CN100561970C - A kind of method and apparatus of realizing load balancing - Google Patents
A kind of method and apparatus of realizing load balancing Download PDFInfo
- Publication number
- CN100561970C CN100561970C CNB2007100996741A CN200710099674A CN100561970C CN 100561970 C CN100561970 C CN 100561970C CN B2007100996741 A CNB2007100996741 A CN B2007100996741A CN 200710099674 A CN200710099674 A CN 200710099674A CN 100561970 C CN100561970 C CN 100561970C
- Authority
- CN
- China
- Prior art keywords
- port
- vlan
- topology
- place
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
The embodiment of the invention discloses a kind of method and apparatus of realizing load balancing.Method comprises: the different VLAN at each port place are mapped to plural protocol entity, wherein said plural protocol entity comprises at least one Multiple Spanning Tree Protocol MSTP example and at least one quick ring protection agreement RRPP domain or flexible link group; And on each protocol entity after the mapping, move two corresponding etale topology agreements respectively, calculate the state of each port corresponding to each VLAN at its place; Last according to the state of each port corresponding to each VLAN at its place, the vlan data that each port receives is handled.Technical scheme of the present invention, the two etale topology agreements that feasible not holding load is shared can be transmitted along different paths the data of different VLAN, thereby have realized the load balancing by VLAN, and then have improved the utilance of link in the double layer network.
Description
Technical field
The present invention relates to network communications technology field, refer to a kind of method and apparatus of realizing load balancing especially.
Background technology
Two etale topology agreements comprise Multiple Spanning Tree Protocol (MSTP, Multiple Spanning TreeProtocol), quick ring protection agreement (RRPP, Rapid Ring Protection Protocol), flexible link (Smart Link) agreement etc.Two etale topology agreements can prevent the broadcast storm that data loopback causes when Ethernet ring is complete, and link on Ethernet ring can be enabled backup link rapidly to recover the communication path between each node on the looped network when disconnecting.
The load balancing of two etale topology agreements on data flow and forwarding control all are to realize by the register of MSTP.Usually the hardware register relevant with MSTP comprises two parts, and one is VLAN (VLAN, Virtual Local Area Network) allocation table, and another is a Port State Table.The VLAN allocation table is at VLAN and generate between the tree group (STG, Spanning Tree Group) and set up many-to-one mapping relations, i.e. STG that VLAN is corresponding unique, and a STG can corresponding a plurality of VLAN.Table 1 is the schematic diagram of VLAN Allocation Table structures.The port status allocation table provides a state for the combination of each port and STG, as " abandoning (Discarding) " state and " transmitting (Forwarding) " state.Table 2 is Port State Table structural representations.
| The VLAN index | 0 | 1 | 2 | ... | 4092 | 4093 | 4094 | 4095 |
| Affiliated STG number of VLAN | 0 | 5 | 1 | ... | 1 | 0 | 3 | 0 |
Table 1
Table 2
When the inbound port of a switching equipment in the network such as port one are received the message that carries " VLAN 2 ", from the VLAN allocation table, obtain earlier the STG corresponding with VLAN 2, for example corresponding with VLAN 2 in the table 1 is STG1; In Port State Table, search the state corresponding then with port one and STG1, if " Discarding " state then abandons this message, if then carrying out normal Layer 2 data, transmits " Forwarding " state, for example corresponding with port one and STG 1 in table 2 is " Discarding " state, so this switching equipment abandons the received message of port one.Message also can the inspection state corresponding with outbound port and STG on the outbound port of switching equipment, if " Discarding " state then abandons this message, if " Forwarding " state then normally sends.No matter move which kind of two etale topology agreement, the chip data in the switching equipment is all carried out above-mentioned data forwarding flow process.
At present, in two etale topology agreements, be that holding load is shared based on the MSTP of protocol entity, the protocol entity that defines in MSTP is called example (Instance).The public Internal Spanning Tree (CIST) of MSTP and Multiple Spanning Tree Instance (MSTI) are referred to as the example of MSTP, the corresponding one group of VLAN of each example, realize by an example is mapped on the hardware STG, be about to a plurality of VLAN and be mapped on the STG, make STG and example corresponding one by one again.MSTP realizes that by the different conditions that calculates port on different examples data traffic realizes load balancing by different VLAN.
But RRPP does not support data traffic to press the load balancing of different VLAN with Smart Link agreement in the two etale topology agreements.This be because, the protocol entity Smart Link group (Smart LinkGroup) that defines in the protocol entity RRPP domain (RRPPDomain) of the definition among the RRPP and the Smart Link agreement all is defined on one group of port, and each port generally only belongs to a protocol entity, promptly a port only belongs to a Smart Link Group or a RRPP Domain, therefore after operation agreement on the protocol entity, the port that defines for protocol entity can only calculate a kind of fixing data forward state.Like this, for the data of all VLAN, this port can carry out unified processing, promptly all transmits or all abandons, thereby can't realize the load balancing of different vlan datas.For example, Fig. 1 is that of the prior art setting up inserted networking diagram fully, be equipped with operation Smart Link agreement in two up access networkings in shown in Figure 1 setting up, port one on the equipment C and port 2 constitute a Smart Link Group, operation Smart Link agreement rear port 1 is a master port on this Smart Link Group, be in forwarding state, and port 2 is from port, be in the state of getting clogged, therefore, under the normal condition, the data traffic of all VLAN all can only be passed through link 2, and can not pass through link 3, thereby can not effectively utilize the bandwidth of link 3.
Summary of the invention
In view of this, the main purpose of the embodiment of the invention is to provide a kind of method that realizes load balancing, and the two etale topology agreements that the feasible not holding load of this method is shared can realize load balancing, thereby have improved the utilance of link in the double layer network.
The embodiment of the invention also provides a kind of device of realizing load balancing, and the two etale topology agreements that the feasible not holding load of this device is shared can realize load balancing, thereby have improved the utilance of link in the double layer network.
For achieving the above object, technical scheme of the present invention specifically is achieved in that
The embodiment of the invention discloses a kind of method that realizes load balancing, this method comprises:
The different VLAN at each port place are mapped to plural protocol entity; Wherein said plural protocol entity comprises at least one Multiple Spanning Tree Protocol MSTP example and at least one quick ring protection agreement RRPP domain or flexible link group;
Two etale topology agreements of operation correspondence respectively on each protocol entity after the mapping, calculate the state of each port corresponding to each VLAN at its place, according to the state of each port, the vlan data that each port receives is handled corresponding to each VLAN at its place;
Wherein, move two etale topology agreements respectively on described each protocol entity after mapping, calculating each port comprises corresponding to the state of each VLAN at its place: by disposing the initiation protocol parameter of each protocol entity after the described mapping, make on each protocol entity each the loop choke point that calculates after the corresponding two etale topology agreements of operation respectively not on same port.
The embodiment of the invention also discloses a kind of device of realizing load balancing, this device comprises: mapping block, state computation module and data forwarding control module, wherein,
Mapping block is used for the different VLAN at each port place of equipment are mapped to plural protocol entity; Wherein said plural protocol entity comprises at least one Multiple Spanning Tree Protocol MSTP example and at least one quick ring protection agreement RRPP domain or flexible link group;
The state computation module is used for moving two corresponding etale topology agreements respectively on each protocol entity after the mapping, calculates the state of each port of equipment corresponding to each VLAN at its place respectively; Wherein, move two etale topology agreements respectively on described each protocol entity after mapping, calculating each port comprises corresponding to the state of each VLAN at its place: by disposing the initiation protocol parameter of each protocol entity after the described mapping, make on each protocol entity each the loop choke point that calculates after the corresponding two etale topology agreements of operation respectively not on same port;
The data forwarding control module is used for according to the state of each port corresponding to each VLAN at its place the vlan data that each port receives being handled.
As seen from the above technical solution, the embodiment of the invention, by determining the different VLAN at each port place, the different VLAN at each port place are mapped to plural protocol entity, and on each protocol entity after the mapping, move two etale topology agreements respectively, calculate the state of each port respectively corresponding to each VLAN at its place; Last according to the state of each port corresponding to each VLAN at its place, the technical scheme that the vlan data that receives is handled, make and on same port, can move a plurality of protocol entities, thereby same port has different forwarding states to the vlan data that belongs to the different agreement entity, the vlan data that promptly belongs to the different agreement entity transmits along different paths, therefore, realized that two etale topology agreements that holding load is not shared support the load balancing by VLAN, and then improved the utilance of link in the double layer network.
Description of drawings
Fig. 1 is that of the prior art setting up inserted networking diagram fully;
Fig. 2 is a kind of flow chart of realizing the method for load balancing of the embodiment of the invention;
Fig. 3 is the flow chart of the method shared of a kind of RRPP of realization holding load of the embodiment of the invention;
Fig. 4 is that the present invention realizes the embodiment schematic diagram that the RRPP holding load is shared;
Fig. 5 is the flow chart of the method shared of a kind of Smart of realization Link of embodiment of the invention agreement holding load;
Fig. 6 is that the present invention realizes the embodiment schematic diagram that Smart Link agreement and MSTP holding load are shared simultaneously;
Fig. 7 is a kind of structured flowchart of realizing the device of load balancing of the embodiment of the invention.
Embodiment
The embodiment of the invention mainly is mapped to plural protocol entity by the different VLAN with each port place, to realize a multi-protocols entity on the port, make the different agreement entity on same port, calculate different data forward state, and then realize the load balancing of two etale topology agreements different vlan datas.
Fig. 2 is a kind of flow chart of realizing the method for load balancing of the embodiment of the invention, as shown in Figure 2, may further comprise the steps:
For compatible existing ATM hardware device, preferably, the realization of this step can be the notion of setting up the topology group, all VLAN with each port place are mapped to different topology groups respectively earlier, then according to the VLAN at port place, determine the different topology group of each port correspondence, the different topology group of each port correspondence is mapped to plural protocol entity, and on each protocol entity after the mapping, move two etale topology agreements respectively, calculate the state of each port respectively corresponding to each topology group at its place, and, determine the state of each port corresponding to each VLAN at its place according to the mapping relations between VLAN and the topology group.Can obtain, all be many-to-one mapping relations between VLAN and topology group, topology group and the protocol entity, and VLAN is mapped to protocol entity by the topology group, and the topology group is the set of one group of VLAN.
Above-mentioned when all VLAN are mapped to different topology groups respectively, one group of VLAN that can logical topology among all VLAN is identical is mapped to same topology group.One group of VLAN that logical topology is identical is one group of VLAN that its data need be transmitted along same paths.
Above-mentioned when setting up mapping relations between VLAN and the protocol entity indirectly by the topology group, the VLAN that more than one logical topology is identical can be corresponding with a topology group, but some VLAN can only be corresponding with a unique topology group, and can not be corresponding with a plurality of topology groups; Equally on a port, under this port more than one topological group can with a protocol entity correspondence, but some topological group can only with a unique protocol entity correspondence.As for which protocol entity is which topology group be mapped to, do not do qualification here.
In this step, when two etale topology agreements were RRPP, protocol entity referred to RRPP domain (RRPPDomain); When two etale topology agreements were Smart Link agreement, protocol entity referred to flexible link group (Smart Link Group); When two etale topology agreements were MSTP, protocol entity referred to MSTP example (MSTP Instance).
Preferably, in this step, described calculate each port corresponding to the implementation procedure of the state of each VLAN at its place can for: by disposing the initiation protocol parameter of each protocol entity after the described mapping, make and on each protocol entity, move each loop choke point of calculating after the two etale topology agreements respectively not on same port, and, determine the state of each port corresponding to each VLAN at its place according to the mapping relations between each the loop choke point that is calculated and VLAN and topology group, topology group and the protocol entity.
In the embodiment shown in Figure 2, if set up relation between VLAN and the protocol entity indirectly by topology in the step 201, then further can also with many-to-one mapping relations information stores between the VLAN that sets up in the step 201 and the topology group in the VLAN allocation table, each port that will calculate and its place each topologically organize corresponding state information and store in the Port State Table; In the step 202,, the vlan data that each port receives is transmitted control by the information in inquiry VLAN allocation table and the Port State Table.
For making purpose of the present invention, technical scheme and advantage clearer, be example with RRPP and Smart Link agreement below, the present invention is described in more detail.
For the RRPP holding load is shared, the embodiment of the invention defines the notion of topology group, and the more than one VLAN that logical topology is identical is mapped to a topology group, and the different topology group with a port place is mapped to plural RRPP Domain again.At present, RRPP Domain is defined as follows among the RRPP: RRPP Domain is identified by the identifier (ID) of integer representation, a configuration set between the switch of identical Domain ID and control VLAN interconnected port constitute a RRPP Domain.After introducing the notion of topology group in embodiments of the present invention, make a RRPP Domain operate on the part topology group of the appointment on its one group of affiliated port, rather than all topologys on one group of port under it are organized.Therefore, in the embodiment of the invention, the definition of RRPP Domain after the expansion can for: RRPPDomain is identified by the ID of integer representation, a configuration set between the switch of identical Domain ID and control VLAN interconnected port on the more than one topology group of appointment, constitute a RRPPDomain.
Fig. 3 is the flow chart of the method shared of a kind of RRPP of realization holding load of the embodiment of the invention.Fig. 4 is that the present invention realizes the embodiment schematic diagram that the RRPP holding load is shared, in Fig. 4, topological structure between switching equipment 401 and the switching equipment 402 is: between the port one 1 of switching equipment 401 and the port 21 of switching equipment 402 link is arranged, between the port one 2 of switching equipment 401 and the port 22 of switching equipment 402 link is arranged, so the port 21 of the port one 1 of equipment 401 and port one 2 and equipment 402 and port 22 have formed a loop.Be elaborated below in conjunction with Fig. 3 and Fig. 4.
As shown in Figure 3, the method shared of a kind of RRPP of realization holding load may further comprise the steps:
For example, in the embodiment shown in fig. 4, set up 3 topology groups on each switching equipment, numbering is respectively 1,2 and 3, and topology group 1 usefulness solid line represents that topology group 2 dots, and topology group 3 usefulness chain-dotted lines are represented.And VLAN 1, VLAN 2 and VLAN 3 be mapped to respectively on 3 topology groups, form VLAN allocation table as shown in table 3.
| The VLAN index | 1 | 2 | 3 |
| Topological group # under the VLAN | 1 | 2 | 3 |
Table 3
Being man-to-man mapping relations between VLAN and the topology group in the table 3, is a special case of many-one mapping relations between VLAN and the topology group.
For example, in the embodiment shown in fig. 4, the port 21 of the port one 1 of equipment 401 and port one 2 and equipment 402 and port 22 have formed a loop, the topology group 1 at each port place on this loop is mapped to RRPP Domain 1, topology group 2 and topology group 3 are mapped to RRPP Domain 2, be that RRPPDomain 1 operates on the topology group 1 at port one 1, port one 2, port 21 and port 22 places, RRPPDomain 2 operates on the topology group 2 and topology group 3 at port one 1, port one 2, port 21 and port 22 places.The RRPP parameter of pre-configured RRPP Domain 1 and RRPP Domain, and on RRPPDomain 1 and RRPP Domain 2, move RRPP according to pre-configured RRPP parameter, make the choke point of RRPP Domain 1 intermediate ring road at port 22, and the choke point of RRPP Domain 2 intermediate ring roads is represented with the black round dot in Fig. 4 at port 21.According to the position of above-mentioned choke point and the mapping relations between topology group and the RRPP Domain, calculate each port state corresponding then, form Port State Table as shown in table 4 with the topology group at its place.Because calculating the process of Port State Table at operation RRPP is prior art, and the computational process complexity, no longer describes here, directly provides the result of calculation of Port State Table.
Table 4
For example, in the embodiment shown in fig. 4, when carrying the message of " VLAN 1 " when port 22 receives, from the VLAN allocation table shown in the table 3, find out earlier the topology group 1 corresponding with VLAN 1, in the Port State Table shown in the table 4, search then with port 22 and topology and organize 1 corresponding state, its result is " Discarding " state, abandons this message.
When carrying the message of " VLAN 2 " when port 22 receives, from the VLAN allocation table shown in the table 3, find out earlier the topology group 2 corresponding with VLAN 2, in the Port State Table shown in the table 4, search then with port 22 and topology and organize 2 corresponding states, its result is " Forwarding " state, normally transmits this message.Same method when carrying the message of " VLAN 3 " when port 22 receives, is normally transmitted this message.
All normally transmit received message when by that analogy, port one 1 and port one 2 carry the message of " VLAN 1 " or " VLAN 2 " or " VLAN 3 " when receiving; And the port 21 normal messages that carry " VLAN 1 " that received of transmitting abandon the message that carries " VLAN 2 " or " VLAN 3 " that is received.As seen the data flow of VLAN 1 is passed through from the link between port one 1 and the port 21, and the data flow of VLAN 2 and VLAN 3 is passed through from the link between port one 2 and the port 22, and promptly the link between link between port one 1 and the port 21 and port one 2 and the port 22 has been realized load balancing to the data flow of VLAN 1, VLAN 2 and VLAN 3.
Certainly, in the embodiment shown in fig. 4, when moving RRPP on RRPP Domain 1 and RRPP Domain 2, each port state corresponding with the topology group that calculates according to different pre-configured RRPP parameter operation RRPP also can be not quite similar.For example, because the port one 1 of equipment 401 and the port 21 and the port 22 of port one 2 and equipment 402 have formed a loop, therefore, can make the choke point of RRPP Domain 1 intermediate ring road at port one 2 by adjusting pre-configured RRPP parameter, and the choke point of RRPPDomain 2 intermediate ring roads is at port one 1 etc., and this does not influence the solution of the present invention and realizes load balancing.RRPP parameter described here is meant the initiation protocol parameter that need dispose when moving RRPP on RRPP Domain.
Share in order to make Smart Link agreement holding load, the embodiment of the invention has defined topological group notion equally, the more than one VLAN that logical topology is identical is mapped to a topology group, and the different topology group with a port place is mapped to plural Smart Link Group again.At present, Smart Link Group is defined as follows in the SmartLink agreement: Smart Link Group is identified by the ID of integer representation, and two ports on equipment constitute a Smart Link Group.After introducing the notion of topology group in embodiments of the present invention, make a Smart Link Group operate on the part topology group of the appointment on two ports of an equipment under it, rather than on all the topology groups on two ports under it.Therefore, in the embodiment of the invention, the definition of Smart Link Group after the expansion can for: Smart Link Group is identified by the ID of integer representation, and two ports on equipment constitute a Smart Link Group on the more than one topology group of appointment.
Fig. 5 is the flow chart of the method shared of a kind of Smart of realization Link of embodiment of the invention agreement holding load, as shown in Figure 5, may further comprise the steps:
Step 503 according to VLAN allocation table and Port State Table, is transmitted control to the vlan data that each port receives.
Still being equipped with two up access networking diagrams with shown in Figure 1 setting up below is example, illustrates how the present invention realizes what Smart Link agreement holding load was shared.
Set up 3 topology groups on each switching equipment in Fig. 1, be respectively topology group 1, topology group 2 and topology group 3, and VLAN 1, VLAN 2 and VLAN 3 are mapped to respectively on 3 topology groups.Make port one and port 2 on the equipment C on topology group 1, constitute Smart Link Group 1, port one on the equipment C and port 2 constitute Smart Link Group 2 on topology group 2 and topology group 3, the port one and the topology group 1 on the port 2 that are about on the equipment C are mapped to Smart Link Group 1, and topology group on port one on the equipment C and the port 22 and topology group 3 are mapped to Smart Link Group 2.Initiation protocol parameter by configuration Smart LinkGroup 1 and Smart Link Group 2 is carried out, and respectively after operation Smart Link agreement on Smart LinkGroup 1 and the Smart Link Group 2, make that Smart LinkGroup 1 middle port 1 is a master port, port 2 is from port, and Smart Link Group 2 middle ports 1 are from port, and port 2 is master ports.Therefore, the port one of equipment C is transmitted the data flow of VLAN 1, blocks the data flow of VLAN2 and VLAN 3; The port 2 of equipment C is transmitted the data flow of VLAN2 and VLAN 3, and the data flow of blocking VLAN 1 has realized load balancing.
The front was mentioned, and no matter moved which kind of two etale topology agreement at present, all had two groups of registers to be respectively applied for storage VLAN allocation table as shown in table 1 and the Port State Table shown in the table 2 on the chip in the switching equipment, and carried out the forwarding of data flow according to table 1 and table 2.Therefore the topology group in the embodiment of the invention is corresponding with the STG in table 1 and the table 2, promptly can under the condition that does not change existing any hardware device, use the load balancing that technical scheme of the present invention realizes RRPP or Smart Link agreement.
Technical scheme of the present invention is equally applicable to MSTP.The notion of at first definition topology group, the more than one VLAN that logical topology is identical is mapped to a topology group, sets up many-to-one mapping relations between VLAN and the topology group, obtains the VLAN allocation table; Set up many-to-one mapping relations between topology group and the MSTP Instance again, and on each MSTP Instance, move the MSTP agreement, calculate and each port and the corresponding state of topology group, obtain Port State Table; According to VLAN allocation table and Port State Table data stream is transmitted control then.At this moment, MSTP Instance no longer sets up direct mapping relations with VLAN, but founds mapping relations by setting up with topology, thereby shines upon VLAN indirectly.More than one topology group can be mapped to some MSTP Instance, some topological group can only be arrived a MSTP Instance, can not be mapped to a plurality of MSTP Instance.
Fig. 6 is that the present invention realizes the embodiment schematic diagram that Smart Link agreement and MSTP holding load are shared simultaneously.As shown in Figure 6, a link is arranged between the port 3 of the port one of switching equipment 601 and switching equipment 602, between the port 2 of switching equipment 601 and the port 4 of switching equipment 602 link is arranged, these two links form a loop; Between the port 5 of switching equipment 602 and the port 7 of switching equipment 603 link is arranged, between the port 6 of switching equipment 602 and the port 8 of switching equipment 603 link is arranged, and these two links form a loop.Operation Smart link agreement is moved MSTP between equipment 602 and equipment 603 between switching equipment 601 and switching equipment 602.Set up 3 topology groups on each switching equipment, numbering is respectively 1,2 and 3, and topology group 1 usefulness solid line represents that topology group 2 dots, and topology group 3 usefulness chain-dotted lines are represented.And VLAN 1, VLAN 2 and VLAN 3 be mapped to respectively on 3 topology groups, the VLAN allocation table of acquisition is with shown in the table 3.Smart link Group 1 is operated on the topology group 1, and Smart link Group 2 operates on topology group 2 and the topology group 3; MSTP Instance 1, MSTPInstance 2 and MSTP Instance 3 are operated in respectively on topology group 1, topology group 2 and the topology group 3.Adjustment by Smart Link protocol parameter, can allow the choke point of Smart Link Group 1 intermediate ring road on port 4, and the choke point of Smart Link Group 2 intermediate ring roads is on port 3, and like this, Smart Link agreement has just realized the load balancing by VLAN.Equally, by the adjustment of MSTP parameter, can allow the choke point of MSTP Instance 1 on port 7, and load balancing has also been realized in the choke point of MSTP Instance 2 and MSTP Instance 3 on port 8.In Fig. 6, represent the choke point with the black round dot.
In Fig. 6, also can be between equipment 601 and equipment 602, and move Smart Link agreement and MSTP simultaneously between equipment 602 and the equipment 603.For example VLAN mapping table is still as shown in table 3, and between equipment 601 and 602, MSTP Instance 1 and MSTP Instance 2 operate in respectively on topology group 1 and the topology group 2, on the topology group 3 of Smart Link Group 1 operation; Between equipment 602 and equipment 603, Smart Link Group 2 operates on the topology group 1, on topology group 2 that MSTP Instance 2 and MSTPInstance 3 move respectively and the topology group 3.
As seen, can move the different agreement entity of different two etale topology agreements on the same port, promptly organize in the many-one mapping of protocol entity in topology, the part topology group at a port place can be mapped to the protocol entity of one two etale topology agreement, as Smart Link Group, and another part topology at this port place is organized the protocol entity that is mapped to another two etale topologies agreement, as MSTP Instance.Certainly, the topological group of protocol entity that can shine upon is not limited to above-mentioned Smart Link Group and MSTPInstance, and a part of topological group of protocol entity that can also be mapped to other is as RRPP Domain.
Fig. 7 is a kind of structured flowchart of realizing the device of load balancing of the embodiment of the invention.As shown in Figure 7, this device must comprise: mapping block 701, state computation module 703 and data forwarding control module 705.Wherein, mapping block 701 is used for the different VLAN at each port place of equipment are mapped to plural protocol entity; State computation module 703 is used for moving two etale topology agreements respectively on each protocol entity after the mapping, calculates the state of each port of equipment corresponding to each VLAN at its place respectively; Data forwarding control module 705 is used for according to the state of each port corresponding to each VLAN at its place the vlan data that each port receives being handled.
When the existing hardware device of needs compatibility, mapping block 701, be used for all VLAN at each port place of equipment are mapped to different topology groups respectively, and according to the VLAN of port correspondence, determine the different topology group at each port place, and the different topology group at each port place is mapped to plural protocol entity.State computation module 703 is used for calculating respectively the state of each port corresponding to each topology group at its place, and according to the mapping relations between VLAN and the topology group, determines the state of each port corresponding to each VLAN at its place.Data forwarding control module 705 is used for according to the state of each port corresponding to each VLAN at its place the vlan data that each port receives being handled.
Preferably, state computation module 703, be used to dispose the initiation protocol parameter of each protocol entity after the described mapping, make and on each protocol entity, move each loop choke point of calculating after the two etale topology agreements respectively not on same port, and, determine the state of each port corresponding to each VLAN at its place according to the mapping relations between each the loop choke point that is calculated and VLAN and topology group, topology group and the protocol entity.
Device as shown in Figure 7 can further include: VLAN allocation table register 702 and port status table register 704.Wherein:
VLAN allocation table register 702 is used for VLAN and the mapping relations information between the topology group, i.e. VLAN allocation table that the Storage Mapping module sends.
Port status table register 704 is used for each port of store status computing module 703 transmissions and the corresponding state information of each topology group at its place, i.e. Port State Table.
Described data forwarding control module 705, be used for respectively organizing corresponding state information with each topology at its place, and the vlan data that each port receives handled according to the described information of obtaining from mapping relations information and each port that VLAN allocation table register 702 and port status table register 704 obtain between VLAN and the topology group.
Certainly in the embodiment shown in fig. 7, can be directly that many-to-one mapping relations information between VLAN and the topology group and each port and different topology group is the corresponding state information of mapping block 701 and state computation module 703 sends to data forwarding control module 705.
In the embodiment shown in fig. 7, two etale topology agreements specifically can be RRPP, Smart Link agreement or MSTP, and the corresponding protocol entity is respectively RRPP Domain, Smart Link Group or MSTPInstance.
In sum, technical scheme of the present invention is by setting up the mapping corresponding relation between VLAN and the topology group, and the correspondence mappings relation between topology group and the protocol entity, move a plurality of protocol entities at each port, make each port under the different agreement entity, have different data forward state; When carrying out data forwarding, each port carries out data forwarding according to the forwarding state corresponding with the protocol entity at vlan data place, has realized that the vlan data that belongs to the different agreement entity transmits on different ports and link.And technical scheme of the present invention can be applied to any two etale topology agreements based on protocol entity, make this two etale topologies agreement to share by holding load, and then the link flow of raising double layer network, and with the topology group when corresponding with STG of the prior art, need not to change existing any hardware device, make user network under the situation that does not increase investment, can support bigger number of users.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention, all any modifications of being made within the spirit and principles in the present invention, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1, a kind of method that realizes load balancing is characterized in that, this method comprises:
The different virtual LAN VLAN at each port place is mapped to plural protocol entity; Wherein said plural protocol entity comprises at least one Multiple Spanning Tree Protocol MSTP example and at least one quick ring protection agreement RRPP domain or flexible link group;
Two etale topology agreements of operation correspondence respectively on each protocol entity after the mapping, calculate the state of each port corresponding to each VLAN at its place, according to the state of each port, the vlan data that each port receives is handled corresponding to each VLAN at its place;
Wherein, move two etale topology agreements respectively on described each protocol entity after mapping, calculating each port comprises corresponding to the state of each VLAN at its place: by disposing the initiation protocol parameter of each protocol entity after the described mapping, make on each protocol entity each the loop choke point that calculates after the corresponding two etale topology agreements of operation respectively not on same port.
2, the method for claim 1 is characterized in that,
Described different VLAN with each port place are mapped to plural protocol entity and comprise: all VLAN at each port place are mapped to different topology groups respectively, then according to the VLAN at port place, determine the different topology group of each port correspondence, the different topology group of each port correspondence is mapped to plural protocol entity;
Described each port that calculates further comprises corresponding to the state of each VLAN at its place: calculate each port state corresponding to each topology group at its place respectively, and, determine the state of each port corresponding to each VLAN at its place according to the mapping relations between VLAN and the topology group.
3, method as claimed in claim 2, it is characterized in that, this method further comprises after all VLAN with each port place are mapped to different topology groups respectively: with the mapping relations information stores between VLAN on each port and the topology group in the VLAN allocation table of this port;
After calculating the state of each port corresponding to each topology group at its place, further comprise: each port that will calculate stores in the Port State Table of this port corresponding to the state information of each topology group at its place;
According to the information in VLAN allocation table and the Port State Table, carry out the step that the described vlan data that each port is received is handled.
4, method as claimed in claim 3 is characterized in that, the step that the described vlan data that each port is received is handled comprises:
The corresponding topology group of VLAN under the vlan data that finds and receive according to the VLAN allocation table, and find and port self and the corresponding state information of topology group that found according to Port State Table, according to the state information that finds the vlan data that receives is transmitted or abandoned.
5, method as claimed in claim 2 is characterized in that, the described step that all VLAN are mapped to different topology groups respectively comprises: the VLAN that logical topology is identical among all VLAN is mapped to same topology group.
6, a kind of device of realizing load balancing is characterized in that, this device comprises: mapping block, state computation module and data forwarding control module, wherein,
Mapping block is used for the different VLAN at each port place of equipment are mapped to plural protocol entity; Wherein said plural protocol entity comprises at least one Multiple Spanning Tree Protocol MSTP example and at least one quick ring protection agreement RRPP domain or flexible link group;
The state computation module is used for moving two corresponding etale topology agreements respectively on each protocol entity after the mapping, calculates the state of each port of equipment corresponding to each VLAN at its place respectively; Wherein, move two etale topology agreements respectively on described each protocol entity after mapping, calculating each port comprises corresponding to the state of each VLAN at its place: by disposing the initiation protocol parameter of each protocol entity after the described mapping, make on each protocol entity each the loop choke point that calculates after the corresponding two etale topology agreements of operation respectively not on same port;
The data forwarding control module is used for according to the state of each port corresponding to each VLAN at its place the vlan data that each port receives being handled.
7, device as claimed in claim 6 is characterized in that,
Described mapping block, be further used for all VLAN at each port place of equipment are mapped to different topology groups respectively, and according to the VLAN of port correspondence, determine the different topology group at each port place, and the different topology group at each port place is mapped to plural protocol entity;
Described state computation module is further used for calculating respectively the state of each port corresponding to each topology group at its place, and according to the mapping relations between VLAN and the topology group, determines the state of each port corresponding to each VLAN at its place.
8, device as claimed in claim 7 is characterized in that, this device further comprises: VLAN allocation table register and port status table register, wherein,
VLAN allocation table register is used to store the mapping relations information between VLAN and the topology group;
The port status table register is used to store each port and organizes corresponding state information with each topology at its place;
Described mapping block is further used for the mapping relations information between VLAN and the topology group is sent to VLAN allocation table register;
Described state computation module, each port state information corresponding with each topology group at its place that is further used for being calculated sends to the port status table register;
Described data forwarding control module, be further used for respectively organizing corresponding state information with each topology at its place from mapping relations information and each port that VLAN allocation table register and port status table register obtain between VLAN and the topology group, and according to the described information of obtaining, realize describedly, the vlan data that each port receives is handled according to the state of each port corresponding to each VLAN at its place.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007100996741A CN100561970C (en) | 2007-05-28 | 2007-05-28 | A kind of method and apparatus of realizing load balancing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007100996741A CN100561970C (en) | 2007-05-28 | 2007-05-28 | A kind of method and apparatus of realizing load balancing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101056246A CN101056246A (en) | 2007-10-17 |
| CN100561970C true CN100561970C (en) | 2009-11-18 |
Family
ID=38795862
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2007100996741A Expired - Fee Related CN100561970C (en) | 2007-05-28 | 2007-05-28 | A kind of method and apparatus of realizing load balancing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100561970C (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101355519B (en) * | 2008-09-12 | 2010-11-10 | 杭州华三通信技术有限公司 | Method and apparatus for implementing share of business stream load |
| CN102075410A (en) * | 2009-11-20 | 2011-05-25 | 杭州华三通信技术有限公司 | Port mirroring method and device in stacking equipment |
| CN102055676B (en) * | 2011-01-25 | 2013-11-27 | 杭州华三通信技术有限公司 | Method and device for realizing load sharing in multiple spanning tree protocol (MSTP) networking environment |
| AU2011229566B2 (en) * | 2011-04-27 | 2015-12-10 | Huawei Technologies Co., Ltd. | Load sharing method and apparatus |
| CN102209040B (en) * | 2011-07-12 | 2013-12-25 | 杭州华三通信技术有限公司 | Multi-network-port-based load sharing method and device |
| CN103534985B (en) * | 2011-07-26 | 2017-02-01 | 华为技术有限公司 | Service load allocating method, apparatus and communication system |
| CN102326370B (en) * | 2011-08-05 | 2014-04-30 | 华为技术有限公司 | Message processing method, apparatus and system |
| CN102497324B (en) * | 2011-12-22 | 2015-09-23 | 杭州华三通信技术有限公司 | Data transmission method between a kind of different Multiple Spanning Tree (MST) region and equipment |
| CN103501274B (en) * | 2013-09-24 | 2017-09-12 | 神州数码网络(北京)有限公司 | Flow load sharing method and device in a kind of multilink transparent internet |
| CN105337873B (en) * | 2014-05-30 | 2019-06-07 | 杭州迪普科技股份有限公司 | MSTP networking method for distributing business and device |
| CN104683204B (en) * | 2015-03-17 | 2017-09-29 | 深圳市振兴光通信股份有限公司 | The control method and device of a kind of FastRing looped networks |
| CN105939272B (en) * | 2015-11-25 | 2019-09-06 | 杭州迪普科技股份有限公司 | A kind of calculation method and device of MSTP |
-
2007
- 2007-05-28 CN CNB2007100996741A patent/CN100561970C/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| HP ProLiant BL p-Class GbE2 Interconnect Switch Application Guide. 52-53,Hewlett-Packard Development Company, L.P.. 2003 |
| HP ProLiant BL p-Class GbE2 Interconnect Switch Application Guide. 52-53,Hewlett-Packard Development Company, L.P.. 2003 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101056246A (en) | 2007-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100561970C (en) | A kind of method and apparatus of realizing load balancing | |
| KR101341272B1 (en) | Providing a logical aggregated point-to-point data link incorporating a multi-point link | |
| CN102415065B (en) | The method and apparatus that redundant host connects in the network having route | |
| EP1858212B1 (en) | A MAC Frame Transfer Method, an edge bridge, and a system for reducing flooding | |
| US7286491B1 (en) | Virtual local area network membership registration protocol for multiple spanning tree network environments | |
| US7171504B2 (en) | Transmission unit | |
| CN100586079C (en) | A kind of method for discovering network topology of data-oriented link layer | |
| JP4676528B2 (en) | RPR ring network on / off ring processing, data transfer method and network equipment thereof | |
| JP4777990B2 (en) | Determine and configure paths in the network | |
| JP2010509880A (en) | Hash-based multihoming | |
| CN101960798A (en) | Distributed spanning tree protocol on a multi chassis port channel | |
| JP5295273B2 (en) | Data stream filtering apparatus and method | |
| CN101283550A (en) | Data communication system and method with virtual ports | |
| US20070047546A1 (en) | Packet forwarding apparatus | |
| CA2259829A1 (en) | Scalable logical lan | |
| JP2013503580A (en) | Broadcast packet / multicast control message processing method and apparatus | |
| US20020089978A1 (en) | Control channel implementation in a packet switched communications network | |
| CN1937782B (en) | Frame device and its message transmitting method | |
| CN102223312A (en) | Flow control method and equipment based on link status | |
| CN110830394B (en) | Method for generating routing table based on RapidIO network | |
| EP1983693A1 (en) | Binding interface in rpr and its realizing method and apparatus | |
| CN101360062B (en) | Method and system implementing service of multi-point to multi-point Ethernet with root node | |
| CN101536410A (en) | LAN topology detection and assignment of addresses | |
| JP5449548B2 (en) | Multicast processing method and apparatus | |
| CN100454890C (en) | Data exchange method based on virtual local area network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No. Patentee after: Xinhua three Technology Co., Ltd. Address before: 310053 Hangzhou science and Technology Industrial Park, high tech Industrial Development Zone, Zhejiang Province, No. six and road, No. 310 Patentee before: Huasan Communication Technology Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20091118 Termination date: 20200528 |