root node foundation in the subnet;
Described device includes:
Receiving unit, for receiving the multicast message of multicast source transmission;
Determination unit, for determining the position in the Overlay network of multicast member;
Transmission unit determines group from the multicast tree for the position according to multicast member in the Overlay network Multicast forwarding path is sent to the multicast member according to the multicast forwarding path.
7. device according to claim 6, which is characterized in that the multicast source is in the first subnet of the first website;
The transmission unit is determined from the multicast tree in the position according to multicast member in the Overlay network When multicast forwarding path, if being specifically used for the multicast member in first subnet, determined from the multicast tree Multicast forwarding path on node successively are as follows: the subnet root node of the first subnet, with the direct-connected node of multicast member, multicast at Member.
8. device according to claim 6, which is characterized in that the multicast source is in the first subnet of the first website;
The transmission unit is determined from the multicast tree in the position according to multicast member in the Overlay network When multicast forwarding path, if specifically for the multicast member in the second subnet of the second website, from the multicast tree The node on the multicast forwarding path determined is successively are as follows: the website root of the subnet root node of the first subnet, the first website Node, global root node, the website root node of the second website, the second subnet subnet root node, with the direct-connected section of multicast member Point, the multicast member.
9. device according to claim 6, which is characterized in that the multicast tree in the Overlay network passes through such as lower section Formula is established:
SDN controller in the Overlay network determines the type of each node, and if judging that the node is not subnet root Node, and do not set up subnet root node tunnel, then notify the node to establish the tunnel of the node Yu subnet root node;Sentence If the disconnected node is subnet root node, and does not set up the node to the tunnel of website root node, then the node is notified to build Found the tunnel of the node Yu website root node;If judging that the node is website root node, and the node is not set up to entirely The tunnel of office's root node then notifies the node to establish the tunnel of the node foundation and global root node.
10. device according to claim 6, which is characterized in that the node type of this node is non-root node;
Alternatively,
The node type of this node includes following one or more kinds of combination:
The subnet root node of subnet where this node;
The website root node of website where this node;
Global root node.
CN201811591183.3A 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network Active CN109660441B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811591183.3A CN109660441B (en) 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811591183.3A CN109660441B (en) 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network
CN201510628198.2A CN105162704B (en) 2015-09-28 2015-09-28 The method and device of multicast replication in Overlay network

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN201510628198.2A Division CN105162704B (en) 2015-09-28 2015-09-28 The method and device of multicast replication in Overlay network

Publications (2)

Publication Number Publication Date
CN109660441A true CN109660441A (en) 2019-04-19
CN109660441B CN109660441B (en) 2021-05-28

Family

ID=54803463

Family Applications (4)

Application Number Title Priority Date Filing Date
CN201811591183.3A Active CN109660441B (en) 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network
CN201811590077.3A Active CN109561033B (en) 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network
CN201510628198.2A Active CN105162704B (en) 2015-09-28 2015-09-28 The method and device of multicast replication in Overlay network
CN201811591207.5A Active CN109660442B (en) 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network

Family Applications After (3)

Application Number Title Priority Date Filing Date
CN201811590077.3A Active CN109561033B (en) 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network
CN201510628198.2A Active CN105162704B (en) 2015-09-28 2015-09-28 The method and device of multicast replication in Overlay network
CN201811591207.5A Active CN109660442B (en) 2015-09-28 2015-09-28 Method and device for multicast replication in Overlay network

Country Status (1)

Country Link
CN (4) CN109660441B (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106301941B (en) * 2016-08-29 2019-08-27 广州西麦科技股份有限公司 Mono- key dispositions method of Overlay and system
CN106411732B (en) * 2016-09-07 2020-11-20 新华三技术有限公司 Message forwarding method and device
CN108011800B (en) * 2016-10-31 2020-12-08 中国电信股份有限公司 Virtual extensible local area network VXLAN deployment method and VXLAN gateway
CN109327375B (en) * 2017-08-01 2021-04-30 中国电信股份有限公司 Method, device and system for establishing VXLAN tunnel
CN107872385B (en) * 2017-10-11 2020-10-23 中国电子科技集团公司第三十研究所 SDN network routing calculation and control method
CN109412978A (en) * 2018-10-17 2019-03-01 郑州云海信息技术有限公司 A kind of unicast method, virtual switch, SDN controller and storage medium
CN110445889B (en) * 2019-09-20 2020-06-02 中国海洋大学 Method and system for managing IP address of switch under Ethernet environment
CN113452551B (en) * 2021-06-11 2022-07-08 烽火通信科技股份有限公司 VXLAN tunnel topology monitoring method, device, equipment and storage medium
CN113365253B (en) * 2021-06-15 2023-10-03 上海高仙自动化科技发展有限公司 Node communication method, device, equipment, system and storage medium in network
CN114285679A (en) * 2021-12-09 2022-04-05 武汉船舶通信研究所(中国船舶重工集团公司第七二二研究所) Method and system for realizing heterogeneous network multicast based on centralized control
CN115051890A (en) * 2022-05-20 2022-09-13 中国电信股份有限公司 Message processing method, system, device, electronic equipment and storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1761226A (en) * 2005-11-22 2006-04-19 中国科学院计算机网络信息中心 The separation and the fusion of hierarchical switched network IPv6 address space
CN1791054A (en) * 2004-12-14 2006-06-21 华为技术有限公司 Virtual special network multicast method by virtual router mode
CN1996931A (en) * 2005-12-31 2007-07-11 迈普(四川)通信技术有限公司 Network multicast method
CN101237393A (en) * 2007-01-30 2008-08-06 华为技术有限公司 A method and device and system for realizing quick multicast service switch
CN102195855A (en) * 2010-03-17 2011-09-21 华为技术有限公司 Business routing method and business network
CN102316030A (en) * 2011-09-01 2012-01-11 杭州华三通信技术有限公司 Method for realizing two-layer internetworking of data center and device
CN102739501A (en) * 2011-04-01 2012-10-17 中兴通讯股份有限公司 Message forwarding method in two or three layer virtual private network (VPN) and system thereof
CN104115453A (en) * 2013-12-31 2014-10-22 华为技术有限公司 Method and device for achieving virtual machine communication
CN104253698A (en) * 2013-06-29 2014-12-31 华为技术有限公司 Message multicast processing method and message multicast processing equipment

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1622551A (en) * 2004-12-15 2005-06-01 中国科学院计算机网络信息中心 Internal service system of layered type switching network and management control method thereof
CN102571616B (en) * 2012-03-16 2017-09-19 中兴通讯股份有限公司 Merging, dividing method, tunnel-associated device and the router in tunnel
CN104871495B (en) * 2012-09-26 2018-07-13 华为技术有限公司 Virtual superposition gateway for stacking network
CN104702476B (en) * 2013-12-05 2018-07-31 华为技术有限公司 Message processing method based on distributed network gate and network virtualization marginal point
US10135687B2 (en) * 2014-01-06 2018-11-20 Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Virtual group policy based filtering within an overlay network
CN104301251B (en) * 2014-09-22 2018-04-27 新华三技术有限公司 A kind of QoS processing methods, system and equipment
CN104702480B (en) * 2015-03-24 2018-10-02 华为技术有限公司 The method and apparatus that protecting tunnel group is established in next-generation multicasting virtual private network

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1791054A (en) * 2004-12-14 2006-06-21 华为技术有限公司 Virtual special network multicast method by virtual router mode
CN1761226A (en) * 2005-11-22 2006-04-19 中国科学院计算机网络信息中心 The separation and the fusion of hierarchical switched network IPv6 address space
CN1996931A (en) * 2005-12-31 2007-07-11 迈普(四川)通信技术有限公司 Network multicast method
CN101237393A (en) * 2007-01-30 2008-08-06 华为技术有限公司 A method and device and system for realizing quick multicast service switch
CN102195855A (en) * 2010-03-17 2011-09-21 华为技术有限公司 Business routing method and business network
CN102739501A (en) * 2011-04-01 2012-10-17 中兴通讯股份有限公司 Message forwarding method in two or three layer virtual private network (VPN) and system thereof
CN102316030A (en) * 2011-09-01 2012-01-11 杭州华三通信技术有限公司 Method for realizing two-layer internetworking of data center and device
CN104253698A (en) * 2013-06-29 2014-12-31 华为技术有限公司 Message multicast processing method and message multicast processing equipment
CN104115453A (en) * 2013-12-31 2014-10-22 华为技术有限公司 Method and device for achieving virtual machine communication

Also Published As

Publication number Publication date
CN105162704A (en) 2015-12-16
CN109660442A (en) 2019-04-19
CN109561033B (en) 2021-08-24
CN109660442B (en) 2021-04-27
CN105162704B (en) 2019-01-25
CN109561033A (en) 2019-04-02
CN109660441B (en) 2021-05-28

Similar Documents

Publication Publication Date Title
CN105162704B (en) The method and device of multicast replication in Overlay network
EP3188409B1 (en) Oam mechanisms for evpn active-active services
CN104270298B (en) Message forwarding method and device in a kind of VXLAN networks
CN102549983B (en) Distributed data center access switch
EP2725737B1 (en) Network policy configuration method, management device and network management centre device
US10263808B2 (en) Deployment of virtual extensible local area network
CN110050445B (en) Method, device and system for sending and receiving message
CN107948041B (en) Method and equipment for constructing VXLAN centralized multi-active gateway
EP3069471B1 (en) Optimized multicast routing in a clos-like network
CN104584491A (en) System and method providing distributed virtual routing and switching (DVRS)
CN108964940A (en) Message method and device, storage medium
CN107306215B (en) Data processing method, system and node
CN106357542B (en) The method and software defined network controller of multicast service are provided
CN105577502A (en) Service transport method and device
US20190014035A1 (en) Information transmission method and device
CN104144143B (en) Method and control device that network is set up
CN103138955B (en) Multiple igmp querying device is used in layer 2 network
US20190215191A1 (en) Deployment Of Virtual Extensible Local Area Network
CN104994019B (en) A kind of horizontal direction interface system for SDN controllers
KR20180122513A (en) Method and framework for traffic engineering in network hypervisor of sdn-based network virtualization platform
US20180198708A1 (en) Data center linking system and method therefor
CN108900422A (en) Multicast forward method, device and electronic equipment
JP5063726B2 (en) Configuration control method for virtual node device
CN106878051B (en) Multi-machine backup implementation method and device
KR20150079391A (en) Method and apparatus for network virtualization

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant