CN104468144B - A kind of method of space multicast, apparatus and system - Google Patents
A kind of method of space multicast, apparatus and system Download PDFInfo
- Publication number
- CN104468144B CN104468144B CN201410710701.4A CN201410710701A CN104468144B CN 104468144 B CN104468144 B CN 104468144B CN 201410710701 A CN201410710701 A CN 201410710701A CN 104468144 B CN104468144 B CN 104468144B
- Authority
- CN
- China
- Prior art keywords
- node
- multicast
- intermediate node
- target
- group
- 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.)
- Active
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
A kind of method of space multicast, apparatus and system, wherein method include：First intermediate node receives native multicasting group, and the native multicasting group contains the content-data for needing multicast and at least two multicast destination nodes；Obtain the second set of intermediate nodes and the bandwidth of the second intermediate node；It includes at least two second intermediate nodes that second intermediate node, which is concentrated, second intermediate node is the intermediate node for having between the multicast destination node direct connection relational, and at least one second intermediate node only has direct connection relational between partaof destination node；The content-data is replicated to the second intermediate node for being distributed to second intermediate node and concentrating, and indicates that the second intermediate node for receiving content-data is transmitted to the multicast destination node for having direct connection relational with it.It can realize that multicast linear speed in space replicates in the non-exchange system connected entirely.
Description
Technical field
The present invention relates to field of communication technology, more particularly to a kind of method of space multicast, apparatus and system.
Background technology
In the function of modern communication equipment, support that multicast (also crying multicast) has been a necessary functions, so-called group
Broadcast refers to that identical data are sent to multiple destination addresses.
The data plane of communication equipment is usually made of line card and switching network, and line card generally includes physical interface, physical layer
With link layer process chip, network processing unit, this several part of switching network interface chip, this direction is known as from line card to switching network
The direction uplink (ingress), this direction is known as the direction downlink (egress) from switching network to line card.
Switching network has single-stage exchange and multistage exchange, single-stage to exchange commonly used in small exchange system (such as individual equipment),
A switching network is passed through in communication between all line cards；If building more massive exchange system (such as multimachine frame cluster
System), it usually needs it is exchanged using multistage, such as current industry uses 3 grades of more CLOS or Benes exchange systems.With 3 grades
For CLOS is exchanged, line card and the 1st, 3 grade of switching network chip are directly connected to, 1,3 grade of switching network and 2 grades of switching network chip connections,
Data interchange between line card will be completed by 3 grades of switching network chips, can be separate between 3 grades of switching network chips
The function of certain grades of switching networks can also be integrated into 1 chip and be realized, such as：Usually inside with 1 switching network chip
Realize 1,3 grade of exchange.
In a communications device there are many kinds of the realization methods of multicast, the ground that is occurred in a device according to the action of multicast replication
Fang Butong can be divided into uplink duplication, switching network duplication and downlink and replicate these types, these types of multicast replication can also be combined
Get up to use.Uplink replicates the uplink switching network interface chip typically in line card and completes；Switching network duplication is then in switching network
Chip interior completes multicast replication, and progress multicast replication, which is called, inside switching network does space multicast；And downlink duplication usually exists
It is completed inside downlink network processor chips.
Usual uplink replicates or switching network is replicated and completed from source plate to the duplication of purpose veneer, and downlink duplication is accomplished to mesh
Veneer above duplication between each port.
Multicast replication can be completed once, can also be completed by multiple copies.Multicast source node is usually called root section
The final purpose node of point, multicast is known as leaf node, and copy-point when multistage replicates in addition to root node and leaf node claims
For intermediate node.
Current space multicast implementation is required when selection carries out the intermediate node of multicast duplication must select energy
See that the intermediate node of all leaf nodes could be replicated, and the intermediate node that can only see some leaf node cannot be into
Row multicast replicates.In this way in the non-exchange system connected entirely, multicast can have replicability loss of energy when replicating, and cannot achieve
Multicast linear speed in space replicates.
Connection switching network refers to entirely：
1) in single-stage switching system, own in all switching network interface chips (being referred to as switching port) and system
There is physical connection between switching network chip.
2) in multi-stage switching system, all switching network interface chips and the switching network chip being connected directly therewith can divide
At multigroup (each group is defined as a full link field), the switching network interface chip inside each full link field and the institute in the group
It is also totally interconnected for having between switching network chip.It is then realized by another grade of switching network chip between group totally interconnected.It is complete as shown in Figure 1
1~N of line card and 1~M of switching network chip has physical connection in link field.
Conversely, when above-mentioned full link field inside points switching network interface chip without and the domain between all switching network chips
It is just the non-switching network system connected entirely when having physical connection.In the full link field of the left side as shown in Figure 1 square region, line
The switching network interface chip of card 1 without and switching network chip 1~M realize and connect entirely, and only with which part switching network chip it
Between have physical connection (dotted arrow is illustrated as no connection).
Using the above space Multicast Scheme, when carrying out multicast duplication in switching network, all leaf nodes can be only seen
Intermediate node could carry out multicast duplication, and the intermediate node that can only see some leaf node cannot carry out multicast duplication.
In this way in the non-exchange system connected entirely, can there are replicability loss of energy, nothing between each leaf node when multicast replicates
Method realizes that multicast linear speed in space replicates.
Invention content
An embodiment of the present invention provides a kind of method of space multicast, apparatus and systems, in the non-exchange connected entirely
In system, realize that multicast linear speed in space replicates.
On the one hand the embodiment of the present invention provides a kind of method of space multicast, including：
First intermediate node receive native multicasting group, the native multicasting group contain need the content-data of multicast with
And at least two multicast destination node；
First intermediate node obtains the second set of intermediate nodes and the bandwidth of the second intermediate node；In described second
It includes at least two second intermediate nodes that intermediate node, which is concentrated, and second intermediate node is to have between the multicast destination node
The intermediate node of direct connection relational, and at least one second intermediate node only has between partaof destination node directly
Connection relation；
First intermediate node, which replicates the content-data, to be distributed in second intermediate node is concentrated second
Intermediate node, and indicate that the second intermediate node for receiving content-data is transmitted to the multicast purpose section for having direct connection relational with it
Point.
In conjunction with the realization method of one side, in the first possible implementation, first intermediate node will be described
Content-data, which replicates, is distributed to the second intermediate node that second intermediate node is concentrated, and indicates to receive the of content-data
Two intermediate nodes are transmitted to the multicast destination node for having direct connection relational with it：
The native multicasting group is split as at least two target multicast groups by first intermediate node；At least one mesh
Multicast group is marked only comprising the partaof destination node in at least two multicasts destination node；
The multicast destination node and target multicast group that target multicast group is included by first intermediate node are corresponding interior
Hold data and is sent to the second intermediate node of target；The second intermediate node of the target is the multicast purpose for including with target multicast group
There is the intermediate node of direct connection relational between node.
In conjunction with the first possible realization method of one side, in second of possible realization method, if in the presence of at least
Two native multicasting groups, it is described native multicasting group is split as at least two target multicast groups to include：
Strategy is split to determine according to the result of calculation of the load-balancing mechanism of intermediate node, and according to fractionation strategy
Each native multicasting group is split as at least two target multicast groups.
In conjunction with the first possible realization method of one side, in the third possible realization method, if in the presence of at least
The combination of two kinds of target multicast groups includes the complete or collected works of at least two multicasts destination node, and first intermediate node is by target
The multicast destination node and the corresponding content-data of target multicast group that multicast group includes are sent to target the second intermediate node packet
It includes：
First intermediate node determines target middle node according to the result of calculation of the load-balancing mechanism of intermediate node
Point, and multicast destination node and the corresponding content-data of target multicast group that target multicast group includes are sent to target
Two intermediate nodes.
In conjunction on the one hand, one side the first, second or the third possible realization method, in the 4th kind of possibility
Realization method in, first intermediate node be second level switching network node, the second intermediate node be first order switching network section
Point or third level switching network node；The multicast destination node is line card.
In conjunction with the realization method of one side, in the 5th kind of possible realization method, if in the presence of with more than two multicasts
There is the second intermediate node of direct connection relational between destination node；Then first intermediate node replicates the content-data
Being distributed to the second intermediate node that second intermediate node is concentrated includes：
Send the content-data after merging has direct connection relational to described between more than two multicast destination nodes
The second intermediate node.
Two aspect of the embodiment of the present invention provides a kind of device of space multicast, is used as the first intermediate node, including：
Multicast group receiving unit, for receiving native multicasting group, the native multicasting group contains the content for needing multicast
Data and at least two multicast destination nodes；
Node acquiring unit, for obtaining the second set of intermediate nodes and the bandwidth of the second intermediate node；In described second
It includes at least two second intermediate nodes that intermediate node, which is concentrated, and second intermediate node is to have between the multicast destination node
The intermediate node of direct connection relational, and at least one second intermediate node only has between partaof destination node directly
Connection relation；
Dispatching Unit, for the content-data to be replicated to the second middle node for being distributed to second intermediate node and concentrating
Point, and indicate that the second intermediate node for receiving content-data is transmitted to the multicast destination node for having direct connection relational with it.
In conjunction with the realization method of two aspects, in the first possible implementation, the Dispatching Unit includes：
Split cells, for the native multicasting group to be split as at least two target multicast groups；At least one target
Multicast group only includes the partaof destination node in at least two multicasts destination node；
Transmission sub-unit, the multicast destination node for including by target multicast group and the corresponding content of target multicast group
Data are sent to the second intermediate node of target；The second intermediate node of the target is the multicast purpose section for including with target multicast group
There is the intermediate node of direct connection relational between point.
In conjunction with the first possible realization method of two aspects, in second of possible realization method, if in the presence of at least
Two native multicasting groups；
The split cells, specifically for determining fractionation plan according to the result of calculation of the load-balancing mechanism of intermediate node
Slightly, and according to fractionation strategy by each native multicasting group at least two target multicast groups are split as.
In conjunction with the first possible realization method of two aspects, in the third possible realization method, if in the presence of at least
The combination of two kinds of target multicast groups includes the complete or collected works of at least two multicasts destination node；
The transmission sub-unit, specifically for determining target according to the result of calculation of the load-balancing mechanism of intermediate node
Intermediate node, and the multicast destination node and the corresponding content-data of target multicast group that target multicast group includes are sent to mesh
Mark the second intermediate node.
In conjunction with two aspect, two aspect the first, second or the third possible realization method, in the 4th kind of possibility
Realization method in, first intermediate node be second level switching network node, second intermediate node be the first order exchange
Net node or third level switching network node；The multicast destination node is line card.
In conjunction with two aspect realization method, in the 5th kind of possible realization method, if in the presence of with more than two multicasts
There is the second intermediate node of direct connection relational between destination node；
The Dispatching Unit, for sends the content-data after merging to described and more than two multicast destination nodes it
Between have the second intermediate node of direct connection relational.
Three aspect of the embodiment of the present invention additionally provides a kind of system of space multicast, including：First intermediate node, in second
Intermediate node, root node and leaf node；Wherein root node is multicast source node, and leaf node is multicast destination node；It is described
Second intermediate node is the intermediate node for having between the multicast destination node direct connection relational；And second middle node
Point is non-full connection with leaf node；First intermediate node is the device of any one provided in an embodiment of the present invention.
As can be seen from the above technical solutions, the embodiment of the present invention has the following advantages：It determines in a full link field
The second set of intermediate nodes, and there is the second intermediate node of part only to be directly connected to close between partaof destination node
System, the duplication of the second intermediate node in the non-switching network system connected entirely can be made full use of using the embodiment of the present invention
Performance, therefore can realize that multicast linear speed in space replicates in the non-exchange system connected entirely.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without having to pay creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is prior art exchange system structural schematic diagram；
Fig. 2 is present invention method flow diagram；
Fig. 3 is the flow direction structure schematic diagram of the non-multi-stage switching system connected entirely of the embodiment of the present invention；
Fig. 4 is the flow direction structure schematic diagram of the non-multi-stage switching system connected entirely of the embodiment of the present invention；
Fig. 5 is the flow direction structure schematic diagram of the non-multi-stage switching system connected entirely of the embodiment of the present invention；
Fig. 6 is the flow direction structure schematic diagram of the non-multi-stage switching system connected entirely of the embodiment of the present invention；
Fig. 7 is the flow direction structure schematic diagram of the non-multi-stage switching system connected entirely of the embodiment of the present invention；
Fig. 8 is the flow direction structure schematic diagram of the non-multi-stage switching system connected entirely of the embodiment of the present invention；
Fig. 9 is the flow direction structure schematic diagram of the non-multi-stage switching system connected entirely of the embodiment of the present invention；
Figure 10 is apparatus structure schematic diagram of the embodiment of the present invention；
Figure 11 is apparatus structure schematic diagram of the embodiment of the present invention；
Figure 12 is apparatus structure schematic diagram of the embodiment of the present invention；
Figure 13 is apparatus structure schematic diagram of the embodiment of the present invention.
Specific implementation mode
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into
It is described in detail to one step, it is clear that the described embodiments are only some of the embodiments of the present invention, rather than whole implementation
Example.Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts
All other embodiment, shall fall within the protection scope of the present invention.
An embodiment of the present invention provides a kind of methods of space multicast, as shown in Fig. 2, including：
101：First intermediate node receives native multicasting group, and above-mentioned native multicasting group contains the content-data for needing multicast
And at least two multicast destination node；
The present embodiments relate to be the non-exchange system connected entirely, multicast destination node is exactly leaf node, second
Intermediate node is the intermediate node for having between multicast destination node direct connection relational, and the second intermediate node is not
There can be direct connection relational between all multicast destination nodes；In addition, the first intermediate node is and multicast purpose section
There is no the intermediate node of direct connection relational between point, it can be considered that being other middle nodes in addition to the second intermediate node
Point.
102：Above-mentioned first intermediate node obtains the second set of intermediate nodes and the bandwidth of the second intermediate node；Above-mentioned
Two intermediate nodes concentrate include at least two second intermediate nodes, above-mentioned second intermediate node be with above-mentioned multicast destination node it
Between have the intermediate node of direct connection relational, and at least one second intermediate node only has between partaof destination node
Direct connection relational；
Preferably, the second intermediate node that above-mentioned second intermediate node is concentrated belongs to the same full link field.
103：Above-mentioned first intermediate node the above Data replication and distribution is concentrated to above-mentioned second intermediate node the
Two intermediate nodes, and indicate that the second intermediate node for receiving content-data is transmitted to the multicast mesh for having direct connection relational with it
Node.
The embodiment of the present invention, it is determined that the second set of intermediate nodes in a full link field, and have among part second
Node only has direct connection relational between partaof destination node, can be made full use of using the embodiment of the present invention non-
The replication performance of the second intermediate node in the switching network system connected entirely, thus can in the non-exchange system connected entirely,
Realize that multicast linear speed in space replicates.
The embodiment of the present invention additionally provides the specific implementation for replicating distribution content-data, as follows：Among above-mentioned first
The above Data replication and distribution is given the second intermediate node that above-mentioned second intermediate node is concentrated by node, and indicates to receive interior
Second intermediate node of appearance data is transmitted to the multicast destination node for having direct connection relational with it and includes：
Above-mentioned native multicasting group is split as at least two target multicast groups by above-mentioned first intermediate node；At least one mesh
Multicast group is marked only comprising the partaof destination node in above-mentioned at least two multicasts destination node；
The multicast destination node and target multicast group that target multicast group is included by above-mentioned first intermediate node are corresponding interior
Hold data and is sent to the second intermediate node of target；The second intermediate node of above-mentioned target is the multicast purpose for including with target multicast group
There is the intermediate node of direct connection relational between node.
In the present embodiment, native multicasting group target multicast group has been split as, the management to multicast group can have been facilitated, separately
Outside, it is compatible can also to realize that the mechanism with existing space multicast carries out for next stage intermediate node.
When native multicasting group has very much, fractionation mode has very much, and an embodiment of the present invention provides specific
Fractionation mode selects as follows：If above-mentioned that native multicasting group is split as at least two targets there are at least two native multicasting groups
Multicast group includes：
Strategy is split to determine according to the result of calculation of the load-balancing mechanism of intermediate node, and according to above-mentioned fractionation strategy
Each native multicasting group is split as at least two target multicast groups.
The present embodiment determines the fractionation strategy to multicast group by way of load balancing, can preferably utilize and exchange
Physical link bandwidth resources between the replication capacity of net and switching node at different levels, to improve system data throughput.
After native multicasting group is split as target multicast group, if target multicast group more than two, then it is possible that meeting
The case where capable of realizing native multicasting group function there are the combination of the second intermediate node of plurality of target, therefore base of the embodiment of the present invention
It is specific as follows in the selection scheme that this gives the second intermediate node of specific target：If there are at least two target multicasts
The combination of group includes the complete or collected works of above-mentioned at least two multicasts destination node, and target multicast group is included by above-mentioned first intermediate node
Multicast destination node and the corresponding content-data of target multicast group are sent to the second intermediate node of target：
Above-mentioned first intermediate node determines target middle node according to the result of calculation of the load-balancing mechanism of intermediate node
Point, and the multicast destination node and the corresponding content-data of target multicast group that target multicast group includes are sent to target second
Intermediate node.
The present embodiment determines the selection to the second intermediate node of target by way of load balancing, can be preferably sharp
With the physical link bandwidth resources between the replication capacity and switching node at different levels of switching network, handle up to improve system data
Amount.
As a typical realization method, switching network can be for 3 grades of exchange systems so, and above-mentioned first intermediate node is
Second level switching network node, the second intermediate node are first order switching network node or third level switching network node.It needs to illustrate
, the series of switching network can be more in the embodiment of the present invention, therefore above example should not be construed as implementing the present invention
The uniqueness of example limits.Such as：Switching network is 5 grades, then the first intermediate node can be second, third, fourth stage switching network
The general designation of node, the second intermediate node are first or the 5th switching network node.
As the realization limited a citing, above-mentioned multicast destination node is line card.Multicast destination node is multicast packet
Equipment where the destination address to be sent to, therefore it is not limited in line card, line card should not be managed as the citing of a typical scenario
Solution is to the restriction of the uniqueness of the embodiment of the present invention.
Directly connect since the second intermediate node may have between the multicast destination node of more than two (including two)
Relationship is connect, there can be the phenomenon that bandwidth waste if the first intermediate node replicates more parts of data, the embodiment of the present invention can be right
This is optimized：If in the presence of the second intermediate node for having direct connection relational between more than two multicast destination nodes；Then
The above Data replication and distribution is given the second intermediate node packet that above-mentioned second intermediate node is concentrated by above-mentioned first intermediate node
It includes：
Send the content-data after merging has direct connection relational to above-mentioned between more than two multicast destination nodes
The second intermediate node.
In embodiments of the present invention, combined content-data only has a content-data, thus can be by above-mentioned two
The corresponding more than two content-datas of above multicast destination node are merged into portion.
Following embodiment will provide a specific example, be carried out to the embodiment of the present invention more detailed for example, with 3 grades
For exchange system：
As shown in figure 3, for the non-multi-stage switching system connected entirely, partial line card is only in N number of line card and S13#1 has physics company
It connects, other line cards then have physical connection with all S13 switching networks chips.This exchange system is the non-exchange connected entirely
Net.Assuming that the bandwidth of line card 1,2 is A, the bandwidth of line card 3 and N are 2A, and line card N sends out two multicast data flows, correspond to two groups respectively
Group G1 and G2 is broadcast, member's (i.e. multicast destination node) of G1 includes line card 1 and line card 3, and G2 includes line card 2 and line card 3.Two groups
The bandwidth for broadcasting flow is all A, according to the space multicast implementation of background technology, due to not having object between S13#2 and line card 1,2
Reason connection, therefore it cannot carry out multicast replication, only could execute multicast replication in S2 and S13#1.Specific data flow is such as
Solid line between arrow direction shown in Fig. 3, line card and switching network node and other switching network net nodes indicates connection relation, real
Line arrow indicates the upward propagation process of the data flow of G1 and G2, and dotted arrows are the space multicast reproduction process of G2 flows, empty
Line arrow is the space multicast reproduction process of G1 flows.3 downlink bandwidth is insufficient from S13#1 to line card as seen from Figure 3
To support the flux of multicast of 2A bandwidth, that is to say, that the ability of line card N to multicast replication between line card 3 only has A.And it is practical uncommon
Prestige can reach 2A, therefore can not realize the space multicast replication performance requirement of this scene.
It should also be noted that Fig. 3 and its citing are Multicast Replication Performance defect in this non-exchange system connected entirely
1 kind of scene, other examples can also be enumerated, such as the flux of multicast for being A from line card 1 and 2 each one bandwidth of hair, corresponding G1 and
G2, G1 include line card 2 and 3, and G2 includes line card 1 and 3, at this moment it is expected that the outlet flux of multicast bandwidth of line card 3 is 2A, but practical
Upper only A, and cannot achieve.Therefore the above citing should not be construed as limiting the uniqueness of the embodiment of the present invention.
Following embodiment will be provided realizes that the realization method of space multicast is illustrated using strategy of the embodiment of the present invention：
One,
The flow of multicast group G1 can be inquired each after line card N uplinks reach switching network S2 before S2 is replicated to S13
Interconnected relationship (be referred to as routing table) of the destination address (being exactly line card here) between each S13 switching networks chip, according to
Query result, which is can be found that between multicast member and S13 switching network chips, 2 kinds of connection relations, and G1 is then split into two mesh
Multicast group is marked, G1-1 and G1-2 is labeled as, wherein G1-1 is corresponded in former multicast group and connected pass with 1 switching network having the same of line card
The multicast member of system combines；G1-2 is corresponded in former multicast group and the multicast member of 3 switching network connection relation having the same of line card
Combination.Then new flux of multicast load balancing table is accordingly generated, the corresponding flux of multicast of G1-1 only copies to S13#1, G1-2
Corresponding flux of multicast then carries out load balancing between S13#1 and S13#2, finally copies to corresponding target multicast group in S13
Member.
Specific data flow arrow direction as shown in Figure 4, line card and switching network node and other switching network net nodes it
Between solid line indicate connection relation, solid arrow indicate G1 data flow upward propagation process, dotted arrows be G1-2 flows
Space multicast reproduction process, dotted arrow be G1-1 flows space multicast reproduction process.
Space multicasting technology based on the present invention is identical with the operation of G1 in the present embodiment to the duplication of G2 in previous example,
The flow of multicast group G2 can inquire each destination address after line card N uplinks reach switching network S2 before S2 is replicated to S13
The interconnected relationship (being referred to as routing table) of (being exactly line card here) between each S13 switching networks chip, according to query result
It was found that there are 2 kinds of connection relations between multicast member and S13 switching network chips, G2 is then split into two target multicast groups, is marked
Be denoted as G2-1 and G2-2, wherein G2-1 correspond in former multicast group and the multicast of 2 switching network connection relation having the same of line card at
Member's combination；G2-2 is corresponded in former multicast group and the multicast member of 3 switching network connection relation having the same of line card combines.Then evidence
This generates new flux of multicast load balancing table, and the corresponding flux of multicast of G2-1 only copies to S13#1, the corresponding multicast data flows of G2-2
Amount then carries out load balancing between S13#1 and S13#2, and the member of corresponding target multicast group is finally copied in S13.
Specific data flow arrow direction as shown in Figure 5, line card and switching network node and other switching network net nodes it
Between solid line indicate connection relation, solid arrow indicate G2 data flow upward propagation process, dotted arrows be G2-2 flows
Space multicast reproduction process, the arrow of dashed interval point is the space multicast reproduction process of G2-1 flows.
In summary the implementation description of G1 and G2, it can be seen that multiple using the space multicast of strategy of the embodiment of the present invention
After technology processed, the multicast replication bandwidth of line card N to line card 3 can reach expected 2A rather than the attainable A of background technology institute.
The space multicasting technology of the embodiment of the present invention can also advanced optimize, to reduce between S2 and S13 switching networks
The occupied bandwidth resources of multicast copy flow, further increase Multicast Replication Performance.Optimization method illustrates the description seen below.
Two,
Data flowchart after being optimized to the implementation described in previous embodiment is as shown in fig. 6, Fig. 6 is to preceding
State the schematic diagram of the duplication flow of G1 multicast groups in example.The difference is that it can when being replicated from S2 to S13 with embodiment one
To be controlled number of copies according to the interconnected relationship between multicast member and S13, number of copies is reduced to the greatest extent to reduce
Demand to switching network intraconnection bandwidth.By taking G1 as an example, after G1 is split into G1-1 and G1-2 by S2, according to routing table lookup
As a result know that the member of G1-1 can only be reached by S13#1, and the member of G1-2 can then be reached by S13#1 and S13#2, that
S2, can be right the flow load sharing of G1-2 to S13#1 and S13#2 when being replicated to the corresponding flux of multicast of G1-2
In the flow for sharing S13#1, S2 can merge with G1-1's multicast replication to reduce object between S2 and S13#1 G1-2's
The flow (flux of multicast as shown in bold dashed lines in Fig. 6 between S2 to S13#1) for managing chain road, for this merging
Flow has corresponding mark on cell or heading and informs downstream chip (being exactly S13#1 here), and downstream chip is right
The flow of this merging can stamp correct multicast to every part of copy according to this mark when copying to final multicast member and identify, example
Such as the copy for being copied to line card 1, the multicast of filling is identified as G1-1, and for being copied to the copy of line card 3, the group of filling
It broadcasts and is identified as G1-2.
S2 shares G1-2 corresponding flows the part flow of S13#2 in the middle, can not merge with the flow of G1-1,
These flows respectively must replicate portion to S13#1 and S13#2 simultaneously, as shown in Figure 7.
Specific data flow arrow direction as shown in Figure 6, between line card and switching network node and other switching network nodes
Solid line indicate that connection relation, solid arrow indicate that the upward propagation process of the data flow of G1, dotted arrows are G1-2 flows
Space multicast reproduction process, dotted arrow are the space multicast reproduction process of G1-1 flows.Thicker dotted arrow be G1-2 and
G1-1 merges the space multicast reproduction process of flow.Fig. 6 is the part flux of multicast that can merge after G1 multicast groups are split
Flow direction and reproduction process schematic diagram
Specific data flow arrow direction as shown in Figure 7, between line card and switching network node and other switching network nodes
Solid line indicate that connection relation, solid arrow are expressed as the upward propagation process of the data flow of G1, dotted arrows are G1-2 flows
Space multicast reproduction process, dotted arrow be G1-1 flows space multicast reproduction process Fig. 7 be G1 multicast groups split after not
The flow direction and reproduction process schematic diagram of combined part flux of multicast.
Identical to the optimization processing and G1 of G2, which is not described herein again, is carried out to G2 multicast traffics using after Optimized Measures
Shown in data flowchart following Fig. 8 and Fig. 9 of duplication.
Specific data flow arrow direction as shown in Figure 8, between line card and switching network node and other switching network nodes
Solid line indicate that connection relation, solid arrow indicate that the upward propagation process of the data flow of G2, dotted arrows are G2-2 flows
Space multicast reproduction process, the arrow of dashed interval point are the space multicast reproduction process of G2-1 flows.Fig. 8 is G1 multicast assembling and dismantling
Divide the flow direction and reproduction process schematic diagram of latter incorporated part flux of multicast.Thicker dotted arrow is that G2-2 and G2-1 merges
The space multicast reproduction process of flow.
Specific data flow arrow direction as shown in Figure 9, line card and switching network node and other switching network net nodes it
Between solid line indicate connection relation, solid arrow indicate G2 data flow upward propagation process, dotted arrows be the spaces G2-2
The arrow of multicast reproduction process, dashed interval point is the multicast reproduction process of the spaces G2-1.Fig. 9 is nonjoinder after G2 multicast groups are split
Space multicast reproduction process.
The present invention solves in the non-exchange system connected entirely arbitrary multicast member group between each space multicast leaf node
The problem of closing the linear speed replication performance in the multicast group of (multicast group) so that networking mode can very flexibly, and can solve
When being caused using the non-switching network scheme connected entirely due to the difference of different generation switching network chip interface specifications when system upgrade
Multicast availability issue.
The embodiment of the present invention additionally provides a kind of device of space multicast, is used as the first intermediate node, such as Figure 10 institutes
Show, including：
Multicast group receiving unit 1001, for receiving native multicasting group, above-mentioned native multicasting group, which contains, needs multicast
Content-data and at least two multicast destination nodes；
Node acquiring unit 1002, for obtaining the second set of intermediate nodes and the bandwidth of the second intermediate node；Above-mentioned
Two intermediate nodes concentrate include at least two second intermediate nodes, above-mentioned second intermediate node be with above-mentioned multicast destination node it
Between have the intermediate node of direct connection relational, and at least one second intermediate node only has between partaof destination node
Direct connection relational；
Dispatching Unit 1003, for giving the above Data replication and distribution in the second of above-mentioned second intermediate node concentration
Intermediate node, and indicate that the second intermediate node for receiving content-data is transmitted to the multicast purpose section for having direct connection relational with it
Point.
Optionally, as shown in figure 11, above-mentioned Dispatching Unit 1003 includes：
Split cells 1101, for above-mentioned native multicasting group to be split as at least two target multicast groups；At least one
Target multicast group only includes the partaof destination node in above-mentioned at least two multicasts destination node；
Transmission sub-unit 1102, the multicast destination node and target multicast group for including by target multicast group are corresponding
Content-data is sent to the second intermediate node of target；The second intermediate node of above-mentioned target is the multicast mesh for including with target multicast group
Node between have the intermediate node of direct connection relational.
Optionally, if there are at least two native multicasting groups；
Above-mentioned split cells 1101 is torn open specifically for the result of calculation of the load-balancing mechanism according to intermediate node to determine
Divide strategy, and each native multicasting group is split as at least two target multicast groups according to above-mentioned fractionation strategy.
Optionally, if the combination there are at least two target multicast groups is complete comprising above-mentioned at least two multicasts destination node
Collection；
Above-mentioned transmission sub-unit 1102, specifically for being determined according to the result of calculation of the load-balancing mechanism of intermediate node
Target intermediate node, and the corresponding content-data of multicast destination node and target multicast group for including by target multicast group is sent
Give target the second intermediate node.
Optionally, above-mentioned first intermediate node is second level switching network node, and above-mentioned second intermediate node is handed over for the first order
Draping node or third level switching network node, above-mentioned multicast destination node are line card.
Optionally, if in the presence of the second middle node for having direct connection relational between more than two multicast destination nodes
Point；
Above-mentioned Dispatching Unit 1003, for sending the content-data after merging to above-mentioned and more than two multicast purpose sections
There is the second intermediate node of direct connection relational between point.
The embodiment of the present invention additionally provides a kind of system of space multicast, as shown in figure 12, including：First intermediate node
1201, the second intermediate node 1202, root node 1203 and leaf node 1204；Wherein root node 1203 is multicast source node,
Leaf node 1204 is multicast destination node；Above-mentioned second intermediate node 1202 is to have between above-mentioned multicast destination node directly
The intermediate node of connection relation；And above-mentioned second intermediate node 1202 and leaf node are non-full connection；Above-mentioned first middle node
Point 1201 is the device of any one provided in an embodiment of the present invention.
The embodiment of the present invention, it is determined that the second set of intermediate nodes in a full link field, and have among part second
Node only has direct connection relational between partaof destination node, can be made full use of using the embodiment of the present invention non-
The replication performance of the second intermediate node in the switching network system connected entirely, thus can in the non-exchange system connected entirely,
Realize that multicast linear speed in space replicates.
The embodiment of the present invention additionally provides the device of another space multicast, is used as the first intermediate node, such as Figure 13
It is shown, including：Receiver 1301, transmitter 1302, processor 1303 and memory 1304；
Wherein, processor 1303, for receiving native multicasting group, above-mentioned native multicasting group contains the content for needing multicast
Data and at least two multicast destination nodes；Obtain the second set of intermediate nodes and the bandwidth of the second intermediate node；Above-mentioned
Two intermediate nodes concentrate include at least two second intermediate nodes, above-mentioned second intermediate node be with above-mentioned multicast destination node it
Between have the intermediate node of direct connection relational, and at least one second intermediate node only has between partaof destination node
Direct connection relational；It gives the above Data replication and distribution to the second intermediate node that above-mentioned second intermediate node is concentrated, and refers to
Show that the second intermediate node for receiving content-data is transmitted to the multicast destination node for having direct connection relational with it.
The present embodiments relate to be the non-exchange system connected entirely, multicast destination node is exactly leaf node, second
Intermediate node is the intermediate node for having between multicast destination node direct connection relational, and the second intermediate node is not all
There can be direct connection relational between all multicast destination nodes；In addition, the first intermediate node is and multicast destination node
Between there is no the intermediate node of direct connection relational, it can be considered that being other middle nodes in addition to the second intermediate node
Point.
The embodiment of the present invention, it is determined that the second set of intermediate nodes in a full link field, and have among part second
Node only has direct connection relational between partaof destination node, can be made full use of using the embodiment of the present invention non-
The replication performance of the second intermediate node in the switching network system connected entirely, thus can in the non-exchange system connected entirely,
Realize that multicast linear speed in space replicates.
The embodiment of the present invention additionally provides the specific implementation for replicating distribution content-data, as follows：
Above-mentioned processor 1303, for giving the second of above-mentioned second intermediate node concentration by the above Data replication and distribution
Intermediate node, and indicate that the second intermediate node for receiving content-data is transmitted to the multicast purpose for having direct connection relational with it
Node includes：For above-mentioned native multicasting group to be split as at least two target multicast groups；At least one target multicast group is only
Including the partaof destination node in above-mentioned at least two multicasts destination node；The multicast purpose for including by target multicast group
Node and the corresponding content-data of target multicast group are sent to the second intermediate node of target；The second intermediate node of above-mentioned target is
There is the intermediate node of direct connection relational between the multicast destination node for including with target multicast group.
In the present embodiment, native multicasting group target multicast group has been split as, the management to multicast group can have been facilitated, separately
Outside, it is compatible can also to realize that the mechanism with existing space multicast carries out for next stage intermediate node.
When native multicasting group has very much, fractionation mode has very much, and an embodiment of the present invention provides specific
Fractionation mode selects as follows：If there are at least two native multicasting groups, above-mentioned processor 1303, for splitting native multicasting group
Include at least two target multicast groups：For determining fractionation plan according to the result of calculation of the load-balancing mechanism of intermediate node
Slightly, and each native multicasting group is split as at least two target multicast groups according to above-mentioned fractionation strategy.
The present embodiment determines the fractionation strategy to multicast group by way of load balancing, can preferably utilize and exchange
Physical link bandwidth resources between the replication capacity of net and switching node at different levels, to improve system data throughput.
After native multicasting group is split as target multicast group, if target multicast group more than two, then it is possible that meeting
The case where capable of realizing native multicasting group function there are the combination of the second intermediate node of plurality of target, therefore base of the embodiment of the present invention
It is specific as follows in the selection scheme that this gives the second intermediate node of specific target：If there are at least two target multicasts
The combination of group includes the complete or collected works of above-mentioned at least two multicasts destination node, and above-mentioned processor 1303 will for the first intermediate node
The multicast destination node and the corresponding content-data of target multicast group that target multicast group includes are sent to the second middle node of target
It puts and includes：For determining target intermediate node according to the result of calculation of the load-balancing mechanism of intermediate node, and by target group
It broadcasts multicast destination node that group includes and the corresponding content-data of target multicast group is sent to the second intermediate node of target.
The present embodiment determines the selection to the second intermediate node of target by way of load balancing, can be preferably sharp
With the physical link bandwidth resources between the replication capacity and switching node at different levels of switching network, handle up to improve system data
Amount.
It is worth noting that, above-mentioned apparatus is only divided according to function logic, but it is not limited to above-mentioned draw
Point, as long as corresponding function can be realized；In addition, the specific name of each functional unit is also only to facilitate mutual area
Point, the protection domain being not intended to restrict the invention.
In addition, one of ordinary skill in the art will appreciate that realizing all or part of step in above-mentioned each method embodiment
It is that relevant hardware can be instructed to complete by program, corresponding program can be stored in a kind of computer readable storage medium
In, storage medium mentioned above can be read-only memory, disk or CD etc..
It these are only the preferable specific implementation mode of the present invention, but scope of protection of the present invention is not limited thereto, it is any
Those familiar with the art the variation that can readily occur in or replaces in the technical scope that the embodiment of the present invention discloses
It changes, should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of claim
Subject to enclosing.
Claims (11)
1. a kind of method of space multicast, which is characterized in that including：
First intermediate node receives native multicasting group, and the native multicasting group contains the content-data for needing multicast and at least
Two multicast destination nodes；
First intermediate node obtains the second set of intermediate nodes and the bandwidth of the second intermediate node；Second middle node
It includes at least two second intermediate nodes that point, which is concentrated, and second intermediate node is to have between the multicast destination node directly
The intermediate node of connection relation, and at least one second intermediate node is only directly connected between partaof destination node
Relationship；
The content-data is replicated the second middle node for being distributed to second intermediate node and concentrating by first intermediate node
Point, and indicate that the second intermediate node for receiving content-data is transmitted to the multicast destination node for having direct connection relational with it；
The content-data is replicated the second middle node for being distributed to second intermediate node and concentrating by first intermediate node
Point, and indicate that the second intermediate node for receiving content-data is transmitted to the multicast destination node packet for having direct connection relational with it
It includes：
The native multicasting group is split as at least two target multicast groups by first intermediate node；At least one target group
Group is broadcast only comprising the partaof destination node in at least two multicasts destination node；
The corresponding content number of multicast destination node and target multicast group that target multicast group is included by first intermediate node
According to being sent to the second intermediate node of target；The second intermediate node of the target is the multicast destination node for including with target multicast group
Between have the intermediate node of direct connection relational.
2. method according to claim 1, which is characterized in that if there are at least two native multicasting groups, it is described by original group
It broadcasts group and is split as at least two target multicast groups and include：
Strategy is split to determine according to the result of calculation of the load-balancing mechanism of intermediate node, and will be each according to the fractionation strategy
Native multicasting group is split as at least two target multicast groups.
3. method according to claim 1, which is characterized in that if including described there are the combination of at least two target multicast groups
The complete or collected works of at least two multicast destination nodes, multicast destination node that target multicast group is included by first intermediate node and
The corresponding content-data of target multicast group is sent to the second intermediate node of target：
First intermediate node determines target intermediate node according to the result of calculation of the load-balancing mechanism of intermediate node, and
Multicast destination node and the corresponding content-data of target multicast group that target multicast group includes are sent among target second
Node.
4. according to claims 1 to 3 any one the method, which is characterized in that first intermediate node is handed over for the second level
Draping node, the second intermediate node are first order switching network node or third level switching network node；The multicast destination node
For line card.
5. method according to claim 1, which is characterized in that if in the presence of the group for being included with more than two target multicast groups
Broadcasting has the second intermediate node of direct connection relational between destination node；Then first intermediate node answers the content-data
System is distributed to the second intermediate node that second intermediate node is concentrated：
Will content-data corresponding at least two target multicast groups after fractionation merge after be sent to it is described with it is more than two
There is the second intermediate node of direct connection relational between the multicast destination node that target multicast group is included.
6. a kind of device of space multicast, uses as the first intermediate node, which is characterized in that including：
Multicast group receiving unit, for receiving native multicasting group, the native multicasting group contains the content-data for needing multicast
And at least two multicast destination node；
Node acquiring unit, for obtaining the second set of intermediate nodes and the bandwidth of the second intermediate node；Second middle node
It includes at least two second intermediate nodes that point, which is concentrated, and second intermediate node is to have between the multicast destination node directly
The intermediate node of connection relation, and at least one second intermediate node is only directly connected between partaof destination node
Relationship；
Dispatching Unit, for the content-data to be replicated to the second intermediate node for being distributed to second intermediate node and concentrating,
And indicate that the second intermediate node for receiving content-data is transmitted to the multicast destination node for having direct connection relational with it；
The Dispatching Unit includes：
Split cells, for the native multicasting group to be split as at least two target multicast groups；At least one target multicast
Group only includes the partaof destination node in at least two multicasts destination node；
Transmission sub-unit, the multicast destination node for including by target multicast group and the corresponding content-data of target multicast group
It is sent to the second intermediate node of target；The second intermediate node of the target be include with target multicast group multicast destination node it
Between have the intermediate node of direct connection relational.
7. device according to claim 6, which is characterized in that if there are at least two native multicasting groups；
The split cells, specifically for splitting strategy according to the result of calculation of the load-balancing mechanism of intermediate node to determine,
And each native multicasting group is split as at least two target multicast groups according to fractionation strategy.
8. device according to claim 6, which is characterized in that if including described there are the combination of at least two target multicast groups
The complete or collected works of at least two multicast destination nodes；
The transmission sub-unit, specifically for being determined among target according to the result of calculation of the load-balancing mechanism of intermediate node
Node, and multicast destination node and the corresponding content-data of target multicast group that target multicast group includes are sent to target
Two intermediate nodes.
9. according to claim 6 to 8 any one described device, which is characterized in that first intermediate node is handed over for the second level
Draping node, second intermediate node are first order switching network node or third level switching network node；The multicast purpose
Node is line card.
10. device according to claim 6, which is characterized in that if being included in the presence of with more than two target multicast groups
There is the second intermediate node of direct connection relational between multicast destination node；
The Dispatching Unit, for being sent to after merging content-data corresponding at least two target multicast groups after fractionation
There is the second middle node of direct connection relational between the multicast destination node for being included with more than two target multicast groups
Point.
11. a kind of system of space multicast, including：First intermediate node, the second intermediate node, root node and leaf node；
Wherein root node is multicast source node, and leaf node is multicast destination node；Second intermediate node be and the multicast mesh
Node between have the intermediate node of direct connection relational；And second intermediate node and leaf node are non-full connection；Its
It is characterized in that, first intermediate node is the device described in claim 6~10 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410710701.4A CN104468144B (en) | 2014-11-28 | 2014-11-28 | A kind of method of space multicast, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410710701.4A CN104468144B (en) | 2014-11-28 | 2014-11-28 | A kind of method of space multicast, apparatus and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104468144A CN104468144A (en) | 2015-03-25 |
| CN104468144B true CN104468144B (en) | 2018-08-14 |
Family
ID=52913573
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410710701.4A Active CN104468144B (en) | 2014-11-28 | 2014-11-28 | A kind of method of space multicast, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104468144B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113645120A (en) * | 2021-10-15 | 2021-11-12 | 中兴通讯股份有限公司 | Message processing method, line card, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1716904A (en) * | 2004-06-30 | 2006-01-04 | 华为技术有限公司 | Group broadcast realizing method based on multiple service transmission platform |
| CN101242342A (en) * | 2007-02-05 | 2008-08-13 | 华为技术有限公司 | Multicast method and multicast route method |
| CN101286925A (en) * | 2007-04-13 | 2008-10-15 | 上海贝尔阿尔卡特股份有限公司 | Method and device for detecting whether repeated multicasting datagram appears |
| CN101425973A (en) * | 2008-11-25 | 2009-05-06 | 中兴通讯股份有限公司 | Method for route selection by IP multicast on equivalent multiple paths |
| WO2011076065A1 (en) * | 2009-12-22 | 2011-06-30 | 华为技术有限公司 | Method and device for processing multicast messages |
| CN103731279A (en) * | 2012-10-11 | 2014-04-16 | 中兴通讯股份有限公司 | Operation executing method and device |
-
2014
- 2014-11-28 CN CN201410710701.4A patent/CN104468144B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1716904A (en) * | 2004-06-30 | 2006-01-04 | 华为技术有限公司 | Group broadcast realizing method based on multiple service transmission platform |
| CN101242342A (en) * | 2007-02-05 | 2008-08-13 | 华为技术有限公司 | Multicast method and multicast route method |
| CN101286925A (en) * | 2007-04-13 | 2008-10-15 | 上海贝尔阿尔卡特股份有限公司 | Method and device for detecting whether repeated multicasting datagram appears |
| CN101425973A (en) * | 2008-11-25 | 2009-05-06 | 中兴通讯股份有限公司 | Method for route selection by IP multicast on equivalent multiple paths |
| WO2011076065A1 (en) * | 2009-12-22 | 2011-06-30 | 华为技术有限公司 | Method and device for processing multicast messages |
| CN103731279A (en) * | 2012-10-11 | 2014-04-16 | 中兴通讯股份有限公司 | Operation executing method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104468144A (en) | 2015-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11677611B2 (en) | Host side method of using a controller assignment list | |
| CN103581274B (en) | Message forwarding method and device in stacking system | |
| US10218523B2 (en) | Using a central controller cluster to configure a distributed multicast logical router | |
| CN103763135B (en) | PE equipment traffic scheduling method and device | |
| US20190020490A1 (en) | Distributed multicast logical router | |
| CN104144135B (en) | Resource allocation methods for multicast virtual network and anti-ruin resource allocation methods | |
| CN103125102A (en) | System and method for providing ethernet over infiniband virtual hub scalability in a middleware machine environment | |
| CN102986172A (en) | Virtual cluster switching | |
| CN105721354B (en) | Network-on-chip interconnected method and device | |
| DE112013001426B4 (en) | Dynamic optimization of a multicast tree hierarchy for a distributed switch | |
| CN104836736A (en) | Spectrum sensing-based energy efficiency resource allocation and path reconfiguration method in elastic optical network | |
| CN104243323B (en) | Switching network multi-broadcast routing method and system | |
| CN104869081B (en) | MESSAGE EXCHANGE processing method, business board and internet exchange system | |
| CN103560967A (en) | Virtual data center mapping method sensing service requirement | |
| CN105357124B (en) | A kind of MapReduce bandwidth optimization methods | |
| CN105991445A (en) | Link aggregation group setting method and device | |
| CN103457852A (en) | Invulnerability mapping method of multicast virtual network | |
| CN112887134B (en) | Network service grid arranging method, device, equipment and medium | |
| CN110493137A (en) | A kind of Tiered broadcast method, apparatus and system based on fixed route network | |
| CN101330411B (en) | Method and system for simulating large-scale network topological | |
| CN102857446A (en) | Cache management method and cache management apparatus for Ethernet switching chip | |
| CN105530179A (en) | Data message processing method and network equipment | |
| CN110958133B (en) | Network slice mapping method, device, server and storage medium | |
| CN110535673B (en) | Virtual subnet construction method, equipment and storage medium | |
| CN104468144B (en) | A kind of method of space multicast, apparatus and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211228 Address after: 450046 Floor 9, building 1, Zhengshang Boya Plaza, Longzihu wisdom Island, Zhengdong New Area, Zhengzhou City, Henan Province Patentee after: Super fusion Digital Technology Co.,Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd. Effective date of registration: 20211228 Address after: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee after: HUAWEI TECHNOLOGIES Co.,Ltd. Address before: 100085, No. 3, information road, Haidian District, Beijing Patentee before: Beijing Huawei Digital Technology Co.,Ltd. |