CN109831388A - Method and apparatus for optimizing flow load balance - Google Patents
Method and apparatus for optimizing flow load balance Download PDFInfo
- Publication number
- CN109831388A CN109831388A CN201711179444.6A CN201711179444A CN109831388A CN 109831388 A CN109831388 A CN 109831388A CN 201711179444 A CN201711179444 A CN 201711179444A CN 109831388 A CN109831388 A CN 109831388A
- Authority
- CN
- China
- Prior art keywords
- token bucket
- physical port
- flow
- data flow
- mean square
- 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.)
- Pending
Links
Abstract
Present disclose provides a kind of for optimizing the method and apparatus of flow load balance.This method comprises: configuring token bucket quantity for each physical port, and data flow is made to correspond to token bucket；According to the token bucket quantity configured for each physical port, the token bucket of respective numbers is distributed for each physical port；The flow velocity of the data flow of each token bucket is flowed through in acquisition, the flow velocity for the data flow for flowing through each physical port is calculated；According to the initial mean square deviation of the flow relocity calculation flow velocity for the data flow for flowing through each physical port；In the case where initial mean square deviation is greater than or equal to threshold value, the mean square deviation for the flow velocity that token bucket is combined with all mappings of physical port is calculated by the method for traversal, and selects the smallest mapping group cooperation of mean square deviation in all mappings combination to combine optimal solution；And the mapping relations of token bucket and physical port are adjusted according to the combination optimal solution, to be optimized to flow load balance.
Description
Technical field
This disclosure relates to which technical field of communication network, in particular to a kind of for optimizing the method and dress of flow load balance
It sets.
Background technique
Existing flow load balance technology is all based on the calculating of Hash (hash) algorithm.Based on calculating source and destination IP
Under the scene that address Hash value is evenly distributed, the effect of flow equalization distribution Kazakhstan that is poor, and being carried out based on data packet
Uncommon calculate will cause biggish pressure to equipment performance.
Fig. 1 is the data stream transmitting schematic diagram for schematically showing flow load balance method in the prior art.Such as Fig. 1
Shown, the data flow after the carry out Hash calculation based on source and destination IP address directly corresponds to physical port.Such as such as Fig. 1 institute
Show, using existing flow load balance technology, the flow for flowing through the data flow of each physical port 1~4 is not completely
Deng, the effect of flow equalization distribution is poor.
Summary of the invention
The technical problem that the embodiment of the present disclosure solves is: providing a kind of method or dress for optimizing flow load balance
It sets.
According to the one aspect of the embodiment of the present disclosure, provide a kind of for optimizing the method for flow load balance, comprising:
Token bucket quantity is configured for each physical port of multiple physical ports, and data flow is made to correspond to token bucket；According to being every
The token bucket quantity of a physical port configuration, the token bucket of respective numbers is distributed for each physical port；Each order is flowed through in acquisition
The flow velocity of the data flow of board bucket, the flow velocity for the data flow for flowing through each physical port is calculated；According to flowing through each physics
The initial mean square deviation of the flow relocity calculation flow velocity of the data flow of port；It is greater than or equal to the feelings of threshold value in the initial mean square deviation
Under condition, the mean square deviation for the flow velocity that token bucket is combined with all mappings of physical port is calculated by the method for traversal, and
The smallest mapping group cooperation of mean square deviation is selected in all mappings combination to combine optimal solution；And it is optimal according to the combination
The mapping relations of solution adjustment token bucket and physical port, to be optimized to flow load balance.
Optionally, the step of flow velocity for the data flow that each token bucket is flowed through in acquisition includes: setting delayer, thus every
Acquisition is sampled every flow velocity of the predetermined time to the data flow for flowing through each token bucket.
Optionally, the step of distributing the token bucket of respective numbers for each physical port includes: according to the initial of token bucket
It puts in order and distributes the token bucket of respective numbers for each physical port.
Optionally, the step of configuring token bucket quantity for each physical port of multiple physical ports includes: according to each
The bandwidth of link corresponding to physical port is that each physical port configures token bucket quantity.
Optionally, the method also includes: the initial mean square deviation be less than the threshold value in the case where, resurvey
Flow through the flow velocity of the data flow of each token bucket.
Optionally, the data flow is the data flow after hash algorithm calculates.
According to the other side of the embodiment of the present disclosure, a kind of for optimizing the device of flow load balance, packet is provided
Include: configuration unit for configuring token bucket quantity for each physical port of multiple physical ports, and corresponds to data flow
Token bucket distributes the token bucket of respective numbers for each physical port according to the token bucket quantity configured for each physical port；
Acquisition unit, for acquiring the flow velocity for flowing through the data flow of each token bucket；Computing unit flows through each token bucket for basis
The flow relocity calculation of data flow obtain flowing through the flow velocity of the data flow of each physical port, and according to flowing through each physical port
The initial mean square deviation of the flow relocity calculation flow velocity of data flow, and the case where the initial mean square deviation is greater than or equal to threshold value
Under, it is calculated the mean square deviation for the flow velocity that token bucket is combined with all mappings of physical port by the method for traversal, and
The smallest mapping group cooperation of mean square deviation is selected in all mapping combinations to combine optimal solution；And adjustment unit, it is used for basis
The mapping relations of optimal solution adjustment token bucket and physical port are combined, to optimize to flow load balance.
Optionally, the acquisition unit is set with delayer, thus at interval of the predetermined time to flowing through each token bucket
The flow velocity of data flow is sampled acquisition.
Optionally, the configuration unit is that each physical port distributes respective numbers according to the initial arrangement sequence of token bucket
Token bucket.
Optionally, the bandwidth of configuration unit link according to corresponding to each physical port is each physical port
Configure token bucket quantity.
Optionally, the acquisition unit resurveys stream in the case where the initial mean square deviation is less than the threshold value
Cross the flow velocity of the data flow of each token bucket.
Optionally, the data flow is the data flow after hash algorithm calculates.
According to the other side of the embodiment of the present disclosure, a kind of for optimizing the device of flow load balance, packet is provided
It includes: memory；And it is coupled to the processor of the memory, the processor is configured to based on the memory is stored in
The foregoing method of instruction execution.
According to the other side of the embodiment of the present disclosure, a kind of computer readable storage medium is provided, is stored thereon with
The step of computer program instructions, which realizes foregoing method when being executed by processor.
In above-described embodiment of the disclosure, token bucket quantity is configured for each physical port, and makes data flow corresponding
To token bucket；According to the token bucket quantity configured for each physical port, the token of respective numbers is distributed for each physical port
Bucket；The flow velocity of the data flow of each token bucket is flowed through in acquisition, the flow velocity for the data flow for flowing through each physical port is calculated；
According to the initial mean square deviation of the flow relocity calculation flow velocity for the data flow for flowing through each physical port；Be greater than in initial mean square deviation or
In the case where equal to threshold value, the flow velocity that token bucket is combined with all mappings of physical port is calculated by the method for traversal
Mean square deviation, and the smallest mapping group cooperation of mean square deviation is selected in all mappings combination to combine optimal solution；And according to
The combination optimal solution adjusts the mapping relations of token bucket and physical port, to optimize to flow load balance.The disclosure is real
The method or apparatus for applying example can improve the isotropism of flow load balance under the premise of not influencing equipment performance.
By the detailed description referring to the drawings to the exemplary embodiment of the disclosure, the other feature of the disclosure and its
Advantage will become apparent.
Detailed description of the invention
The attached drawing for constituting part of specification describes embodiment of the disclosure, and together with the description for solving
Release the principle of the disclosure.
The disclosure can be more clearly understood according to following detailed description referring to attached drawing, in which:
Fig. 1 is the data stream transmitting schematic diagram for schematically showing flow load balance method in the prior art.
Fig. 2 is the flow chart for showing the method for optimizing flow load balance according to some embodiments of the disclosure.
Fig. 3 is the number for schematically showing the method for optimizing flow load balance according to some embodiments of the disclosure
According to spreading defeated schematic diagram.
Fig. 4 is the knot for schematically showing the device for optimizing flow load balance according to some embodiments of the disclosure
Composition.
Fig. 5 is the device for being used to optimize flow load balance schematically shown according to other embodiments of the disclosure
Structure chart.
Fig. 6 is the device for being used to optimize flow load balance schematically shown according to other embodiments of the disclosure
Structure chart.
Specific embodiment
The various exemplary embodiments of the disclosure are described in detail now with reference to attached drawing.It should also be noted that unless in addition having
Body explanation, the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
Scope of disclosure.
Simultaneously, it should be appreciated that for ease of description, the size of various pieces shown in attached drawing is not according to reality
Proportionate relationship draw.
Be to the description only actually of at least one exemplary embodiment below it is illustrative, never as to the disclosure
And its application or any restrictions used.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as part of specification.
It is shown here and discuss all examples in, any occurrence should be construed as merely illustratively, without
It is as limitation.Therefore, the other examples of exemplary embodiment can have different values.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need that it is further discussed.
Fig. 2 is the flow chart for showing the method for optimizing flow load balance according to some embodiments of the disclosure.Fig. 3
It is to schematically show to be shown according to the data stream transmitting of the method for optimizing flow load balance of some embodiments of the disclosure
It is intended to.Below with reference to Fig. 2 and Fig. 3 detailed description according to some embodiments of the disclosure for optimizing the side of flow load balance
Method.
As shown in Fig. 2, being that each physical port of multiple physical ports configures token bucket quantity, and makes in step S202
It obtains data flow and corresponds to token bucket.
Such as shown in figure 3, token bucket quantity is respectively configured for each physical port 1~4, for example, each physical port
Weight 4 is configured, i.e., distributes 4 token buckets respectively.Data flow passes through token bucket respectively and is transferred to corresponding physical port.For example,
Here data flow is the data flow after hash algorithm calculates.In some embodiments, data flow can be with token bucket one
One is corresponding.I.e. every corresponding token bucket of group data stream, the combination that each token bucket can permit a group data stream are flowed through.
In embodiment of the disclosure, the step of configuring token bucket quantity for each physical port of multiple physical ports can
With include: can the bandwidth of the link according to corresponding to each physical port be that each physical port configures token bucket quantity.
It in some embodiments, can be each physics if the bandwidth of link corresponding to each physical port is equal
Port configures equal token bucket quantity (i.e. the identical token bucket quantity of constant speed link configuration).Such as two physical ports are right respectively
The bandwidth for the both links answered is 10G, in order to enable load balancing, both links can configure the token bucket of identical quantity.
Certainly, unequal token bucket quantity can also be configured in order to realize certain special purposes for the link of constant speed.Such as
In order to realize the purpose for reducing utilization rate on certain corresponding road Tiao Lian of port, less token can be configured on the link
Bucket.
It in further embodiments, can be each if the bandwidth of link corresponding to each physical port is unequal
Physical port configures unequal token bucket quantity.I.e. if being non-constant speed link, such as the corresponding link of a physical port
Bandwidth is 10G, and the corresponding link bandwidth of another physical port is 40G, then the token bucket quantity configured on 40G chain road can
To be 4 times of the token bucket quantity configured on 10G chain road.
Fig. 2 is returned to, in step S204, according to the token bucket quantity configured for each physical port, for each physical port
Distribute the token bucket of respective numbers.The token bucket of respective numbers is mapped to corresponding physical port, this makes each token
Bucket forms mapping relations with corresponding physical port.
In some embodiments, the step of distributing the token bucket of respective numbers for each physical port may include: can be with
Initial arrangement sequence according to token bucket is the token bucket of each physical port distribution respective numbers.I.e. according to the initial of token bucket
It puts in order and configures the pointer of physical port.For example, token bucket 1~4 can be distributed to port 1, token bucket 5~8 is distributed
To port 2, token bucket 9~12 is distributed into port 3, token bucket 13~16 is distributed into port 4.According to the initial row of token bucket
Column sequence is each physical port default configuration pointer, and the configuration in initialization can be made convenient.
It in further embodiments, can not also be that each physical port distributes phase according to the initial arrangement sequence of token bucket
Answer the token bucket of quantity.For example, token bucket 1,2,5,6 can be distributed to port 1, token bucket 3,4,7,8 distributes to port 2,
Token bucket 9,10,13,14 distributes to port 3, and token bucket 11,12,15,16 distributes to port 4.
As shown in Fig. 2, the flow velocity of the data flow of each token bucket is flowed through in acquisition in step S206, flowed through with being calculated
The flow velocity of the data flow of each physical port.
In some embodiments, the step of flow velocity for the data flow that each token bucket is flowed through in acquisition may include: that setting is prolonged
When device, to be sampled acquisition at interval of flow velocity of the predetermined time to the data flow for flowing through each token bucket.Then basis is adopted
The flow relocity calculation of the data flow for flowing through each token bucket of collection obtains flowing through the flow velocity of the data flow of each physical port.For example,
For some physical port, will be added with the flow velocity of the data flow of all token buckets of the Physical ports map, it is obtained and
Exactly flow through the flow velocity of the data flow of the physical port.
As shown in Fig. 2, in step S208, according to the initial of the flow relocity calculation flow velocity for the data flow for flowing through each physical port
Mean square deviation.I.e. after the flow velocity that the data flow for flowing through each physical port is calculated, according to known mean square deviation formula meter
The mean square deviation for calculating flow velocity, is denoted as initial mean square deviation.
As shown in Fig. 2, judging whether initial mean square deviation is greater than or equal to threshold value in step S210.If it is, process
S212 is entered step, otherwise return step S206.Here, threshold value can be set by the user according to the actual situation.
In this embodiment, judge whether the flow velocity of port differs greatly using mean square deviation algorithm.I.e. if be calculated
The initial mean square deviation of flow velocity be greater than or equal to threshold value, then show that the flow velocity of each physical port differs greatly, i.e. flow is negative
The effect for carrying equilibrium is poor, then needs to optimize by subsequent step the effect of flow load balance., whereas if flow velocity is initial
Mean square deviation is less than threshold value, then shows that the flow velocity difference of each physical port is smaller, i.e. the effect of flow load balance is preferable, then
It may return to step S206, i.e., in the case where initial mean square deviation is less than threshold value, resurvey and flow through each token bucket
The flow velocity of data flow.
As shown in Fig. 2, all mappings of token bucket and physical port are calculated by the method for traversal in step S212
The mean square deviation of combined flow velocity, and select the smallest mapping group cooperation of mean square deviation optimal for combination in all mappings combination
Solution.
In the embodiment the step of, token bucket is calculated by the method for traversal and is combined with all mappings of physical port
In each map the mean square deviation of combined flow velocity, then in all mappings are combined, the selection the smallest mapping of mean square deviation
Combination is as combination optimal solution, to reconfigure token bucket.
As shown in Fig. 2, the mapping relations of token bucket and physical port are adjusted according to combination optimal solution in step S214, with
Flow load balance is optimized.
In some embodiments, according to the pointer of the corresponding physical port of combination optimal solution adjustment token bucket.For example, as schemed
Shown in 3, token bucket 1,3,5,10 is distributed into physical port 1, token bucket 2,6,8,15 is distributed into physical port 2, by token
Bucket 4,11,13,16 distributes to physical port 3, and token bucket 7,9,12,14 is distributed to physical port 4.
In the above method of the embodiment of the present disclosure, token bucket quantity is configured for each physical port, and make data flow
Correspond to token bucket；According to the token bucket quantity configured for each physical port, respective numbers are distributed for each physical port
Token bucket；The flow velocity of the data flow of each token bucket is flowed through in acquisition, the data flow for flowing through each physical port is calculated
Flow velocity；According to the initial mean square deviation of the flow relocity calculation flow velocity for the data flow for flowing through each physical port；In initial mean square deviation
In the case where more than or equal to threshold value, it is calculated what token bucket was combined with all mappings of physical port by the method for traversal
The mean square deviation of flow velocity, and the smallest mapping group cooperation of mean square deviation is selected in all mappings combination to combine optimal solution；With
And the mapping relations of token bucket and physical port are adjusted according to the combination optimal solution, to be optimized to flow load balance.This
The method of open embodiment can improve the isotropism of flow load balance under the premise of not influencing equipment performance.
In some embodiments of the present disclosure, after the mapping relations of adjustment token bucket and physical port, by predetermined
The interval of time acquires the flow velocity for flowing through the data flow of each token bucket again, to carry out subsequent flow load balance again
Optimization processing.
Fig. 4 is the knot for schematically showing the device for optimizing flow load balance according to some embodiments of the disclosure
Composition.As shown in figure 4, the apparatus may include: configuration unit 402, acquisition unit 404, computing unit 406 and adjustment unit
408。
The configuration unit 402 can be used for configuring token bucket quantity for each physical port of multiple physical ports, and make
It obtains data flow and corresponds to token bucket, according to the token bucket quantity configured for each physical port, distribute phase for each physical port
Answer the token bucket of quantity.For example, the data flow can be the data flow after hash algorithm calculates.
The acquisition unit 404 can be used for acquiring the flow velocity for the data flow for flowing through each token bucket.
The computing unit 406 can be used for obtaining flowing through according to the flow relocity calculation for the data flow for flowing through each token bucket each
The flow velocity of the data flow of physical port, and according to the initial square of the flow relocity calculation flow velocity for the data flow for flowing through each physical port
Difference, and in the case where the initial mean square deviation is greater than or equal to threshold value, token bucket is calculated by the method for traversal
The mean square deviation of the flow velocity combined with all mappings of physical port, and selection mean square deviation is the smallest in all mappings combination
Mapping group cooperation is combination optimal solution.
The adjustment unit 408 can be used for the mapping relations according to combination optimal solution adjustment token bucket and physical port, with
Flow load balance is optimized.
In the above-mentioned apparatus of the embodiment of the present disclosure, configuration unit is that each physical port configures token bucket quantity, and makes
It obtains data flow and corresponds to token bucket, according to the token bucket quantity configured for each physical port, distribute phase for each physical port
Answer the token bucket of quantity；The flow velocity of the data flow of each token bucket is flowed through in acquisition unit acquisition；Computing unit is each according to flowing through
The flow relocity calculation of the data flow of token bucket obtains flowing through the flow velocity of the data flow of each physical port, according to flowing through each physics end
The initial mean square deviation of the flow relocity calculation flow velocity of the data flow of mouth, in the case where initial mean square deviation is greater than or equal to threshold value,
The mean square deviation for the flow velocity that token bucket is combined with all mappings of physical port is calculated by the method for traversal, and all
The smallest mapping group cooperation of mean square deviation is selected in mapping combination to combine optimal solution；Adjustment unit is according to the combination optimal demodulation
The mapping relations of whole token bucket and physical port, to be optimized to flow load balance.The device of the embodiment of the present disclosure can be with
Under the premise of not influencing equipment performance, the isotropism of flow load balance is improved.
In some embodiments, configuration unit 402 can be each physical port according to the initial arrangement sequence of token bucket
Distribute the token bucket of respective numbers.
In some embodiments, configuration unit 402 can the bandwidth of the link according to corresponding to each physical port be each
Physical port configures token bucket quantity.
In some embodiments, acquisition unit 404 can be set with delayer, thus every to flowing through at interval of the predetermined time
The flow velocity of the data flow of a token bucket is sampled acquisition.
In some embodiments, acquisition unit 404 can be resurveyed in the case where initial mean square deviation is less than threshold value
Flow through the flow velocity of the data flow of each token bucket.
In the method or apparatus of the embodiment of the present disclosure, the data flow after Hash corresponds to token bucket, and each token bucket reflects
It is mapped to corresponding physical port, periodically acquires the flow information of token bucket, is calculated by mean square deviation algorithm and traversal method
It obtains token bucket combination optimal solution and realizes overall load to periodically adjust the mapping relations of token bucket and physical port
Balanced optimization.
Fig. 5 is the device for being used to optimize flow load balance schematically shown according to other embodiments of the disclosure
Structure chart.The device includes memory 510 and processor 520.Wherein:
Memory 510 can be disk, flash memory or other any non-volatile memory mediums.Memory is for storing Fig. 2
Instruction in corresponding embodiment.
Processor 520 is coupled to memory 510, can be used as one or more integrated circuits to implement, such as micro process
Device or microcontroller.The processor 520 is for executing the instruction stored in memory, by the way that data flow is corresponded to token bucket,
Each token bucket is mapped to corresponding physical port, acquires the flow information of token bucket, passes through mean square deviation algorithm and traversal method
Token bucket combination optimal solution is calculated, to adjust the mapping relations of token bucket and physical port, realizes overall load equilibrium
Optimization.
It in one embodiment, can be as shown in fig. 6, the device 600 includes memory 610 and processor 620.Processing
Device 620 is coupled to memory 610 by BUS bus 630.The device 600 can also be connected to outside by memory interface 640 and deposit
Storage device 650 can also be connected to network or an other department of computer science to call external data by network interface 660
System (not shown), no longer describes in detail herein.
In this embodiment, it is instructed by memory stores data, then above-metioned instruction is handled by processor, by that will count
Token bucket is corresponded to according to stream, each token bucket is mapped to corresponding physical port, the flow information of token bucket is acquired, by square
Token bucket combination optimal solution is calculated in difference algorithm and traversal method, real to adjust the mapping relations of token bucket and physical port
The optimization of existing overall load equilibrium.
In another embodiment, the disclosure additionally provides a kind of computer readable storage medium, is stored thereon with calculating
The step of machine program instruction, which realizes the method in embodiment corresponding to Fig. 2 when being executed by processor.Skill in the art
Art personnel are it should be appreciated that embodiment of the disclosure can provide as method, apparatus or computer program product.Therefore, the disclosure can adopt
With the form of complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects.Moreover, this public affairs
Opening, which can be used, can use non-transient storage medium in the computer that one or more wherein includes computer usable program code
The form for the computer program product implemented in (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.).
The disclosure is reference according to the method for the embodiment of the present disclosure, the flow chart of equipment (system) and computer program product
And/or block diagram describes.It should be understood that each process in flowchart and/or the block diagram can be realized by computer program instructions
And/or the combination of the process and/or box in box and flowchart and/or the block diagram.It can provide these computer programs to refer to
Enable the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to generate
One machine so that by the instruction that the processor of computer or other programmable data processing devices executes generate for realizing
The device for the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
So far, the disclosure is described in detail.In order to avoid covering the design of the disclosure, it is public that this field institute is not described
The some details known.Those skilled in the art as described above, completely it can be appreciated how implementing technology disclosed herein
Scheme.
Disclosed method and system may be achieved in many ways.For example, can by software, hardware, firmware or
Software, hardware, firmware any combination realize disclosed method and system.The said sequence of the step of for the method
Merely to be illustrated, the step of disclosed method, is not limited to sequence described in detail above, special unless otherwise
It does not mentionlet alone bright.
Although being described in detail by some specific embodiments of the example to the disclosure, the skill of this field
Art personnel it should be understood that above example merely to be illustrated, rather than in order to limit the scope of the present disclosure.The skill of this field
Art personnel are it should be understood that can modify to above embodiments in the case where not departing from the scope of the present disclosure and spirit.This public affairs
The range opened is defined by the following claims.
Claims (14)
1. a kind of for optimizing the method for flow load balance, comprising:
Token bucket quantity is configured for each physical port of multiple physical ports, and data flow is made to correspond to token bucket；
According to the token bucket quantity configured for each physical port, the token bucket of respective numbers is distributed for each physical port；
The flow velocity of the data flow of each token bucket is flowed through in acquisition, the stream for the data flow for flowing through each physical port is calculated
Speed；
According to the initial mean square deviation of the flow relocity calculation flow velocity for the data flow for flowing through each physical port；
In the case where the initial mean square deviation is greater than or equal to threshold value, token bucket and object are calculated by the method for traversal
The mean square deviation of the flow velocity of all mappings combination of port is managed, and selects the smallest mapping of mean square deviation in all mappings combination
Combination is as combination optimal solution；And
It is excellent to be carried out to flow load balance according to the mapping relations of combination optimal solution the adjustment token bucket and physical port
Change.
2. according to the method described in claim 1, wherein, the step of acquiring the flow velocity for the data flow for flowing through each token bucket packet
It includes:
Delayer is set, to be sampled acquisition at interval of flow velocity of the predetermined time to the data flow for flowing through each token bucket.
3. according to the method described in claim 1, wherein, the step of distributing the token bucket of respective numbers for each physical port, is wrapped
It includes:
Initial arrangement sequence according to token bucket is the token bucket of each physical port distribution respective numbers.
4. according to the method described in claim 1, being that each physical port of multiple physical ports configures token bucket quantity wherein
The step of include:
It is that each physical port configures token bucket quantity according to the bandwidth of link corresponding to each physical port.
5. according to the method described in claim 1, further include:
In the case where the initial mean square deviation is less than the threshold value, the stream for flowing through the data flow of each token bucket is resurveyed
Speed.
6. according to the method described in claim 1, wherein,
The data flow is the data flow after hash algorithm calculates.
7. a kind of for optimizing the device of flow load balance, comprising:
Configuration unit for configuring token bucket quantity for each physical port of multiple physical ports, and makes data flow corresponding
The token of respective numbers is distributed for each physical port according to the token bucket quantity configured for each physical port to token bucket
Bucket；
Acquisition unit, for acquiring the flow velocity for flowing through the data flow of each token bucket；
Computing unit, for obtaining flowing through the number of each physical port according to the flow relocity calculation for the data flow for flowing through each token bucket
According to the flow velocity of stream, and the initial mean square deviation of the flow relocity calculation flow velocity according to the data flow for flowing through each physical port, Yi Ji
In the case that the initial mean square deviation is greater than or equal to threshold value, token bucket and physical port are calculated by the method for traversal
All mappings combination flow velocity mean square deviation, and all mappings combination in selection the smallest mapping group cooperation of mean square deviation
To combine optimal solution；And
Adjustment unit, for the mapping relations according to combination optimal solution adjustment token bucket and physical port, with equal to flow load
Weighing apparatus optimizes.
8. device according to claim 7, wherein
The acquisition unit is set with delayer, thus the flow velocity at interval of the predetermined time to the data flow for flowing through each token bucket
It is sampled acquisition.
9. device according to claim 7, wherein
The configuration unit is according to the token bucket that the initial arrangement sequence of token bucket is that each physical port distributes respective numbers.
10. device according to claim 7, wherein
The bandwidth of configuration unit link according to corresponding to each physical port is that each physical port configures token bucket
Quantity.
11. device according to claim 7, wherein
The acquisition unit resurveys in the case where the initial mean square deviation is less than the threshold value and flows through each token bucket
Data flow flow velocity.
12. device according to claim 7, wherein
The data flow is the data flow after hash algorithm calculates.
13. a kind of for optimizing the device of flow load balance, comprising:
Memory；And
It is coupled to the processor of the memory, the processor is configured to based on the instruction execution for being stored in the memory
Method as described in claim 1 to 6 any one.
14. a kind of computer readable storage medium, is stored thereon with computer program instructions, real when which is executed by processor
Now the step of method as described in claim 1 to 6 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711179444.6A CN109831388A (en) | 2017-11-23 | 2017-11-23 | Method and apparatus for optimizing flow load balance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711179444.6A CN109831388A (en) | 2017-11-23 | 2017-11-23 | Method and apparatus for optimizing flow load balance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109831388A true CN109831388A (en) | 2019-05-31 |
Family
ID=66858357
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711179444.6A Pending CN109831388A (en) | 2017-11-23 | 2017-11-23 | Method and apparatus for optimizing flow load balance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109831388A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111030938A (en) * | 2019-12-20 | 2020-04-17 | 锐捷网络股份有限公司 | Network equipment load balancing method and device based on CLOS framework |
| CN113452620A (en) * | 2021-06-22 | 2021-09-28 | 新华三信息安全技术有限公司 | Load sharing method and device |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030028514A1 (en) * | 2001-06-05 | 2003-02-06 | Lord Stephen Philip | Extended attribute caching in clustered filesystem |
| CN1622532A (en) * | 2003-11-25 | 2005-06-01 | 华为技术有限公司 | A dynamic equilibrium distributing method for port data flow |
| CN101051939A (en) * | 2006-06-15 | 2007-10-10 | 华为技术有限公司 | Method and device for realizing network flow load sharing |
| CN101494611A (en) * | 2009-02-20 | 2009-07-29 | 华为技术有限公司 | Method and apparatus for regulating load share |
| CN102065022A (en) * | 2011-02-17 | 2011-05-18 | 福建星网锐捷网络有限公司 | Method and device for flow balance of aggregation port, aggregation port and network equipment |
| CN102904826A (en) * | 2012-09-28 | 2013-01-30 | 华为技术有限公司 | Method and device for sharing traffic load |
| CN103259743A (en) * | 2012-02-15 | 2013-08-21 | 中兴通讯股份有限公司 | Method and device for controlling output flow based on token bucket |
| US9239719B1 (en) * | 2013-01-23 | 2016-01-19 | Amazon Technologies, Inc. | Task management system |
| CN105721577A (en) * | 2016-02-15 | 2016-06-29 | 安徽大学 | Server load balancing method for software defined network |
| US20160366250A1 (en) * | 2015-06-11 | 2016-12-15 | Comcast Cable Communications, Llc | Application Peering |
| CN106411733A (en) * | 2016-09-08 | 2017-02-15 | 重庆邮电大学 | SDN dynamic load balanced scheduling method based on real-time load of link |
| CN106656812A (en) * | 2016-12-13 | 2017-05-10 | 北京锐安科技有限公司 | Mapping table constructing method, shunting method of shunting device and corresponding devices |
| CN107294868A (en) * | 2017-07-10 | 2017-10-24 | 西安电子科技大学 | The load-balancing method of polymorphic type egress gateways |
-
2017
- 2017-11-23 CN CN201711179444.6A patent/CN109831388A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030028514A1 (en) * | 2001-06-05 | 2003-02-06 | Lord Stephen Philip | Extended attribute caching in clustered filesystem |
| CN1622532A (en) * | 2003-11-25 | 2005-06-01 | 华为技术有限公司 | A dynamic equilibrium distributing method for port data flow |
| CN101051939A (en) * | 2006-06-15 | 2007-10-10 | 华为技术有限公司 | Method and device for realizing network flow load sharing |
| CN101494611A (en) * | 2009-02-20 | 2009-07-29 | 华为技术有限公司 | Method and apparatus for regulating load share |
| CN102065022A (en) * | 2011-02-17 | 2011-05-18 | 福建星网锐捷网络有限公司 | Method and device for flow balance of aggregation port, aggregation port and network equipment |
| CN103259743A (en) * | 2012-02-15 | 2013-08-21 | 中兴通讯股份有限公司 | Method and device for controlling output flow based on token bucket |
| CN102904826A (en) * | 2012-09-28 | 2013-01-30 | 华为技术有限公司 | Method and device for sharing traffic load |
| US9239719B1 (en) * | 2013-01-23 | 2016-01-19 | Amazon Technologies, Inc. | Task management system |
| US20160366250A1 (en) * | 2015-06-11 | 2016-12-15 | Comcast Cable Communications, Llc | Application Peering |
| CN105721577A (en) * | 2016-02-15 | 2016-06-29 | 安徽大学 | Server load balancing method for software defined network |
| CN106411733A (en) * | 2016-09-08 | 2017-02-15 | 重庆邮电大学 | SDN dynamic load balanced scheduling method based on real-time load of link |
| CN106656812A (en) * | 2016-12-13 | 2017-05-10 | 北京锐安科技有限公司 | Mapping table constructing method, shunting method of shunting device and corresponding devices |
| CN107294868A (en) * | 2017-07-10 | 2017-10-24 | 西安电子科技大学 | The load-balancing method of polymorphic type egress gateways |
Non-Patent Citations (1)
| Title |
|---|
| 柳佳佳: "基于策略的转发系统的设计与实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111030938A (en) * | 2019-12-20 | 2020-04-17 | 锐捷网络股份有限公司 | Network equipment load balancing method and device based on CLOS framework |
| CN113452620A (en) * | 2021-06-22 | 2021-09-28 | 新华三信息安全技术有限公司 | Load sharing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Leventhal et al. | A column generation algorithm for optimal traffic assignment | |
| CN104954253B (en) | For the mainframe network accelerator (HNA) based on PCIe of data center's overlay network | |
| CN104954247B (en) | Mainframe network accelerator for data center's overlay network | |
| CN103873377B (en) | Network Status Mapping | |
| CN104428752B (en) | Virtual machine stream is offloaded to physical queue | |
| Kuo et al. | Optimal approximation algorithm of virtual machine placement for data latency minimization in cloud systems | |
| WO2015192627A1 (en) | Service scheduling method, apparatus, and system | |
| CN110392999A (en) | Virtual filter platform in distributed computing system | |
| CN107729147A (en) | Data processing method, control node and stream calculation system in stream calculation system | |
| CN104954252B (en) | Flow control in high-performance, expansible and data center's switching fabric without call drop | |
| CN105740199B (en) | Time sequence power estimation device and method of network on chip | |
| CN105634974B (en) | Route determining methods and device in software defined network | |
| CN105282043A (en) | Global network load balancing system, device and method | |
| US20160065449A1 (en) | Bandwidth-Weighted Equal Cost Multi-Path Routing | |
| CN109831388A (en) | Method and apparatus for optimizing flow load balance | |
| CN106031094A (en) | Accurate measurement of distributed counters | |
| CN103036792A (en) | Transmitting and scheduling method for maximizing minimal equity multiple data streams | |
| CN110214436A (en) | A kind of multicore is without lock rate limitation devices and method | |
| CN101527685B (en) | Method for assigning message transmission link and device | |
| KR101841026B1 (en) | Service function chaining network system for path optimization | |
| Ke et al. | Aggregation on the fly: Reducing traffic for big data in the cloud | |
| US9749219B2 (en) | Method of optimizing routing in a cluster comprising static communication links and computer program implementing that method | |
| CN108259221A (en) | Flow control methods, device, system, storage medium and processor | |
| CN105511956B (en) | A kind of method for scheduling task and system based on shared scheduling information | |
| JP5742981B1 (en) | Route calculation apparatus, method, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190531 |
|
| RJ01 | Rejection of invention patent application after publication |