CN101227402B - Method and apparatus for sharing polymerization link circuit flow - Google Patents
Method and apparatus for sharing polymerization link circuit flow Download PDFInfo
- Publication number
- CN101227402B CN101227402B CN2008100079268A CN200810007926A CN101227402B CN 101227402 B CN101227402 B CN 101227402B CN 2008100079268 A CN2008100079268 A CN 2008100079268A CN 200810007926 A CN200810007926 A CN 200810007926A CN 101227402 B CN101227402 B CN 101227402B
- Authority
- CN
- China
- Prior art keywords
- port
- sharing
- value
- link circuit
- circuit flow
- 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
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a sharing method of polymeric link circuit flow, which is used to share flow among polymeric link circuit member ports. The method comprises the following steps: getting to know at least a dynamic parameter value which can dynamically reflect forwarding performance of the port of the member ports, leading static state parameter value of message which is waiting to be transmitted and the dynamic parameter value to be used as the input of polymeric link circuit flow sharing algorithm together, operating the polymeric link circuit flow sharing algorithm, transmitting the message which is waiting to be retransmitted on an outlet port which is selected by the sharing algorithm. The invention also discloses a sharing device of polymeric link circuit flow. The sharing method of polymeric link circuit flow of the invention leads flow load on the polymeric link circuit to get effective sharing.
Description
Technical field
The present invention relates to a kind of network technology, especially with a kind of sharing polymerization link circuit flow method and install relevant.
Background technology
Link aggregation (Link Aggregation) is that a plurality of physical ports are bundled into an aggregation port group, to realize load flow sharing in converging group membership's port, also provides higher connection reliability simultaneously.The link aggregation technology is a kind of Ethernet interconnection technique commonly used, can be used for connection between the ethernet device or ethernet device to the connection between the high-speed servers.It is by tying up a plurality of physical ports, provides a kind of mode of cheap and simple to come the bandwidth of Extended Capabilities Port, and higher connection reliability also is provided simultaneously.The one group of port that bundles is formed an aggregation group, an aggregation group is regarded a logic port, the message of the aggregation group of coming in and going out can distribute with the flow between the balance ports according to certain mode on the member port in aggregation group, steadily increases port bandwidth.Have a connection to break down in aggregation group, the flow on the non-working port will distribute on other ports, and the assurance business can not interrupted; After fault connect to be recovered, flow can be automatically balance again, the connection after guaranteeing to recover can continue to share whole flow.
How between the aggregated links member port, to carry out effective traffic sharing, become an emphasis in the link aggregation technology.In existing traffic sharing technology, the general input of adopting the one or more conducts in the information such as source MAC to be E-Packeted, target MAC (Media Access Control) address, source IP address, purpose IP address, source L4 port numbers, purpose L4 port numbers, reception physical port, reception type of message to share algorithm.Obviously, for certain network application of determining, these information are not subjected to the influence of time, are changeless, are static parameter value described to be E-Packeted.By the simple process to these information, for example intercepting wait the to E-Packet last 3bit of source MAC and target MAC (Media Access Control) address carries out XOR (XOR) operation, just can obtain the different outbound ports that message is transmitted, the sharing of realization flow.Hence one can see that, the input information of the traffic sharing method of prior art is reflection wait the to E-Packet static information of attribute, the actual flow load condition that can not reflect each member port in the aggregated links, more can not share mode adjustment to polylink flow according to these actual load situations, though the function of energy simple realization traffic sharing, but can not realize effectively sharing of flow, can bring between the aggregation group member port traffic load to share inequality, the situation of difference great disparity, make the little port of load have bandwidth waste, especially can be more serious to the excessive port problem of loading, meeting occurs congested and causes packet loss, and this makes the link aggregation technology also be very limited in practical application.For example, the physical port of two phase same rates is formed an aggregation group, carries out xor operation according to source MAC and the last 1bit of target MAC (Media Access Control) address and selects the outbound port that E-Packets, is 0 and transmits from 1 port, is 1 and transmits from 2 ports.When the flow of 1 port has reached fully loaded 99%, and the flow of 2 ports only is 1% o'clock, there is a new Business Stream to add again, and the result is 1 port according to this traffic sharing mode, this flow will be added in original with regard on the 1 crowded port link so, very likely cause 1 port congested, and 2 port flows only are 1%.Caused great waste.As seen, existing traffic sharing method can not realize that the port bandwidth resources effective is shared, causes inevitable bandwidth conflict and waste.
Summary of the invention
At the sharing polymerization link circuit flow inequality that exists in the prior art, the problem that causes the port bandwidth resource fully effectively not utilize, the object of the present invention is to provide a kind of sharing polymerization link circuit flow equilibrium, each member port bandwidth resources can access the sharing polymerization link circuit flow method of fully effectively utilizing.
Another object of the present invention is to provide a kind of sharing polymerization link circuit flow device of realizing sharing polymerization link circuit flow method of the present invention.
For achieving the above object, technical scheme of the present invention is as follows:
MAC Address sharing polymerization link circuit flow method of the present invention may further comprise the steps:
A kind of sharing polymerization link circuit flow method is used for the traffic sharing between the aggregated links member port, may further comprise the steps:
Step S100: know the dynamic parameter value that at least one of each described member port can this port forwarding performance of dynamic reflection;
Step S200: static parameter value that will be to be E-Packeted and described dynamic parameter value, common input as the sharing polymerization link circuit flow algorithm, move described sharing polymerization link circuit flow algorithm, wait to E-Packet during selective polymerization group outbound port for described, carry out probability according to the transfer capability value of member port and select;
Step S300: with described wait to E-Packet transmit in the described selected outbound port of algorithm of sharing.
A kind of sharing polymerization link circuit flow device of realizing sharing polymerization link circuit flow method of the present invention, described sharing polymerization link circuit flow device comprise parameter acquisition module, port selection module and forwarding module;
Parameter acquisition module: be used to obtain the dynamic parameter value that at least one of each described member port can this port forwarding performance of dynamic reflection;
Port is selected module: be connected in described parameter acquisition module, the dynamic parameter value that static parameter value described to be E-Packeted and described parameter acquisition module are transmitted, common input as the sharing polymerization link circuit flow algorithm, move described sharing polymerization link circuit flow algorithm, wait to E-Packet during selective polymerization group outbound port for described, carry out probability according to the transfer capability value of member port and select;
Forwarding module: be connected in described port and select module, what receive that described port selects that module transmits shares the outbound port that algorithm is selected by described, with described wait to E-Packet transmit in described outbound port.
As shown from the above technical solution, the present invention has following beneficial effect: sharing polymerization link circuit flow method of the present invention when carrying out sharing polymerization link circuit flow, determines to share mode jointly by fixing application message and aggregation group member port flow.When carrying out sharing polymerization link circuit flow, consider that the ability situation of port changes, dynamically adjust the traffic sharing mode.The present invention can make the traffic load on the aggregated links effectively be shared, and has made full use of port bandwidth, has improved bandwidth utilization.
Below by drawings and Examples, technical scheme of the present invention is described in further detail.
Description of drawings
Fig. 1 is the flow chart of sharing polymerization link circuit flow method of the present invention;
Fig. 2 is the flow chart that the port ability value of sharing polymerization link circuit flow method of the present invention regularly upgrades;
The flow chart that Fig. 3 selects for the forwarding port of sharing polymerization link circuit flow method of the present invention;
Fig. 4 is a sharing polymerization link circuit flow schematic representation of apparatus of the present invention.
Embodiment
The invention will be further described below in conjunction with the accompanying drawings and the specific embodiments.
Sharing polymerization link circuit flow method of the present invention is used for the traffic sharing between the aggregated links member port as shown in Figure 1, be may further comprise the steps:
Step S100: know the dynamic parameter value that at least one of each described member port can this port forwarding performance of dynamic reflection;
Step S200: at least one static parameter value and described dynamic parameter value that will be to be E-Packeted, as the input of sharing polymerization link circuit flow algorithm, move described sharing polymerization link circuit flow algorithm jointly;
Step S300: with described wait to E-Packet transmit in the described selected outbound port of algorithm of sharing.
The sharing polymerization link circuit flow method of the embodiment of the invention, described dynamic parameter value is the port ability value that is used for the current available remaining bandwidth of the described port of nominal, and described current available remaining bandwidth can be deducted the present flow rate occupied bandwidth by the maximum bandwidth of described port support and obtain.
In order to reach the purpose of flow equalization, the aggregated links sharing method of the embodiment of the invention, not simple with static parameter to be E-Packeted, source/target MAC (Media Access Control) address promptly to be E-Packeted, source/purpose IP address, source/purpose L4 port numbers, reception physical port etc. is used as the foundation of sharing polymerization link circuit flow, but this dynamic parameter of port ability value that adopts at least one static parameter and the current available remaining bandwidth of nominal polymerization member port decides the mode of traffic sharing jointly, is about at least one static parameter and port ability value jointly as the input of traffic sharing algorithm.At present, the study of Ethernet two layer MAC address is all learnt on the aggregation group index, therefore, sharing polymerization link circuit flow method of the present invention, port flow mainly is meant the flow of outgoing direction.
For getting access to the real-time outflow information of each member port of aggregation group, promptly obtain current outflow occupied bandwidth, and then obtain the port ability value, at first need to be statistical counter of each member port configuration of aggregation group.The statistical counter of port can be the counter of hardware, also can be the counter of software.From promptness and the minimizing guaranteeing to add up the consumption of cpu resource is considered, preferably hardware counter.In addition, statistical counter can also can be added up based on byte based on counting messages, if reach the high-precision load balancing effect of sharing polymerization link circuit flow, be preferably based on byte and carry out statistical counting, but the hardware counter bit wide that the byte statistical counter need take is bigger than counting messages counter, and cost must improve, in actual applications, can be according to actual conditions, take all factors into consideration according to cost performance and to choose.
Carry out the statistical value of real-time statistics for statistical counter, the link aggregation of the embodiment of the invention is shared algorithm, can read (Direct MemoryAccess by software timing inquiry or the regularly direct internal memory of hardware, be called for short DMA) transmit the statistical technique value that gets access to the port statistics counter, regularly or receive current bandwidth value of taking of obtaining each member port of aggregation group after waiting to E-Packet in real time.After the current bandwidth value of taking that gets access to each member port of port trunking group, calculate the current available remaining bandwidth of each member port of aggregation group, its maximum bandwidth by port support separately deducts present flow rate occupied bandwidth value and obtains.The link aggregation of the embodiment of the invention is shared algorithm, and the inlet port ability value comes the current available remaining bandwidth of nominal.For example, be two ports of 100M and 10M for the speed that comprises full duplex in the aggregation group, when the port statistics counter was 0, the ability value of 100M port was decided to be 100, and the ability value of 10M port is decided to be 10.
After the port ability value of each member port of aggregation group is determined, can calculate the total capacity value of aggregation group.For example, the aggregation group that above-mentioned two ports are formed, its current total capacity value is 110.The aggregated links sharing method of the embodiment of the invention, in described sharing polymerization link circuit flow algorithm, receive one described wait to E-Packet after, to wait to E-Packet to this and transmit the selection of port, equal the port ability value of this member port at the shared percentage of the total capacity value of all of the port and each member port of aggregation group is selected as the probability of outbound port.In other words, the likelihood ratio that each member port of aggregation group is selected as outbound port is the ratio of the port ability value of each port, and promptly the port ability value is that 100 member port has 10 times to be that the probability of 10 member port is selected as one and waits to E-Packet or the outbound port of a new stream to the port ability value; When the port statistics counter by each member port calculates the present bandwidth consumed 95M of 100M port, when 10M port bandwidth consumed is 2M, 100M port corresponding port ability value is 5, the ability value of 10M port becomes 8, the likelihood ratio that this time, two ports were selected as outbound port becomes 5: 8, and promptly the 10M port more may be selected as the outbound port of a new forwarding flow.
The sharing polymerization link circuit flow algorithm of the embodiment of the invention, the described algorithm of sharing can be realized by hardware logic, also can realize by software, for example can set up one by the mapping of described static parameter collection to described each member port collection, the mapped probability of two described member ports equals the ratio of the port ability value of described two member's ports.Choose a static parameter to be E-Packeted, for example the source address of message is got four binary value thereafter, and the port ability value of supposing selectable port one is 50, the port ability value of port 2 is 25, the port ability value of port 3 is 25; Then this mapping can be mapped to port one to 0000～0111 of binary value, binary value 1000～1011 is mapped to port 2,1100～1111 of binary value is mapped to port 3, back four binary value of source address is 0000～0111 then to select port one be outbound port if promptly wait to E-Packet, back four binary value of source address is 1000～1011 then reflects and select port 2 to be outbound port if wait to E-Packet, if back four binary value of the source address of waiting to E-Packet is 1100～1111 then selects port 3 to be outbound port.
The sharing polymerization link circuit flow method of another embodiment of the present invention, in described sharing polymerization link circuit flow algorithm, describedly wait to E-Packet or message flow to one, a member port is selected as describedly waiting to E-Packet or probability that message flow is transmitted port equals the weighting total capacity value shared percentage of the weighting port ability value of this member port at all of the port.Promptly the port ability value is weighted, for example, can be to considering the situation of each member port full and half duplex of aggregation group, the weight coefficient of the described weighting port ability value of the member port of full duplex is 2, the weight coefficient of the described weighting port ability value of semiduplex member port is 1.Also can take all factors into consideration this weight coefficient is set according to the up-downgoing flow proportional.
In polymerization traffic sharing method of the present invention, the input of sharing algorithm increased the port ability value this can this port forwarding performance of dynamic reflection the parameter value of dynamic parameter, with static parameter value described to be E-Packeted, common input as the sharing polymerization link circuit flow algorithm, move described sharing polymerization link circuit flow algorithm, select one to transmit outbound port described to be E-Packeted.Because the port ability value of each member port of aggregation group is based on flow and takies that situation calculates, and therefore is the parameter of a dynamic change.When being certain message or message flow forwarding selective polymerization group outbound port, carrying out probability according to the transfer capability value of member port selects, be equivalent in sharing algorithm, to have introduced the negative feedback of the real-time operating position of link bandwidth, guaranteed fully effectively utilizing of aggregation group bandwidth.
The sharing polymerization link circuit flow method of the embodiment of the invention, when the aggregation group member port broke down, the processing method of prior art need be adjusted effective port of aggregated links, and non-working port is got rid of from effective port.At the aggregated links sharing method of the embodiment of the invention, do not need to upgrade the composition situation of the effective member port of aggregation group, it is 0 just passable that the port ability value that only need break down is set to.Because for ability value is 0 port, its outline that is selected to the aggregation group outbound port is 0, and effect is equivalent to get rid of from aggregation group.In case fault solves, the port ability value obtains upgrading at once, quick-recovery link bandwidth soon, reenter can selecteed aggregation group outbound port in, possess good failover capability.
The sharing polymerization link circuit flow method of the embodiment of the invention, cause aggregation group member port link overload for fear of burst flow, the port ability value can be set when the port flow load of described aggregation group member port meets or exceeds certain upper threshold, be reduced to 0, as fully loaded 85%, promptly the current outflow occupied bandwidth of a member port reaches 85% of maximum bandwidth that described port supports.Concrete upper threshold value can be determined by the flow of common application on the statistics and analysis link.
After choosing outbound port, wait to E-Packet or message flow is transmitted in described outbound port with described.
Sharing polymerization link circuit flow method of the present invention, described dynamic parameter also can be and connect distance or rent link cost or surf time section, perhaps their combination, the perhaps combination of they and port ability value.
As seen, sharing polymerization link circuit flow method of the present invention, the message flow that the timing renewal of the port ability value that step S100 finishes and step S200 finish or the aggregation group outbound port of message are selected.All can handle, also can handle by the ASIC hardware logic by CPU.Their concrete handling process as shown in Figures 2 and 3.
As shown in Figure 2, the sharing polymerization link circuit flow method of the embodiment of the invention, the flow process that the timing of step S100 middle port ability value is upgraded comprises the steps:
Step S110: when setting up aggregation group, be each member port binding statistical counter;
Step S120: the maximum bandwidth of determining described port according to factors such as the physical speed of member port, duplexing situation and burst bandwidth reservations;
Step S130: regularly carry out software query statistic count value or start hardware DMA uploading the statistical counting value;
Step S140: calculate the current bandwidth occupancy value of each member port;
Step S150: calculate and upgrade the port ability value of each member port, it deducts current bandwidth occupancy value gained result by maximum bandwidth and is converted to.
As shown in Figure 3, the sharing polymerization link circuit flow method of the embodiment of the invention, the flow process that the aggregation group outbound port is selected to transmit with the message of step S300 among the step S200 comprises the steps:
Step S210: receive and wait to E-Packet;
Step S220: search address table and determine to transmit port;
Step S230: judge whether to be sent to aggregation group, the forwarding port of promptly putting down in writing in address table then is sent to aggregation group as being the aggregation group port; If desired, execution in step S240 then; As not needing, execution in step S310 then;
Step S240: share algorithm according to the polymerization that with the port ability value is an input and determine port; Execution in step S320 then;
Step S310: outlet E-Packets according to address table; Flow process finishes;
Step S320: share the determined outbound port of algorithm according to described polymerization and E-Packet; Flow process finishes.
Sharing polymerization link circuit flow method of the present invention, the port ability value of introducing is the parameter of a dynamic change.All carry out an aggregation group outbound port according to each message forwarding and select, the definite message flow of certain bar will arrive the destination by different paths so.For the network of two layers of forwarding, can any influence not arranged to application; In the network of three layers of forwarding, the situation of packet out-ordering will certainly appear.At this situation, the sharing polymerization link circuit flow method of the embodiment of the invention is just carried out the selection of aggregation group outbound port when a new message flow arrives, and writes down the outbound port of selecting based on message flow.If this stream continues to take place, will transmit according to current selection result so always.If outbound port breaks down, then need again basis each port ability value at that time to carry out outbound port and upgrade selection.
Again the aggregated links sharing device of the embodiment of the invention is introduced below.
As shown in Figure 4, realize the sharing polymerization link circuit flow device of the embodiment of the invention of sharing polymerization link circuit flow method of the present invention, comprise parameter acquisition module, port selection module and forwarding module;
Parameter acquisition module: be used to obtain each described member port can this port forwarding performance of dynamic reflection the port ability value; Described port ability value is used for the current available remaining bandwidth of the described port of nominal;
Port is selected module: be connected in described parameter acquisition module, the port ability value of each member port that static parameter value that at least one is described to be E-Packeted and described parameter acquisition module transmit, as the input of sharing polymerization link circuit flow algorithm, move described sharing polymerization link circuit flow algorithm jointly;
Forwarding module: receive the selecteed described outbound port that described port selects module to transmit, with described wait to E-Packet transmit in described outbound port.
The sharing polymerization link circuit flow device of the embodiment of the invention comprises computing unit, traffic statistics unit and parameter conversion unit in the described parameter acquisition module;
Described traffic statistics unit is used to obtain the present flow rate occupied bandwidth of each member port; Described traffic statistics unit by receiving or inquire about the statistical counting value of statistical counter that is used to obtain present flow rate bandwidth occupancy value of each member port binding, obtains the present flow rate bandwidth occupancy value of each member port.
Described computing unit receives the present flow rate occupied bandwidth of described traffic statistics unit input, deducts the present flow rate occupied bandwidth by the maximum bandwidth of described port support and obtains current available remaining bandwidth;
Described converting unit, the described current available remaining bandwidth with described computing unit transmits is converted to the port ability value.
The sharing polymerization link circuit flow device of the embodiment of the invention is selected module to comprise at described port and is amounted to unit, probability calculation unit, algorithm unit and map unit;
Amount to the unit and add up the port ability value of each member port, and calculate the total capacity value of all of the port; The probability calculation unit receives port ability value and the described total capacity value that described total unit transmits each member port, calculates port ability value shared percentage in described total capacity value of each member port; Send result of calculation to algorithm unit; Algorithm unit moves described sharing polymerization link circuit flow algorithm, the described percentage that probability of acceptance computing unit transmits, with described percentage serve as select probability be described wait to E-Packet select port.When algorithm unit carries out the outbound port selection, can be with the percentage input map unit of a static parameter and each port ability value, set up one by the mapping of described static parameter collection to described each member port collection by map unit, the mapped probability of two described member ports equals the ratio of the port ability value of described two member's ports.Described algorithm unit receives the mapping relations of the output of described map unit, waits to E-Packet for each or message flow is selected port according to described mapping relations.
The sharing polymerization link circuit flow device of second embodiment of the invention, parameter acquisition module is identical with the sharing polymerization link circuit flow device of first embodiment with forwarding module, and described port is selected to comprise dispensing unit in the module, amount to unit, probability calculation unit, algorithm unit and map unit:
Dispensing unit disposes the weighting coefficient values of each member port; For example the weight coefficient for the described weighting port ability value of the member port of full duplex is configured to 2, and the weight coefficient of the described weighting port ability value of semiduplex member port is configured to 1.Amount to the weighting coefficient values that the unit receives each member port of described dispensing unit input, calculate the weighting total capacity value of the weighting port ability value and all of the port of each member port; With its input probability computing unit; The probability calculation unit receives port weighting ability value and the described weighting total capacity value that described total unit transmits each member port, calculates port ability value shared percentage in described total capacity value of each member port; Algorithm unit moves described sharing polymerization link circuit flow algorithm, the described percentage that probability of acceptance computing unit transmits, with described percentage serve as select probability be described wait to E-Packet select port.Wherein algorithm unit is identical with the algorithm unit and the map unit of the sharing polymerization link circuit flow device of map unit and first embodiment of the invention.
The sharing polymerization link circuit flow device of the first embodiment of the present invention and second embodiment, also comprise the parameter adjustment unit in the described parameter acquisition module, described parameter adjustment unit receives the port ability value of described converting unit input, and the port ability value of the described member port that breaks down is adjusted into 0; When described fault is eliminated, upgrade described port ability value; The port ability value that described parameter adjustment unit also reaches traffic load the member port of threshold value is adjusted into 0.
The sharing polymerization link circuit flow device of the first embodiment of the present invention and second embodiment, select in the module at described port, can also comprise a stream processing unit, described stream processing unit writes down the outbound port of this message flow, and the forwarding port described to be E-Packeted that belongs to this message flow is chosen as this outbound port.
Above-described only is preferable possible embodiments of the present invention; described embodiment is not in order to limit scope of patent protection of the present invention; therefore the equivalent structure done of every utilization specification of the present invention and accompanying drawing content changes, and in like manner all should be included in protection scope of the present invention.
Claims (15)
1. a sharing polymerization link circuit flow method is used for the traffic sharing between the aggregated links member port, may further comprise the steps:
Step S100: know the dynamic parameter value that at least one of each described member port can this port forwarding performance of dynamic reflection;
Step S200: static parameter value that will be to be E-Packeted and described dynamic parameter value, common input as the sharing polymerization link circuit flow algorithm, move described sharing polymerization link circuit flow algorithm, wait to E-Packet during selective polymerization group outbound port for described, carry out probability according to the transfer capability value of member port and select;
Step S300: with described wait to E-Packet transmit in the described selected outbound port of algorithm of sharing.
2. sharing polymerization link circuit flow method as claimed in claim 1, it is characterized in that, in step S 100, described dynamic parameter value is the port ability value of the current available remaining bandwidth of the described port of nominal, described current available remaining bandwidth can be deducted current outflow occupied bandwidth by the maximum bandwidth of described port support and obtain, and described current outflow occupied bandwidth is obtained as calculated by the count value of the statistical counter that is bundled in each described member port.
3. sharing polymerization link circuit flow method as claimed in claim 2, it is characterized in that, in step S200, in described sharing polymerization link circuit flow algorithm, one described waiting E-Packeted, and the probability that a member port is selected as described outbound port equals the timing updated port ability value of this member port at the shared percentage of the total capacity value of all of the port.
4. sharing polymerization link circuit flow method as claimed in claim 3 is characterized in that, in step S200, and the summation of the port ability value that described total capacity value is weighting.
5. sharing polymerization link circuit flow method as claimed in claim 2 is characterized in that: in step S100, the described port ability value of the described member port that breaks down is set to 0, when described fault is eliminated, upgrades described port ability value; When the traffic load of described member port reached a threshold value, the port ability value of this member port was set to 0.
6. sharing polymerization link circuit flow method as claimed in claim 2, it is characterized in that, obtain the mode of described statistical counter and inquire about the statistical counting value of described statistical counter or start the described statistical counting value that hardware DMA uploads described statistical counter for timing.
7. sharing polymerization link circuit flow method as claimed in claim 1 is characterized in that: in step S200, described the waiting that belongs to a message flow E-Packeted, the outbound port of this message is chosen as the outbound port of this message flow that has write down.
8. sharing polymerization link circuit flow method as claimed in claim 1 is characterized in that: described dynamic parameter is for connecting distance, renting link cost or surf time section.
9. a sharing polymerization link circuit flow device of realizing the described sharing polymerization link circuit flow method of claim 1 is characterized in that, described sharing polymerization link circuit flow device comprises parameter acquisition module, port selection module and forwarding module;
Parameter acquisition module: be used to obtain the dynamic parameter value that at least one of each described member port can this port forwarding performance of dynamic reflection;
Port is selected module: be connected in described parameter acquisition module, the dynamic parameter value that static parameter value described to be E-Packeted and described parameter acquisition module are transmitted, common input as the sharing polymerization link circuit flow algorithm, move described sharing polymerization link circuit flow algorithm, wait to E-Packet during selective polymerization group outbound port for described, carry out probability according to the transfer capability value of member port and select;
Forwarding module: be connected in described port and select module, what receive that described port selects that module transmits shares the outbound port that algorithm is selected by described, with described wait to E-Packet transmit in described outbound port.
10. sharing polymerization link circuit flow device as claimed in claim 9 is characterized in that: described dynamic parameter value is the port ability value that is used for the current available remaining bandwidth of the described port of nominal;
Comprise computing unit, traffic statistics unit and parameter conversion unit in the described parameter acquisition module; Wherein:
Described traffic statistics unit is used to obtain the present flow rate occupied bandwidth of each member port; By receiving or inquire about the statistical counting value of statistical counter that is used to obtain present flow rate bandwidth occupancy value of each member port binding, obtain the present flow rate bandwidth occupancy value of each member port;
Described computing unit receives the present flow rate occupied bandwidth of described traffic statistics unit input, deducts the present flow rate occupied bandwidth by the maximum bandwidth of described port support and obtains current available remaining bandwidth;
Described converting unit with the described current available remaining bandwidth that described computing unit transmits, is converted to the port ability value of the current available remaining bandwidth of the described port of nominal.
11. sharing polymerization link circuit flow device as claimed in claim 10 is characterized in that: select to comprise in the module at described port:
Amount to the unit, be used to add up the port ability value of each member port, and calculate the total capacity value of all of the port;
The probability calculation unit receives the port ability value and the described total capacity value of each member port that transmits described total unit, calculates port ability value shared percentage in described total capacity value of each member port;
Algorithm unit is used to move described sharing polymerization link circuit flow algorithm, the described percentage that probability of acceptance computing unit transmits, with described percentage serve as select probability be described wait to E-Packet select port.
12. sharing polymerization link circuit flow device as claimed in claim 11 is characterized in that: select also to comprise in the module at described port:
Dispensing unit is used to dispose the weighting coefficient values of each member port;
Described total unit receives the weighting coefficient values of each member port of described dispensing unit input, calculates the weighting total capacity value of the weighting port ability value and all of the port of each member port.
13. sharing polymerization link circuit flow device as claimed in claim 11, it is characterized in that, described port selects module also to comprise map unit, described map unit receives the described static parameter of described algorithm unit input, a mapped member port is exported to described algorithm unit, and the mapped probability of two described member ports equals the ratio of the port ability value of described two member's ports.
14. sharing polymerization link circuit flow device as claimed in claim 11, it is characterized in that: select in the module at described port, also comprise a stream processing unit, described stream processing unit writes down the outbound port of this message flow, and the forwarding port described to be E-Packeted that belongs to this message flow is chosen as this outbound port.
15. sharing polymerization link circuit flow device as claimed in claim 12, it is characterized in that: described parameter acquisition module also comprises the parameter adjustment unit, described parameter adjustment unit receives the port ability value of described converting unit input, and the port ability value of the described member port that breaks down is adjusted into 0; When described fault is eliminated, upgrade described port ability value; The port ability value that described parameter adjustment unit also reaches traffic load the member port of threshold value is adjusted into 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100079268A CN101227402B (en) | 2008-02-20 | 2008-02-20 | Method and apparatus for sharing polymerization link circuit flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100079268A CN101227402B (en) | 2008-02-20 | 2008-02-20 | Method and apparatus for sharing polymerization link circuit flow |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101227402A CN101227402A (en) | 2008-07-23 |
| CN101227402B true CN101227402B (en) | 2011-03-23 |
Family
ID=39859163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008100079268A Active CN101227402B (en) | 2008-02-20 | 2008-02-20 | Method and apparatus for sharing polymerization link circuit flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101227402B (en) |
Families Citing this family (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101820394B (en) * | 2010-03-22 | 2015-06-03 | 中兴通讯股份有限公司 | Statistical method and equipment of access flow |
| CN102137018A (en) | 2011-03-21 | 2011-07-27 | 华为技术有限公司 | Load sharing method and device thereof |
| CN102347859B (en) * | 2011-09-26 | 2013-04-10 | 杭州华三通信技术有限公司 | Link rapidly recovering method and device |
| CN102447619B (en) * | 2011-11-10 | 2014-09-17 | 华为技术有限公司 | Load sharing mode selecting method, device and system |
| CN102413053A (en) * | 2011-12-12 | 2012-04-11 | 盛科网络（苏州）有限公司 | Method and device for realizing load balance of trunk port |
| CN103312613B (en) * | 2012-03-16 | 2018-08-10 | 南京中兴新软件有限责任公司 | Service message retransmission method and device |
| CN103023804B (en) * | 2012-12-18 | 2015-08-19 | 北京星网锐捷网络技术有限公司 | A kind of aggregated links adaptive flow load-balancing method, device and the network equipment |
| CN105703928A (en) * | 2014-11-25 | 2016-06-22 | 中兴通讯股份有限公司 | Traffic adjustment method and traffic adjustment device |
| CN104580002B (en) * | 2015-01-14 | 2017-07-28 | 盛科网络（苏州）有限公司 | The big balanced retransmission method of current load and device |
| CN104618254A (en) * | 2015-02-12 | 2015-05-13 | 杭州华三通信技术有限公司 | Method and controller capable of avoiding network congestion in software defined network |
| CN106453111B (en) * | 2015-08-11 | 2020-12-22 | 中兴通讯股份有限公司 | Traffic management method and device based on aggregated link |
| CN105490958B (en) * | 2015-12-09 | 2019-08-16 | 上海斐讯数据通信技术有限公司 | A kind of method and system of LACP load balancing |
| CN105848258A (en) * | 2016-01-27 | 2016-08-10 | 乐卡汽车智能科技（北京）有限公司 | Communication method and apparatus on communication device with at least two communication links |
| CN107566299A (en) * | 2016-06-30 | 2018-01-09 | 中兴通讯股份有限公司 | A kind of dynamic link aggregation implementation method and device |
| CN108574961A (en) * | 2017-03-09 | 2018-09-25 | 中国移动通信有限公司研究院 | A kind of resource allocation methods and device |
| CN109995667A (en) * | 2017-12-29 | 2019-07-09 | 北京华为数字技术有限公司 | The method and sending device of transmitting message |
| CN108600100B (en) * | 2018-07-13 | 2021-09-07 | 新华三技术有限公司合肥分公司 | Member port selection method and device |
| CN109218218B (en) * | 2018-08-22 | 2022-08-05 | 烽火通信科技股份有限公司 | Method and system for sharing data flow according to bandwidth based on user-defined template |
| CN111817906B (en) * | 2019-04-11 | 2021-11-30 | 中国移动通信集团山东有限公司 | Data processing method, device, network equipment and storage medium |
| CN110838986B (en) * | 2019-07-29 | 2022-02-22 | 迈普通信技术股份有限公司 | Method, equipment and system for balancing LACP load |
| CN112953742B (en) * | 2019-12-10 | 2023-05-12 | 中盈优创资讯科技有限公司 | Fault positioning method and device for binding port of network equipment |
| CN112019432B (en) * | 2020-07-31 | 2022-02-01 | 深圳市风云实业有限公司 | Uplink input message forwarding system based on multiport binding |
| CN112311671B (en) * | 2020-11-06 | 2022-03-25 | 杭州迪普科技股份有限公司 | Method, apparatus, medium and device for issuing aggregated link configuration to switch chip |
| CN112737945B (en) * | 2020-12-30 | 2022-09-30 | 杭州迪普信息技术有限公司 | Server connection control method and device |
| CN113098790B (en) * | 2021-03-26 | 2022-06-21 | 新华三信息安全技术有限公司 | Flow forwarding method and device |
| CN113645145B (en) * | 2021-08-02 | 2024-07-19 | 迈普通信技术股份有限公司 | Load balancing method, load balancing device, network equipment and computer readable storage medium |
| CN113839883B (en) * | 2021-09-30 | 2023-05-26 | 杭州迪普科技股份有限公司 | Configuration method of port aggregation group |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1484413A (en) * | 2002-09-18 | 2004-03-24 | ����ͨѶ�ɷ�����˾ | Method for realizing polylink flow equalization |
| CN1787489A (en) * | 2005-12-22 | 2006-06-14 | 港湾网络有限公司 | Method for average distributing interface flow at multi network processor engines |
-
2008
- 2008-02-20 CN CN2008100079268A patent/CN101227402B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1484413A (en) * | 2002-09-18 | 2004-03-24 | ����ͨѶ�ɷ�����˾ | Method for realizing polylink flow equalization |
| CN1787489A (en) * | 2005-12-22 | 2006-06-14 | 港湾网络有限公司 | Method for average distributing interface flow at multi network processor engines |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101227402A (en) | 2008-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101227402B (en) | Method and apparatus for sharing polymerization link circuit flow | |
| US7295519B2 (en) | Method of quality of service based flow control within a distributed switch fabric network | |
| US9185047B2 (en) | Hierarchical profiled scheduling and shaping | |
| US6621791B1 (en) | Traffic management and flow prioritization over multiple physical interfaces on a routed computer network | |
| JP5335892B2 (en) | High-speed virtual channel for packet-switched on-chip interconnect networks | |
| EP1810466B1 (en) | Directional and priority based flow control between nodes | |
| CA2329542C (en) | System and method for scheduling message transmission and processing in a digital data network | |
| US7274660B2 (en) | Method of flow control | |
| CN103476062B (en) | Data flow scheduling method, equipment and system | |
| CN100596114C (en) | Credit based flow control system, apparatus and method | |
| CN104378308A (en) | Method and device for detecting message sending rate | |
| CN102047619B (en) | Methods, systems, and computer readable media for dynamically rate limiting slowpath processing of exception packets | |
| US8989037B2 (en) | System for performing data cut-through | |
| CN101002443A (en) | A method for allocating data to at least one packet in an integrated circuit | |
| CN102111327A (en) | Method and system for cell dispatching | |
| JPH03132127A (en) | Packet communication method and communication terminal device | |
| CN112787952B (en) | Service flow adjusting method and device | |
| CN109995664B (en) | Method, equipment and system for transmitting data stream | |
| CN110995607A (en) | Method, device, network equipment and storage medium for transmitting flow | |
| CN109995608B (en) | Network rate calculation method and device | |
| CN112751776B (en) | Congestion control method and related device | |
| JP5307745B2 (en) | Traffic control system and method, program, and communication relay device | |
| CN110601996B (en) | Looped network anti-starvation flow control method adopting token bottom-preserving distributed greedy algorithm | |
| CN116418745A (en) | Data exchange method and device | |
| CN110191069B (en) | Annular network on chip with multiple channels |
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 |
Address after: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No. Patentee after: Xinhua three Technology Co., Ltd. Address before: 310053 Hangzhou hi tech Industrial Development Zone, Zhejiang province science and Technology Industrial Park, No. 310 and No. six road, HUAWEI, Hangzhou production base Patentee before: Huasan Communication Technology Co., Ltd. |
|
| CP03 | Change of name, title or address |