CN103297355A - Data flow forwarding method and system and exchanger - Google Patents
Data flow forwarding method and system and exchanger Download PDFInfo
- Publication number
- CN103297355A CN103297355A CN2013102468524A CN201310246852A CN103297355A CN 103297355 A CN103297355 A CN 103297355A CN 2013102468524 A CN2013102468524 A CN 2013102468524A CN 201310246852 A CN201310246852 A CN 201310246852A CN 103297355 A CN103297355 A CN 103297355A
- Authority
- CN
- China
- Prior art keywords
- exchange chip
- outlet
- data flow
- shunting
- network port
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention discloses a data flow forwarding method and system and an exchanger. The data flow forwarding method includes that a current exchange chip determines the received objective exchange chip of data flows according to a preset two-layer forwarding table; the data flows are shunted to nonstop outlets and corresponding shunting outlets according to the bandwidth ratio of the nonstop outlets and the shunting outlets, wherein the current exchange chip passes through the nonstop outlets and the shunting outlets for reaching the objective exchange chip, the current exchange chip and the objective exchange chip are connected through the nonstop outlets, and the current exchange chip is connected with all transfer exchange chips through the corresponding shunting outlets; and the data flows of the nonstop outlets are sent to the objective exchange chip, and the data flows of all the shunting outlets are sent to the objective exchange chip through the corresponding transfer exchange chips. The data flow forwarding method and system and the exchanger improve the effective forwarding efficiency and reduce the packet loss probability.
Description
Technical field
The present invention relates to networking technology area, refer to a kind of data flow retransmission method, system and switch especially.
Background technology
Full mesh (FULLMESH) fabric switch is by the exchange chip of different slots is interconnected in twos, realizes netted switching network.Figure 1 shows that the structural representation of typical 3 slot FULLMESH fabric switch, a ply-yarn drill is set on each slot, interconnected by 2 HG42 between two ply-yarn drills, the ply-yarn drill that common FULLMESH fabric switch is supported is the ply-yarn drill that chip is changed in single cross, that is to say between two exchange chips interconnected by 2 HG42.
Gbps is an important indicator weighing cabinet formula switch performance between exchange chip, and it has determined switch inside can stride the bandwidth that sheet is transmitted.Current FULLMESH fabric switch is striden the data flow of sheet forwarding all by the direct-connected mode that reaches, and for example mails to the directly HG42 forwarding by connecting between the two of data flow of exchange chip 3 from exchange chip 1.
As can be seen from Figure 1, exchange chip 1 connects 4 HG42 altogether, the bandwidth of each HG42 is 40G, so the Gbps of exchange chip 1 is: 40G * 4=160Gbps, and the Gbps between exchange exchange chip 1 and the exchange chip 3 is 40G * 2=80Gbps, the network bandwidth of exchange chip 1 is generally 120G or 160G usually greater than 80G.Suppose this moment from the network port linear speed of exchange chip 1 to exchange chip 2 data flow of transmitting 160Gbps, so, can only transmit 80Gbps, abandon 80Gbps, this moment, effectively forward rate only was 50%.As seen, because the direct-connected mode that reaches of current employing is transmitted data flow, cause the problem that effective forward rate is not high, packet loss is serious.
Summary of the invention
The embodiment of the invention provides a kind of data flow retransmission method, system and switch, transmits data flow, the problem that effective forward rate is not high, packet loss is serious that causes in order to solve the direct-connected mode that reaches of available technology adopting.
A kind of data flow retransmission method comprises:
Current exchange chip is determined the purpose exchange chip of the data flow that receives according to presetting two-layer retransmitting table;
Arrive the bandwidth ratio of through outlet and the shunting outlet of described purpose exchange chip process according to described current exchange chip, described data flow is diverted to through outlet and corresponding shunting outlet, described current exchange chip is connected by described through outlet with described purpose exchange chip, and described current exchange chip connects by corresponding shunting outlet with each transfer exchange chip;
The data flow of described through outlet is sent to described purpose exchange chip, and the data flow of each the described shunting outlet transfer exchange chip by correspondence is sent to described purpose exchange chip.
A kind of data flow repeater system comprises:
A current exchange chip, a purpose exchange chip and at least one transfer exchange chip, described current exchange chip, purpose exchange chip and at least one transfer exchange chip are interconnected in twos, wherein:
Current exchange chip is used for determining the purpose exchange chip of the data flow that receives according to presetting two-layer retransmitting table; Arrive the bandwidth ratio of through outlet and the shunting outlet of described purpose exchange chip process according to described current exchange chip, described data flow is diverted to through outlet and corresponding shunting outlet, described current exchange chip is connected by described through outlet with described purpose exchange chip, and described current exchange chip connects by corresponding shunting outlet with each transfer exchange chip; The data flow of described through outlet is sent to described purpose exchange chip, and the data flow of each described shunting outlet is sent to corresponding transfer exchange chip;
Described transfer exchange chip, the data flow that is used for receiving by the shunting outlet of described current exchange chip is sent to described purpose exchange chip;
Described purpose exchange chip is used for receiving the data flow from described current exchange chip and the forwarding of described transfer exchange chip.
A kind of switch comprises above-mentioned data flow repeater system.
Beneficial effect of the present invention is as follows:
Data flow retransmission method, system and switch that the embodiment of the invention provides are determined the purpose exchange chip of the data flow that receives by current exchange chip according to presetting two-layer retransmitting table; Arrive the bandwidth ratio of through outlet and the shunting outlet of described purpose exchange chip process according to described current exchange chip, described data flow is diverted to through outlet and corresponding shunting outlet, described current exchange chip is connected by described through outlet with described purpose exchange chip, and described current exchange chip connects by corresponding shunting outlet with each transfer exchange chip; The data flow of described through outlet is sent to described purpose exchange chip, and the data flow of each the described shunting outlet transfer exchange chip by correspondence is sent to described purpose exchange chip.This scheme is shunted the data flow that receives, a part is transmitted by through outlet, a part is transmitted by the corresponding transfer exchange chip of shunting outlet, even the data volume of the data flow that receives is greater than the bandwidth of through outlet, can guarantee that also entire stream successfully is forwarded to the purpose exchange chip, thereby can improve effective forward rate, reduce packet loss.
Description of drawings
Fig. 1 is the structural representation of 3 groove switches in the background technology of the present invention;
Fig. 2 is the flow chart of data flow retransmission method in the embodiment of the invention;
Fig. 3 is the flow chart of data flow shunting in the embodiment of the invention;
The structural representation of data flow repeater system in Fig. 4 embodiment of the invention;
Fig. 5 is the schematic diagram of preferred data flow retransmission method in the embodiment of the invention.
Embodiment
Transmit data flow at the direct-connected mode that reaches of available technology adopting, the problem that effective forward rate is not high, packet loss is serious that causes, the embodiment of the invention provide a kind of data flow retransmission method, the flow process of this method as shown in Figure 2, execution in step is as follows:
S20: current exchange chip is determined the purpose exchange chip of the data flow that receives according to presetting two-layer retransmitting table.
S21: the bandwidth ratio that arrives through outlet and the shunting outlet of purpose exchange chip process according to current exchange chip, data flow is diverted to through outlet and corresponding shunting outlet, current exchange chip is connected by through outlet with the purpose exchange chip, and current exchange chip connects by corresponding shunting outlet with each transfer exchange chip.
Exchange chip in the switch is typically provided with the network port, the data flow that desampler is outer also is provided with the backboard port, and the data that are used between the exchange chip are transmitted, through outlet and shunting outlet are the backboard ports of current exchange chip, are used for the data flow that the transmission network port receives.The backboard port can be HG42 etc.
Each exchange chip is provided with a plurality of backboard ports, and the bandwidth of each backboard port can be certain, and for example, the bandwidth of existing backboard port is generally 40G.
By the bandwidth ratio of through outlet and shunting outlet, data flow is shunted.
Suppose that the exchange chip 1 among Fig. 1 is current exchange chip, exchange chip 2 is the purpose exchange chip, the backboard port that exchange chip 1 is connected with exchange chip 2 is HG1,2, the backboard port that exchange chip 1 is connected with exchange chip 3 is HG3,4, so, HG1,2 is through outlet, and HG3,4 is the shunting outlet.Wherein the shunting outlet can be 1, also can be a plurality of, mainly is to determine according to the number of exchange chip.
S22: the data flow of the outlet of will going directly is sent to the purpose exchange chip, and the data flow of each the shunting outlet transfer exchange chip by correspondence is sent to the purpose exchange chip.
This scheme is shunted the data flow that receives, a part is transmitted by through outlet, a part is transmitted by the corresponding transfer exchange chip of shunting outlet, even the data volume of the data flow that receives is greater than the bandwidth of through outlet, can guarantee that also entire stream successfully is forwarded to the purpose exchange chip, thereby can improve effective forward rate, reduce packet loss.
Concrete, arriving the through outlet of purpose exchange chip process and shunting the bandwidth ratio that exports according to current exchange chip among the above-mentioned S21, the shunting that data flow is diverted to through outlet and correspondence exports, and as shown in Figure 3, specifically comprises:
S30: the ratio of the bandwidth standard configuration value of the bandwidth standard configuration value of the outlet of will going directly and each shunting outlet, as the bandwidth ratio of through outlet and each shunting outlet.
Continue to continue to use example, suppose that current exchange chip 1 has 4 backboard ports, the bandwidth standard configuration value of each backboard port is 40G, and wherein the bandwidth standard configuration value of through outlet is 80G, the bandwidth standard configuration value of shunting outlet is 80G, so the bandwidth ratio of through outlet and shunting outlet is 1:1.
Certainly, bandwidth ratio according to the through outlet of actual conditions and shunting outlet also can be worth for other, for example, have 4 exchange chips, exchange chip 1 is current exchange chip, exchange chip 2 is the purpose exchange chip, the backboard port that exchange chip 1 is connected with exchange chip 2 is HG1,2, the backboard port that exchange chip 1 is connected with exchange chip 3 is HG3,4,5, the backboard port that exchange chip 1 is connected with exchange chip 4 is HG6,7,8,9, so, HG1,2 are through outlet, HG3,4,5 are shunting outlet 1, HG6,7,8,9 are shunting outlet 2, if the bandwidth of each HG is 40G, so, the bandwidth standard configuration value of through outlet is 40G * 2=80G, and the bandwidth standard configuration value of shunting outlet 1 is 40G * 3=120G, the bandwidth standard configuration value of shunting outlet 2 is 40G * 4=160G, through outlet: shunting outlet 1: shunting outlet 2=80:120:160=2:3:4.
S31: according to bandwidth ratio, network port grouping with current exchange chip, determine the network port grouping corresponding with through outlet and the network port grouping corresponding with each shunting outlet, and according to bandwidth ratio, the corresponding network port grouping of through outlet and the corresponding network port grouping of each shunting outlet will be injected into after the data flow shunting that receive.
If the bandwidth ratio of through outlet and shunting outlet is 1:1, the network port has four, then these 4 network ports can be divided into two groups, any two is 1 group, and remaining two is 1 group, and the data volume of event data stream is 40G, so, just can inject the data of 10G at 4 network ports respectively; If the bandwidth ratio of through outlet and shunting outlet is 1:1, the network port has five, then can select 4 network ports and be divided into two groups, any two is 1 group, and remaining two is 1 group, and the data volume of event data stream is 40G, so, just can inject the data of 10G at 4 network ports respectively.
If through outlet: shunting outlet 1: the bandwidth ratio of shunting outlet 2 is 2:3:4, when 48 network ports, can select 45 network ports, 45 network ports selecting is divided into groups again; If 45 network ports are arranged, just can directly divide into groups.From 45 network ports or existing 45 network ports selected, select the corresponding network port grouping of the through outlet of 10 conducts arbitrarily, from 35 remaining network ports, select 15 arbitrarily as the network port grouping of diversion port 1 correspondence, 20 remaining network ports are as the network port grouping of diversion port 2 correspondences, then data flow on average is injected in 45 network ports, thereby has realized the data flow that receives is shunted.
Here only be with through outlet: shunting outlet 1: the bandwidth ratio of shunting outlet 2 is 2:3:4, the network port is that 48 or 45 describe for example, if the bandwidth ratio of through outlet and shunting outlet is other value, when the data of the network port are worth for other, equally according to the method described above the network port is divided into groups, and data stream is shunted.
S32: the data flow during the corresponding network port of through outlet divides into groups is forwarded to through outlet, and the data flow that each shunting exports in the corresponding network port grouping is forwarded to corresponding shunting outlet.
So just can finish the forwarding of data flow.
Concrete, the data flow that each shunting is exported among the above-mentioned S22 is sent to the purpose exchange chip by the transfer exchange chip of correspondence, specifically comprises: the backboard port that the data flow of each shunting outlet is forwarded to corresponding transfer exchange chip; Each transfer exchange chip is by transmitting the data flow that receives with the direct-connected backboard port of purpose exchange chip.
Data flow of each shunting outlet is forwarded to corresponding transfer exchange chip, in the middle of deliver change chip and receive data flow by the backboard port after, just forward it to the purpose exchange chip by direct-connected backboard port.
Wherein, the corresponding relation of through outlet, diversion port and network port group, the backboard port that goes directly between transfer exchange chip and purpose exchange chip can be arranged on during clean culture transmits.
Preferably, the data flow in above-mentioned S21 is diverted to before through outlet and the shunting outlet, also comprises: the data volume of specified data stream is greater than the bandwidth of through outlet.
A kind of preferred scheme, can during greater than the bandwidth of through outlet, adopt the method for the embodiment of the invention to transmit in the data volume of data flow, in the data volume of data flow during smaller or equal to the bandwidth of through outlet, only use through outlet to transmit, just adopt existing method to transmit.
Based on same inventive concept, the embodiment of the invention provides a kind of data flow repeater system, can be arranged in the switch, the structure of this system as shown in Figure 4, comprise a current exchange chip 40, a purpose exchange chip 42 and at least one transfer exchange chip 41, current exchange chip 40, purpose exchange chip 42 and at least one transfer exchange chip 41 are interconnected in twos, wherein:
Concrete, current exchange chip 40 is used for arriving the through outlet of purpose exchange chip process and shunting the bandwidth ratio that exports according to current exchange chip, and the shunting that data flow is diverted to through outlet and correspondence exports, and specifically is used for:
The through bandwidth standard configuration value that exports and each are shunted the ratio of the bandwidth standard configuration value of outlet, shunt the bandwidth ratio of outlet as through outlet and each;
According to bandwidth ratio, network port grouping with current exchange chip, determine the network port grouping corresponding with through outlet and the network port grouping corresponding with each shunting outlet, and according to bandwidth ratio, the corresponding network port grouping of through outlet and the corresponding network port grouping of each shunting outlet will be injected into after the data flow shunting that receive;
Data flow during the corresponding network port of through outlet divides into groups is forwarded to through outlet, and the data flow that each shunting exports in the corresponding network port grouping is forwarded to corresponding shunting outlet.
Concrete, above-mentioned current exchange chip 40, the concrete backboard port that is used for the data flow of each shunting outlet is forwarded to corresponding transfer exchange chip 41.
Above-mentioned transfer exchange chip 41, concrete being used for by transmitting the data flow that receives with the direct-connected backboard port of purpose exchange chip 42.
Preferably, above-mentioned current exchange chip 40 also was used for before data flow is diverted to through outlet and shunting outlet, and the data volume of specified data stream is greater than the bandwidth of through outlet.
Be that example specifies the data flow retransmission method in the embodiment of the invention with the switch that comprises 5 exchange chips below, as shown in Figure 5, there is shown 4 paths that data flow that exchange chip 1 receives is forwarded to exchange chip 5, represent with the line of 4 band arrows.Wherein, the quantity of the quantity of exchange chip and the slot of switch does not have positive connection, can dispose 1 exchange chip by each slot, can be at the exchange chip of 1 slot deployment more than 2 yet.
Suppose to connect by 3 HG42 between any two exchange chips, each exchange chip is connected with 4 exchange chips, and namely each exchange chip connects 12 HG42, so each exchange chip is supported the bandwidth of 3 * 4 * 40=480Gbps.Suppose that 5 exchange chip signs are respectively 1,2,3,4,5, and shown in the backboard connected mode table 1 of 5 exchange chips:
Table 1
The network port of supposing exchange chip is numbered: 1～48(bandwidth is 10G), the backboard port numbering is HG1～12.
Can pre-configured clean culture transmit, the layoutprocedure that the clean culture of 5 exchange chips is transmitted is the same, and transmitting layoutprocedure with the clean culture of exchange chip 1 below is that example describes:
The network port is divided into 4 groups that bandwidth equates, Class-1={1,2 ..., 12}; Class-2={13,14 ..., 24}; Class-3={25,26 ..., 36}; Class-4={37,38 ..., 48}.
As above, the bandwidth of every group network port is 12 * 10=120Gbps.
The backboard port is divided into 4 groups that bandwidth equates, TRUNK-1={HG1,2,3}; TRUNK-2={HG4,5,6}; TRUNK-3={HG7,8,9}; TRUNK-4={HG10,11,12}.
Clean culture after the configuration is transmitted as shown in table 2:
Table 2
Wherein, the forwarding behavior of the network port arranges as shown in table 3:
Table 3
The forwarding behavior of backboard port is set to:
Table 4
When supposing to receive the data flow of 480Gbps from 48 network ports of exchange chip 1, determine that by presetting two-layer retransmitting table the purpose exchange chip is exchange chip 2, can enter the data flow of 120Gbps from the network port of Class1～4 respectively:
Can know by look-up table 3, the data flow that Class-1 injects is transmitted according to UC_MODPORT_MAP-1, can know by look-up table 2, and the purpose exchange chip is that 2 outlet information is TRUNK-1, therefore, the data flow of Class-1 injection is forwarded to TRUNK-1 and arrives purpose exchange chip 2;
Can know by look-up table 3, the data flow that Class-2 injects is transmitted according to UC_MODPORT_MAP-2, can know by look-up table 2, the purpose exchange chip is that 2 outlet information is TRUNK-2, therefore, the data flow that Class-2 injects is forwarded to TRUNK-2 and reaches transfer exchange chip 3, therefrom deliver the backboard port that changes chip 3 to enter because be, can know by look-up table 4, transmit data flow according to UC_MODPORT_MAP-5, can know that by UC_MODPORT_MAP-5 the purpose exchange chip is that 2 outlet information is TRUNK-1, namely the TRUNK-1 by transfer exchange chip 3 is forwarded to purpose exchange chip 2 with data flow;
Can know by look-up table 3, the data flow that Class-3 injects is transmitted according to UC_MODPORT_MAP-3, can know by look-up table 2, the purpose exchange chip is that 2 outlet information is TRUNK-3, therefore, the data flow that Class-3 injects is forwarded to TRUNK-3 and reaches transfer exchange chip 4, therefrom deliver the backboard port that changes chip 4 to enter because be, can know by look-up table 4, to transmit data flow according to the UC_MODPORT_MAP-5 of transfer exchange chip 4, can know that by UC_MODPORT_MAP-5 the purpose exchange chip is that 2 outlet information is TRUNK-1, namely the TRUNK-1 by transfer exchange chip 4 is forwarded to purpose exchange chip 2 with data flow;
Can know by look-up table 3, the data flow that Class-4 injects is transmitted according to UC_MODPORT_MAP-4, can know by look-up table 2, the purpose exchange chip is that 2 outlet information is TRUNK-4, therefore, the data flow that Class-4 injects is forwarded to TRUNK-4 and reaches transfer exchange chip 5, therefrom deliver the backboard port that changes chip 5 to enter because be, can know by look-up table 4, to transmit data flow according to the UC_MODPORT_MAP-5 of transfer exchange chip 5, can know that by UC_MODPORT_MAP-5 the purpose exchange chip is that 2 outlet information is TRUNK-1, namely the TRUNK-1 by transfer exchange chip 4 is forwarded to purpose exchange chip 2 with data flow;
So far finished from exchange chip 1 to exchange chip the forwarding of 2 data flow.
Therefore the data flow of the through purpose exchange chip 2 of the TRUNK-1 by exchange chip 1 is 120Gbps; TRUNK-2 by exchange chip 1 reaches transfer exchange chip 3, and the data flow that arrives purpose exchange chip 2 by transfer exchange chip 3 is 120Gbps again; TRUNK-3 by exchange chip 1 reaches transfer exchange chip 4, and the data flow that arrives purpose exchange chip 2 by transfer exchange chip 4 is 120Gbps again; TRUNK-4 by exchange chip 1 reaches transfer exchange chip 5, and the data flow that arrives purpose exchange chip 2 by transfer exchange chip 5 is 120Gbps again; Effectively forward rate is 100%, and packet loss is 0.
If with traditional method, exchange chip 1 is 3 * 40=120Gbps to the bandwidth of exchange chip 2, this moment is if exchange chip 1 all backboard ports are transmitted data flow to exchange chip 2, tentation data stream is 480Gbps, only can transmit 120Gbps, abandon 360Gbps, this moment is forward rate only 25% effectively, and packet loss is 75%.
Therefore, the data flow retransmission method that provides in the embodiment of the invention has improved effective forward rate, has reduced packet loss.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (9)
1. a data flow retransmission method is characterized in that, comprising:
Current exchange chip is determined the purpose exchange chip of the data flow that receives according to presetting two-layer retransmitting table;
Arrive the bandwidth ratio of through outlet and the shunting outlet of described purpose exchange chip process according to described current exchange chip, described data flow is diverted to through outlet and corresponding shunting outlet, described current exchange chip is connected by described through outlet with described purpose exchange chip, and described current exchange chip connects by corresponding shunting outlet with each transfer exchange chip;
The data flow of described through outlet is sent to described purpose exchange chip, and the data flow of each the described shunting outlet transfer exchange chip by correspondence is sent to described purpose exchange chip.
2. the method for claim 1, it is characterized in that, arrive the through outlet of described purpose exchange chip process and the bandwidth ratio that shunting exports according to described current exchange chip, described data flow be diverted to through outlet export with corresponding shunting, specifically comprise:
With the ratio of the bandwidth standard configuration value of the bandwidth standard configuration value of described through outlet and each described shunting outlet, as the bandwidth ratio of described through outlet and each described shunting outlet;
According to described bandwidth ratio, network port grouping with described current exchange chip, determine the network port grouping corresponding with described through outlet and the network port grouping corresponding with each described shunting outlet, and according to described bandwidth ratio, network port grouping and the corresponding network port grouping of each described shunting outlet of described through outlet correspondence will be injected into after the data flow shunting that receive;
Data flow in the network port grouping of described through outlet correspondence is forwarded to described through outlet, and the data flow during the corresponding network port of each described shunting outlet divides into groups is forwarded to corresponding shunting outlet.
3. the method for claim 1 is characterized in that, the data flow of each the described shunting outlet transfer exchange chip by correspondence is sent to described purpose exchange chip, specifically comprises:
The data flow of each described shunting outlet is forwarded to the backboard port of corresponding transfer exchange chip;
Each described transfer exchange chip is by transmitting the data flow that receives with the direct-connected backboard port of described purpose exchange chip.
4. as the arbitrary described method of claim 1-3, it is characterized in that described data flow is diverted to before the shunting outlet of through outlet and correspondence, also comprises:
Determine that the data volume of described data flow is greater than the bandwidth of described through outlet.
5. data flow repeater system, it is characterized in that, comprise a current exchange chip, a purpose exchange chip and at least one transfer exchange chip, described current exchange chip, purpose exchange chip and at least one transfer exchange chip are interconnected in twos, wherein:
Current exchange chip is used for determining the purpose exchange chip of the data flow that receives according to presetting two-layer retransmitting table; Arrive the bandwidth ratio of through outlet and the shunting outlet of described purpose exchange chip process according to described current exchange chip, described data flow is diverted to through outlet and corresponding shunting outlet, described current exchange chip is connected by described through outlet with described purpose exchange chip, and described current exchange chip connects by corresponding shunting outlet with each transfer exchange chip; The data flow of described through outlet is sent to described purpose exchange chip, and the data flow of each described shunting outlet is sent to corresponding transfer exchange chip;
Described transfer exchange chip, the data flow that is used for receiving by the shunting outlet of described current exchange chip is sent to described purpose exchange chip;
Described purpose exchange chip is used for receiving the data flow from described current exchange chip and the forwarding of described transfer exchange chip.
6. system as claimed in claim 5, it is characterized in that, described current exchange chip, be used for arriving according to described current exchange chip the bandwidth ratio of through outlet and the shunting outlet of described purpose exchange chip process, described data flow is diverted to through outlet and corresponding shunting outlet, specifically is used for:
With the ratio of the bandwidth standard configuration value of the bandwidth standard configuration value of described through outlet and each described shunting outlet, as the bandwidth ratio of described through outlet and each described shunting outlet;
According to described bandwidth ratio, network port grouping with described current exchange chip, determine the network port grouping corresponding with described through outlet and the network port grouping corresponding with each described shunting outlet, and according to described bandwidth ratio, network port grouping and the corresponding network port grouping of each described shunting outlet of described through outlet correspondence will be injected into after the data flow shunting that receive;
Data flow in the network port grouping of described through outlet correspondence is forwarded to described through outlet, and the data flow during the corresponding network port of each described shunting outlet divides into groups is forwarded to corresponding shunting outlet.
7. system as claimed in claim 6 is characterized in that, described current exchange chip, the concrete backboard port that is used for the data flow of each described shunting outlet is forwarded to corresponding transfer exchange chip;
Described transfer exchange chip, concrete being used for by transmitting the data flow that receives with the direct-connected backboard port of described purpose exchange chip.
8. as the arbitrary described system of claim 5-7, it is characterized in that described current exchange chip also is used for determining that the data volume of described data flow was greater than the bandwidth of described through outlet before described data flow is diverted to through outlet and shunting outlet.
9. a switch is characterized in that, comprises arbitrary described data flow repeater system as claim 5-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310246852.4A CN103297355B (en) | 2013-06-20 | 2013-06-20 | A kind of stream compression forwarding method, system and switch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310246852.4A CN103297355B (en) | 2013-06-20 | 2013-06-20 | A kind of stream compression forwarding method, system and switch |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103297355A true CN103297355A (en) | 2013-09-11 |
| CN103297355B CN103297355B (en) | 2016-08-10 |
Family
ID=49097693
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310246852.4A Active CN103297355B (en) | 2013-06-20 | 2013-06-20 | A kind of stream compression forwarding method, system and switch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103297355B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104683249A (en) * | 2015-02-26 | 2015-06-03 | 浪潮电子信息产业股份有限公司 | Independent configurable interconnection module implementing method for multi-chip interconnection system |
| CN106936737A (en) * | 2017-04-24 | 2017-07-07 | 北京星网锐捷网络技术有限公司 | Port Mirroring implementation method and device based on the network equipment |
| CN107196802A (en) * | 2017-05-27 | 2017-09-22 | 全球能源互联网研究院 | A kind of Outband network management system |
| WO2021139286A1 (en) * | 2020-01-07 | 2021-07-15 | 华为技术有限公司 | Load sharing method, device, and network system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101262436A (en) * | 2008-04-08 | 2008-09-10 | 华为技术有限公司 | Multicast duplication method, device and line card |
| CN101789875A (en) * | 2010-02-10 | 2010-07-28 | 华为技术有限公司 | Method, device and router for forwarding data flow |
| CN102123080A (en) * | 2010-12-31 | 2011-07-13 | 杭州华三通信技术有限公司 | Data forwarding device and cable fastener plate and data forwarding method |
-
2013
- 2013-06-20 CN CN201310246852.4A patent/CN103297355B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101262436A (en) * | 2008-04-08 | 2008-09-10 | 华为技术有限公司 | Multicast duplication method, device and line card |
| CN101789875A (en) * | 2010-02-10 | 2010-07-28 | 华为技术有限公司 | Method, device and router for forwarding data flow |
| CN102123080A (en) * | 2010-12-31 | 2011-07-13 | 杭州华三通信技术有限公司 | Data forwarding device and cable fastener plate and data forwarding method |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104683249A (en) * | 2015-02-26 | 2015-06-03 | 浪潮电子信息产业股份有限公司 | Independent configurable interconnection module implementing method for multi-chip interconnection system |
| CN106936737A (en) * | 2017-04-24 | 2017-07-07 | 北京星网锐捷网络技术有限公司 | Port Mirroring implementation method and device based on the network equipment |
| CN106936737B (en) * | 2017-04-24 | 2020-01-21 | 北京星网锐捷网络技术有限公司 | Port mirroring implementation method and device based on network equipment |
| CN107196802A (en) * | 2017-05-27 | 2017-09-22 | 全球能源互联网研究院 | A kind of Outband network management system |
| WO2021139286A1 (en) * | 2020-01-07 | 2021-07-15 | 华为技术有限公司 | Load sharing method, device, and network system |
| US11824781B2 (en) | 2020-01-07 | 2023-11-21 | Huawei Technologies Co., Ltd. | Method, device, and network system for load balancing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103297355B (en) | 2016-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103501236B (en) | Network control plane logical topology generates method and device | |
| CN102726009B (en) | Data flow exchange control method and relative device | |
| CN103297355A (en) | Data flow forwarding method and system and exchanger | |
| CN101873269B (en) | Data retransmission device and method for distributing buffer to ports | |
| CN101764729B (en) | Serial server having switching function | |
| RU2007111857A (en) | RING NETWORK, COMMUNICATION DEVICE AND OPERATIONAL MANAGEMENT METHOD USED FOR THE RING NETWORK AND COMMUNICATION DEVICE | |
| CN102238072A (en) | Method for dynamically selecting routing and CLOS (Charles Clos) switching network system | |
| CN103200101B (en) | Based on method and the device of the openflow multi-table query of exchange chip | |
| CN103746922B (en) | Method and device for forwarding multicast messages | |
| CN103095568A (en) | System and method for achieving stacking of rack type switching devices | |
| CN103414801A (en) | Method and device for synchronizing medium access control addresses in stacking system | |
| CN102111327A (en) | Method and system for cell dispatching | |
| CN102870381B (en) | A kind of PCIE switching system, device and switching method | |
| CN103546397B (en) | Support out of order from route Omega network | |
| CN104169903B (en) | A kind of method and system realizing multi-channel synchronous parallel transmission | |
| CN102447640A (en) | Method for carrying out aggregate routing in IRF (intelligent resilient framework) system and machine frame switch | |
| CN104796343A (en) | Communication structure based on network-on-chip | |
| CN104699649A (en) | Multi-branch serial bus interface and data exchange method | |
| CN104917704B (en) | 10GBase R PCS and 40GBase R PCS method and system are multiplexed in same framework | |
| CN101778044B (en) | Switched network system structure with adjustable throughput rate | |
| CN104580028B (en) | Exchange chip message processing method and device based on expansion interface | |
| CN101539871A (en) | Device for balancedly processing load, data processing system and method and main control panel | |
| CN108712347A (en) | A kind of data transmission system and its transmission method of interchanger | |
| CN105791141B (en) | A kind of flow table method of replacing and device | |
| CN100563189C (en) | A kind of Ethernet transmitting device and method that is lower than the gigabit transmission rate that be suitable for |
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 | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Cangshan District of Fuzhou City, Fujian province 350002 Jinshan Road No. 618 Garden State Industrial Park 19 floor Patentee after: RUIJIE NETWORKS CO., LTD. Address before: Cangshan District of Fuzhou City, Fujian province 350002 Jinshan Road No. 618 Garden State Industrial Park 19 floor Patentee before: Fujian Xingwangruijie Network Co., Ltd. |
|
| CP01 | Change in the name or title of a patent holder |