CN112600762A - Dynamic flow distribution method for accelerating EF service forwarding - Google Patents
Dynamic flow distribution method for accelerating EF service forwarding Download PDFInfo
- Publication number
- CN112600762A CN112600762A CN202011424766.4A CN202011424766A CN112600762A CN 112600762 A CN112600762 A CN 112600762A CN 202011424766 A CN202011424766 A CN 202011424766A CN 112600762 A CN112600762 A CN 112600762A
- Authority
- CN
- China
- Prior art keywords
- traffic
- flow
- weighted average
- discarded
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2425—Traffic characterised by specific attributes, e.g. priority or QoS for supporting services specification, e.g. SLA
- H04L47/2433—Allocation of priorities to traffic types
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/32—Flow control; Congestion control by discarding or delaying data units, e.g. packets or frames
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/76—Admission control; Resource allocation using dynamic resource allocation, e.g. in-call renegotiation requested by the user or requested by the network in response to changing network conditions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/80—Actions related to the user profile or the type of traffic
- H04L47/805—QOS or priority aware
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a dynamic flow distribution method for accelerating the forwarding of EF services, which comprises the following steps: setting a cycle time on a network node for realizing DiffServ; setting a weighted average service flow parameter; and repeatedly acquiring service flow information in the period time according to the period, and calculating, setting and starting the basic bandwidth reserved for the AF service and the EF service limit bandwidth by adopting a dynamic flow algorithm. The invention adopts a new dynamic flow distribution method to dynamically adjust the flow limiting rate by acquiring the EF and AF flow information in the period time, thereby solving the problems that the flow limitation of the EF service depends on the experience of a network administrator to carry out configuration, the quality of the EF service is influenced, and the bandwidth of a network node is wasted.
Description
Technical Field
The invention relates to the technical field of computer network service quality, in particular to a dynamic flow distribution method for accelerating the forwarding of EF (electronic flow) services.
Background
In the service quality of the IP protocol, a priority Forwarding service with low delay, low jitter, low packet loss rate, and guaranteed bandwidth is defined as an EF (Expedited Forwarding) service, which mainly carries voice and video and is sensitive to jitter and delay.
In the header of IPv4 packet, DSCP (Differentiated Services CodePoint) is the first 6 bits of the ToS (Type of Service) field, and DSCP is used in Differentiated Services (DiffServ) -based network to support QoS (Quality of Service) Service, and its role is to select PHB (Per-Hop Behavior), which guarantees QoS of IPv4 message.
When DSCP is (101110)2When the packet is forwarded, the IPv4 packet is shown to carry EF service, in a differentiated services (DiffServ) -based network, a network node (DS node) for realizing DiffServ is used for realizing a PHB strategy, the EF service is forwarded preferentially, and the PHB has the characteristics of single-hop and independence, only acts on the node, and does not influence other nodes and domains in the network.
To ensure that the EF service does not have an excessive impact on other services, the PHB will limit the flow of the EF service, and the flow exceeding the limit will be discarded. In RFC3246, it is specified that the flow limit of EF traffic is set by the network administrator, i.e. the configuration is done directly depending on the experience of the network administrator. If there is no or little flow of AF (Assured Forwarding) traffic in a network node implementing DiffServ, when the flow of EF traffic suddenly exceeds the flow limit rate, even if the bandwidth of the network node is still sufficient, the flow exceeding the limit in the EF traffic is still discarded, and this policy affects the quality of the EF traffic and wastes the bandwidth of the network node.
It can be seen that, in the prior art, the flow limitation of the EF service depends on the experience of the network administrator for configuration, which affects the quality of the EF service and wastes the bandwidth of the network node.
The above drawbacks are expected to be overcome by those skilled in the art.
Disclosure of Invention
Technical problem to be solved
In order to solve the above problems in the prior art, the present invention provides a dynamic traffic allocation method for accelerating forwarding of an EF service, which dynamically adjusts a traffic limitation rate, and solves the problems that the traffic limitation of the EF service depends on the experience of a network administrator for configuration, the quality of the EF service is affected, and the bandwidth of a network node is wasted.
(II) technical scheme
In order to achieve the purpose, the invention adopts the main technical scheme that:
an embodiment of the present invention provides a dynamic flow allocation method for accelerating forwarding of an EF service, including the following steps:
step S100, setting unit time delta t on a DS node;
step S200, two parameters Z are set on the DS nodeEF(t) and ZAF(t),ZEF(t) is the weighted average EF flow at time t, ZAF(t) is the weighted average AF flow at time t;
step S300, two parameters D are set on the DS nodeEF(t) and DAF(t),DEF(t) EF traffic discarded as a weighted average at time t, DAF(t) weighted average discarded AF traffic at time t;
step S400, obtaining EF flow sample R at the moment t on the DS node in unit time delta tEF(t) and AF flow samples RAF(t);
Step S500, obtaining EF flow sample L discarded in the DS node at the moment t and within the unit time delta tEF(t) and discarded AF traffic samples LAF(t);
Step S600, calculating a new weighted average EF flow ZEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAF(t+1);
Step S700, recalculating basic bandwidth B reserved by DS node for AF servicemax(t +1), setting and enabling Bmax(t+1);
Step S800, recalculating EF service limit bandwidth R of DS nodemax(t +1), set and enable Rmax(t+1);
Step S900, make ZEF(t)=ZEF(t+1),ZAF(t)=ZAF(t+1),DEF(t)=DEF(t+1),DAF(t)=DAF(t +1), wait for unit time Δ t, revolutionGo to step S400.
In an embodiment of the present invention, the unit time Δ t in step S100 further includes:
the unit time delta t is the updating period of the basic bandwidth reserved by the AF service and the EF service limited bandwidth on the DS node, and the default value of the delta t is 6 seconds.
In an embodiment of the present invention, the weighted average EF flow Z at the time t in the step S200EF(t) and the weighted average AF flow Z at time tAF(t), further comprising:
in the DS node, when t is 0, that is, when the dynamic traffic distribution is first implemented by using the method, the weighted average EF traffic Z is obtainedEF(0) 0.8 xW, weighted average AF flow ZAF(0) 0.2 × W, where W is the maximum bandwidth that can be provided by the DS node.
In an embodiment of the present invention, the weighted average discarded EF traffic D at the time t in the step S300EF(t), weighted average of discarded AF traffic D at time tAF(t), further comprising:
in the DS node, when t is 0, that is, when the dynamic traffic allocation is first implemented by using the method, the weighted average discarded EF traffic D is obtainedEF(0) Weighted average of the discarded AF traffic D at 0AF(0)=0。
In an embodiment of the invention, the EF flow samples R in the unit time Δ t in the step S400EF(t) and AF flow samples RAF(t), further comprising:
in the DS node, EF flow and AF flow statistical data are extracted by obtaining flow logs in the current period and the last period, and the absolute value of the difference value of the respective statistical data of the EF flow and the AF flow in the two periods is an EF flow sample R in unit time delta tEF(t) and AF flow samples RAF(t);
When t is equal to 0, namely when the method is adopted to realize dynamic flow distribution for the first time, EF flow samples R in unit time delta tEF(0) AF flow sample R ═ 0.8 × WAF(0) 0.2 × W, where W is the maximum bandwidth that can be provided by the DS node.
In one embodiment of the present invention, in the step S500, a single stepEF traffic samples L discarded within a bit time Δ tEF(t) and discarded AF traffic samples LAF(t), further comprising:
in the DS node, the discarded EF traffic and the discarded AF traffic statistical data are extracted by obtaining the traffic logs in the current period and the previous period, and the absolute value of the difference value of the respective statistical data of the discarded EF traffic and the discarded AF traffic in the two periods is the discarded EF traffic sample L in the unit time delta tEF(t) and discarded AF traffic samples LAF(t);
When t is equal to 0, namely when the method is adopted to realize dynamic flow distribution for the first time, the EF flow sample L discarded in unit time delta tEF(0) Discarded AF traffic sample L-0AF(0)=0。
In an embodiment of the present invention, the new weighted average EF flow Z is calculated in step S600EF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAF(t +1), further comprising:
new weighted average EF flow ZEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAFThe calculation formula of (t +1) is as follows:
ZEF(t+1)=(1-α)×ZEF(t)+ ×REF(t)
ZAF(t+1)=(1-α)×ZAF(t)+α×RAF(t)
DEF(t+1)=(1-β)×DEF(t)+β×LEF(t)
DAF(t+1)=(1-β)×DAF(t)+β×LAF(t)
wherein Z isEF(t) EF flow, which is the weighted average of the last period, ZAF(t) is the weighted average AF flow from the previous cycle, DEF(t) EF traffic discarded as a weighted average of the last cycle, DAF(t) AF traffic discarded as a weighted average of the last cycle, REF(t) EF traffic of the previous cycleSample, RAF(t) is the AF flow sample of the previous cycle, LEF(t) is the EF flow sample of the previous cycle, LAF(t) is the discarded AF flow sample of the previous cycle, α and β are adjustment factors, 0 ≦ α, β<1, alpha is 0.125 by default and beta is 0.25 by default.
In an embodiment of the present invention, in the step S700, a basic bandwidth B reserved by the DS node for the AF service is calculatedmax(t +1), further comprising:
basic bandwidth B reserved by DS node for AF servicemaxThe calculation formula of (t +1) is as follows:
Bmax(t+1)=ZAF(t+1)+γ×DAF(t+1)
wherein Z isAF(t +1) is the new weighted average AF flow, DAF(t +1) is the new weighted average discarded AF traffic, γ is the adjustment factor, 0 ≦ γ, with a default value of 2.
In an embodiment of the present invention, in the step S800, the EF service restriction bandwidth R of the DS node is calculatedmax(t +1), further comprising:
EF traffic limited bandwidth R of DS nodemaxThe calculation formula of (t +1) is as follows:
Rmax(t+1)=min[ZEF(t+1)+γ×DEF(t+1),W-Bmax(t+Δt)]
wherein Z isEF(t +1) is the new weighted average EF flow, DEF(t +1) is the new weighted average discarded EF traffic, Bmax(t +1) is the basic bandwidth reserved by the DS node for the AF service, W is the maximum bandwidth provided by the DS node, gamma is an adjustment factor, gamma is not less than 0, and the default value of gamma is 2.
(III) advantageous effects
The invention has the beneficial effects that: the dynamic flow allocation method for accelerating the forwarding of the EF service, provided by the embodiment of the invention, provides a dynamic flow allocation method for accelerating the forwarding of the EF service, dynamically adjusts the flow limitation rate, and solves the problems that the flow limitation of the EF service depends on the experience of a network administrator for configuration, the quality of the EF service is influenced, and the bandwidth of a network node is wasted.
Drawings
Fig. 1 is a flowchart of a dynamic traffic allocation method for expedited forwarding of an EF service according to an embodiment of the present invention.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
In the related embodiment of the invention, a dynamic flow allocation method for accelerating the forwarding of the EF service is provided, the flow limitation rate is dynamically adjusted, and the problems that the flow limitation of the EF service is configured depending on the experience of a network administrator, the quality of the EF service is influenced, and the bandwidth of a network node is wasted are solved.
Fig. 1 is a flowchart of a dynamic traffic allocation method for accelerating forwarding of an EF service according to an embodiment of the present invention, and as shown in fig. 1, the method includes the following steps:
as shown in fig. 1, step S100, a unit time Δ t is set on the DS node;
as shown in FIG. 1, two parameters Z are set on the DS node in step S200EF(t) and ZAF(t),ZEF(t) is the weighted average EF flow at time t, ZAF(t) is the weighted average AF flow at time t;
as shown in FIG. 1, two parameters D are set on the DS node in step S300EF(t) and DAF(t),DEF(t) EF traffic discarded as a weighted average at time t, DAF(t) weighted average discarded AF traffic at time t;
as shown in fig. 1, in step S400, an EF flow sample R at time t on the DS node within unit time Δ t is obtainedEF(t) and AF flow samplesRAF(t);
As shown in fig. 1, in step S500, an EF traffic sample L discarded at time t and within unit time Δ t on the DS node is obtainedEF(t) and discarded AF traffic samples LAF(t);
As shown in FIG. 1, step S600 calculates a new weighted average EF flow ZEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAF(t+1);
As shown in fig. 1, step S700 is to recalculate the basic bandwidth B reserved by the DS node for the AF servicemax(t +1), setting and enabling Bmax(t+1);
As shown in FIG. 1, step S800 is to recalculate the EF traffic limitation bandwidth R of the DS nodemax(t +1), set and enable Rmax(t+1);
As shown in FIG. 1, step S900, let ZEF(t)=ZEF(t+1),ZAF(t)=ZAF(t+1),DEF(t)=DEF(t+1),DAF(t)=DAF(t +1), the unit time Δ t is waited for, and the process goes to step S400.
The specific implementation of the steps of the embodiment shown in fig. 1 is described in detail below:
in step S100, the unit time Δ t is set on the DS node.
In an embodiment of the present invention, the unit time Δ t in step S100 further includes:
the unit time Δ t is an update period of the basic bandwidth reserved by the AF service and the EF service restriction bandwidth on the DS node, and if the unit time Δ t is set too small, the CPU load on the DS node will be too high, and if the unit time Δ t is set too large, the basic bandwidth reserved by the AF service and the EF service restriction bandwidth on the DS node will be updated in time, which affects the effect of the method, and the default value of Δ t is 6 seconds.
In step S200, two parameters Z are set on the DS nodeEF(t) and ZAF(t),ZEF(t) is the weighted average EF flow at time t, ZAF(t) at time tThe weighted average AF flow.
In an embodiment of the present invention, the weighted average EF flow Z at the time t in the step S200EF(t) and the weighted average AF flow Z at time tAF(t), further comprising:
in the DS node, when t is 0, that is, when the dynamic traffic distribution is first implemented by using the method, the weighted average EF traffic Z is obtainedEF(0) 0.8 xW, weighted average AF flow ZAF(0) 0.2 × W, where W is the maximum bandwidth that can be provided by the DS node.
In step S300, two parameters D are set on the DS nodeEF(t) and DAF(t),DEF(t) EF traffic discarded as a weighted average at time t, DAF(t) is the weighted average of the discarded AF traffic at time t.
In an embodiment of the present invention, the weighted average discarded EF traffic D at the time t in the step S300EF(t), weighted average of discarded AF traffic D at time tAF(t), further comprising:
in the DS node, when t is 0, that is, when the dynamic traffic allocation is first implemented by using the method, the weighted average discarded EF traffic D is obtainedEF(0) Weighted average of the discarded AF traffic D at 0AF(0)=0。
In step S400, an EF flow sample R at time t on the DS node within a unit time Δ t is obtainedEF(t) and AF flow samples RAF(t)。
In an embodiment of the invention, the EF flow samples R in the unit time Δ t in the step S400EF(t) and AF flow samples RAF(t), further comprising:
in the DS node, EF flow and AF flow statistical data are extracted by obtaining flow logs in the current period and the last period, and the absolute value of the difference value of the respective statistical data of the EF flow and the AF flow in the two periods is an EF flow sample R in unit time delta tEF(t) and AF flow samples RAF(t);
When t is equal to 0, namely when the method is adopted to realize dynamic flow distribution for the first time, EF flow samples R in unit time delta tEF(0) AF flow sample R ═ 0.8 × WAF(0)=0.2 xW, wherein W is the maximum bandwidth that can be provided by the DS node.
In step S500, an EF traffic sample L discarded at time t and within unit time Δ t on the DS node is obtainedEF(t) and discarded AF traffic samples LAF(t)。
In an embodiment of the present invention, the EF traffic samples L discarded in the unit time Δ t in the step S500EF(t) and discarded AF traffic samples LAF(t), further comprising:
in the DS node, the discarded EF traffic and the discarded AF traffic statistical data are extracted by obtaining the traffic logs in the current period and the previous period, and the absolute value of the difference value of the respective statistical data of the discarded EF traffic and the discarded AF traffic in the two periods is the discarded EF traffic sample L in the unit time delta tEF(t) and discarded AF traffic samples LAF(t);
When t is equal to 0, namely when the method is adopted to realize dynamic flow distribution for the first time, the EF flow sample L discarded in unit time delta tEF(0) Discarded AF traffic sample L-0AF(0)=0。
In step S600, a new weighted average EF flow Z is calculatedEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAF(t+1)。
In an embodiment of the present invention, the new weighted average EF flow Z is calculated in step S600EF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAF(t +1), further comprising:
new weighted average EF flow ZEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAFThe calculation formula of (t +1) is as follows:
ZEF(t+1)=(1-α)×ZEF(t)+α×REF(t)
ZAF(t+1)=(1-α)×ZAF(t)+α×RAF(t)
DEF(t+1)=(1-β)×DEF(t)+β×LEF(t)
DAF(t+1)=(1-β)×DAF(t)+β×LAF(t)
wherein Z isEF(t) EF flow, which is the weighted average of the last period, ZAF(t) is the weighted average AF flow from the previous cycle, DEF(t) EF traffic discarded as a weighted average of the last cycle, DAF(t) AF traffic discarded as a weighted average of the last cycle, REF(t) is the EF flow sample of the previous cycle, RAF(t) is the AF flow sample of the previous cycle, LEF(t) is the EF flow sample of the previous cycle, LAF(t) is the discarded AF flow sample of the previous cycle, α and β are adjustment factors, 0 ≦ α, β<1, alpha is 0.125 by default and beta is 0.25 by default.
In one embodiment of the present invention, Z in the step S600EF(t+1)、ZAF(t+1)、DEF(t +1) and DAFThe calculation of (t +1) adopts a normalization method for processing, and the calculation is obtained under the action of adjustment factors (alpha and beta), so that the flow conditions of EF and AF services of the current node and the flow conditions of historical EF and AF services of the current node are considered, and Z is enabled to be obtainedEF(t+1)、ZAF(t+1)、DEF(t +1) and DAFThe change of (t +1) is smoother, severe jitter is avoided, the more the adjustment factors alpha and beta approach to 1, the larger the influence of the current flow sample is, the more alpha and beta approach to 0, the larger the influence of the historical flow sample is, and experiments prove that the alpha is 0.125, the beta value is 0.25, and the effect is better.
In step S700, the basic bandwidth B reserved by the DS node for the AF service is recalculatedmax(t +1), and setting is performed.
In an embodiment of the present invention, in the step S700, a basic bandwidth B reserved by the DS node for the AF service is calculatedmax(t +1), further comprising:
basic bandwidth B reserved by DS node for AF servicemaxCalculation of (t +1)The formula is as follows:
Bmax(t+1)=ZAF(t+1)+γ×DAF(t+1)
wherein Z isAF(t +1) is the new weighted average AF flow, DAF(t +1) is the new weighted average dropped AF traffic, γ is the adjustment factor, 0 ≦ γ, and γ has a default value of 2, the purpose of using γ is based on the dropped traffic, based on ZAF(t +1) appropriately expanding the basic bandwidth Bmax(t +1), and experiments prove that the method is reasonable.
In step S800, the EF traffic limit bandwidth R of the DS node is recalculatedmax(t +1), and setting is performed.
In an embodiment of the present invention, in the step S800, the EF service restriction bandwidth R of the DS node is calculatedmax(t +1), further comprising:
EF traffic limited bandwidth R of DS nodemaxThe calculation formula of (t +1) is as follows:
Rmax(t+1)=min[ZEF(t+1)+γ×DEF(t+1),W-Bmax(t+Δt)]
wherein Z isEF(t +1) is the new weighted average EF flow, DEF(t +1) is the new weighted average discarded EF traffic, Bmax(t +1) is the basic bandwidth reserved by the DS node for the AF service, W is the maximum bandwidth provided by the DS node, gamma is an adjustment factor, gamma is more than or equal to 0 and is the default value of 2, and the purpose of using gamma is according to the service discarding situation and according to ZEF(t +1) appropriately expanding EF traffic restriction bandwidth Rmax(t +1), experiments prove that the method is reasonable, and it should be noted that because bandwidth competition relationship exists between the EF service and the AF service, transmission of the AF service should be guaranteed as much as possible, so that R is limitedmax(t +1) should be equal to or less than W-Bmax(t+Δt)。
In step S900, Z is caused to beEF(t)=ZEF(t+1),ZAF(t)=ZAF(t+1),DEF(t)=DEF(t+1),DAF(t)=DAF(t +1), the unit time Δ t is waited for, and the process goes to step S400.
In one embodiment of the present invention, the step S900 is toZEF(t)=ZEF(t+1),ZAF(t)=ZAF(t+1),DEF(t)=DEF(t+1),DAF(t)=DAF(t +1) for saving the intermediate result calculated in the present cycle so that the next cycle is calculated using the history data, and thereafter waiting for the unit time Δ t, the flow goes to step S400, and the process proceeds to the next cycle.
In summary, the method provided in the embodiment of the present invention provides a dynamic traffic allocation method for accelerating forwarding of an EF service, which dynamically adjusts a traffic limitation rate, and solves the problems that the traffic limitation of the EF service depends on the experience of a network administrator for configuration, the quality of the EF service is affected, and the bandwidth of a network node is wasted.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.
Claims (9)
1. A dynamic flow distribution method for accelerating the forwarding of EF services is characterized by comprising the following steps:
step S100, setting unit time delta t on a DS node;
step S200, two parameters Z are set on the DS nodeEF(t) and ZAF(t),ZEF(t) is the weighted average EF flow at time t, ZAF(t) is the weighted average AF flow at time t;
step S300, two parameters D are set on the DS nodeEF(t) and DAF(t),DEF(t) EF traffic discarded as a weighted average at time t, DAF(t) weighted average discarded AF traffic at time t;
step S400, obtaining EF flow sample R at the moment t on the DS node in unit time delta tEF(t) and AF flow samples RAF(t);
Step S500, obtaining EF flow sample L discarded in the DS node at the moment t and within the unit time delta tEF(t) and discarded AF traffic samples LAF(t);
Step S600, calculating a new weighted average EF flow ZEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAF(t+1);
Step S700, recalculating basic bandwidth B reserved by DS node for AF servicemax(t +1), setting and enabling Bmax(t+1);
Step S800, recalculating EF service limit bandwidth R of DS nodemax(t +1), set and enable Rmax(t+1);
Step S900, make ZEF(t)=ZEF(t+1),ZAF(t)=ZAF(t+1),DEF(t)=DEF(t+1),DAF(t)=DAF(t +1), the unit time Δ t is waited for, and the process goes to step S400.
2. The dynamic traffic allocation method for expedited forwarding of EF traffic according to claim 1, wherein the unit time Δ t in step S100 further includes:
the unit time delta t is the updating period of the basic bandwidth reserved by the AF service and the EF service limited bandwidth on the DS node, and the default value of the delta t is 6 seconds.
3. The dynamic traffic distribution method for expedited forwarding of EF traffic as claimed in claim 1, wherein the weighted average EF traffic Z at time t in step S200 isEF(t) and the weighted average AF flow Z at time tAF(t), further comprising:
in the DS node, when t is 0, that is, when the dynamic traffic distribution is first implemented by using the method, the weighted average EF traffic Z is obtainedEF(0) 0.8 xW, weighted average AF flow ZAF(0) 0.2 × W, where W is the maximum bandwidth that can be provided by the DS node.
4. The dynamic traffic distribution method for expedited forwarding of EF traffic as claimed in claim 1, wherein the weighted average discarded EF traffic D at time t in S300 is EF traffic DEF(t), weighted average of discarded AF traffic D at time tAF(t), further comprising:
in the DS node, when t is 0, that is, when the dynamic traffic allocation is first implemented by using the method, the weighted average discarded EF traffic D is obtainedEF(0) Weighted average of the discarded AF traffic D at 0AF(0)=0。
5. The dynamic traffic distribution method for expedited forwarding of EF traffic as claimed in claim 1, wherein the EF traffic samples R in the unit time Δ t in step S400EF(t) and AF flow samples RAF(t), further comprising:
in the DS node, EF flow and AF flow statistical data are extracted by obtaining flow logs in the current period and the last period, and the absolute value of the difference value of the respective statistical data of the EF flow and the AF flow in the two periods is an EF flow sample R in unit time delta tEF(t) and AF flow samples RAF(t);
When t is equal to 0, namely when the method is adopted to realize dynamic flow distribution for the first time, EF flow samples R in unit time delta tEF(0) AF flow sample R ═ 0.8 × WAF(0) 0.2 × W, where W is the maximum bandwidth that can be provided by the DS node.
6. The dynamic traffic distribution method for expedited forwarding of EF traffic as claimed in claim 1, wherein the EF traffic samples L discarded in the unit time Δ t in step S500 areEF(t) and discarded AF traffic samples LAF(t), further comprising:
in the DS node, by acquiringAnd extracting discarded EF flow and discarded AF flow statistical data from the flow logs in the current period and the last period, wherein the absolute value of the difference between the respective statistical data of the discarded EF flow and the discarded AF flow in the two periods is the discarded EF flow sample L in the unit time delta tEF(t) and discarded AF traffic samples LAF(t);
When t is equal to 0, namely when the method is adopted to realize dynamic flow distribution for the first time, the EF flow sample L discarded in unit time delta tEF(0) Discarded AF traffic sample L-0AF(0)=0。
7. The dynamic traffic distribution method for expedited forwarding of EF traffic as claimed in claim 1, wherein the step S600 calculates a new weighted average of EF traffic ZEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAF(t +1), further comprising:
new weighted average EF flow ZEF(t +1), new weighted average AF flow ZAF(t +1), new weighted average discarded EF traffic DEF(t +1) and new weighted average dropped AF traffic DAFThe calculation formula of (t +1) is as follows:
ZEF(t+1)=(1-α)×ZEF(t)+α×REF(t)
ZAF(t+1)=(1-α)×ZAF(t)+α×RAF(t)
DEF(t+1)=(1-β)×DEF(t)+β×LEF(t)
DAF(t+1)=(1-β)×DAF(t)+β×LAF(t)
wherein Z isEF(t) EF flow, which is the weighted average of the last period, ZAF(t) is the weighted average AF flow from the previous cycle, DEF(t) EF traffic discarded as a weighted average of the last cycle, DAF(t) AF traffic discarded as a weighted average of the last cycle, REF(t) is the EF flow sample of the previous cycle, RAF(t) is the AF flow sample of the previous cycle, LEF(t) Is the EF traffic sample of the previous cycle, LAF(t) is the discarded AF flow sample of the previous cycle, α and β are adjustment factors, 0 ≦ α, β < 1, a default of 0.125, and β default 0.25.
8. The dynamic traffic allocation method for expedited forwarding of EF services according to claim 1, wherein the step S700 calculates a basic bandwidth B reserved by the DS node for the AF servicesmax(t +1), further comprising:
basic bandwidth B reserved by DS node for AF servicemaxThe calculation formula of (t +1) is as follows:
Bmax(t+1)=ZAF(t+1)+γ×DAF(t+1)
wherein Z isAF(t +1) is the new weighted average AF flow, DAF(t +1) is the new weighted average discarded AF traffic, γ is the adjustment factor, 0 ≦ γ, with a default value of 2.
9. The dynamic traffic allocation method for expedited forwarding of EF traffic as claimed in claim 1, wherein the step S800 is performed by calculating an EF traffic restriction bandwidth R of the DS nodemax(t +1), further comprising:
EF traffic limited bandwidth R of DS nodemaxThe calculation formula of (t +1) is as follows:
Rmax(t+1)=min[ZEF(t+1)+γ×DEF(t+1),W-Bmax(t+Δt)]
wherein Z isEF(t +1) is the new weighted average EF flow, DEF(t +1) is the new weighted average discarded EF traffic, Bmax(t +1) is the basic bandwidth reserved by the DS node for the AF service, W is the maximum bandwidth provided by the DS node, gamma is an adjustment factor, gamma is not less than 0, and the default value of gamma is 2.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011424766.4A CN112600762B (en) | 2020-12-09 | 2020-12-09 | Dynamic flow distribution method for accelerating EF service forwarding |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011424766.4A CN112600762B (en) | 2020-12-09 | 2020-12-09 | Dynamic flow distribution method for accelerating EF service forwarding |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112600762A true CN112600762A (en) | 2021-04-02 |
CN112600762B CN112600762B (en) | 2023-01-10 |
Family
ID=75191144
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011424766.4A Active CN112600762B (en) | 2020-12-09 | 2020-12-09 | Dynamic flow distribution method for accelerating EF service forwarding |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112600762B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004080207A (en) * | 2002-08-13 | 2004-03-11 | Nippon Telegr & Teleph Corp <Ntt> | System, method and its program for packet transfer control, recording medium, and communication apparatus |
CN1581848A (en) * | 2004-05-18 | 2005-02-16 | 北京邮电大学 | Flow regulating method for ensuring integrated flow fairness of guaranted repeat business |
JP2006222821A (en) * | 2005-02-14 | 2006-08-24 | Alaxala Networks Corp | Packet transfer device |
US20070253438A1 (en) * | 2006-04-28 | 2007-11-01 | Tellabs San Jose, Inc. | Differentiated services using weighted quality of service (QoS) |
CN101808033A (en) * | 2010-03-09 | 2010-08-18 | 杭州华三通信技术有限公司 | Method and apparatus for allocating reservation bandwidth of traffic |
US8391356B1 (en) * | 2009-02-18 | 2013-03-05 | Sprint Communications Company L.P. | Scalable video coding priority marking |
CN105119682A (en) * | 2015-08-27 | 2015-12-02 | 重庆邮电大学 | User level protocol and QoS (Quality of Service) based dynamic wavelength bandwidth distribution method in WDM-EPON (Wavelength Division Multiplexing-Ethernet Passive Optical Network) |
-
2020
- 2020-12-09 CN CN202011424766.4A patent/CN112600762B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004080207A (en) * | 2002-08-13 | 2004-03-11 | Nippon Telegr & Teleph Corp <Ntt> | System, method and its program for packet transfer control, recording medium, and communication apparatus |
CN1581848A (en) * | 2004-05-18 | 2005-02-16 | 北京邮电大学 | Flow regulating method for ensuring integrated flow fairness of guaranted repeat business |
JP2006222821A (en) * | 2005-02-14 | 2006-08-24 | Alaxala Networks Corp | Packet transfer device |
US20070253438A1 (en) * | 2006-04-28 | 2007-11-01 | Tellabs San Jose, Inc. | Differentiated services using weighted quality of service (QoS) |
US8391356B1 (en) * | 2009-02-18 | 2013-03-05 | Sprint Communications Company L.P. | Scalable video coding priority marking |
CN101808033A (en) * | 2010-03-09 | 2010-08-18 | 杭州华三通信技术有限公司 | Method and apparatus for allocating reservation bandwidth of traffic |
CN105119682A (en) * | 2015-08-27 | 2015-12-02 | 重庆邮电大学 | User level protocol and QoS (Quality of Service) based dynamic wavelength bandwidth distribution method in WDM-EPON (Wavelength Division Multiplexing-Ethernet Passive Optical Network) |
Non-Patent Citations (3)
Title |
---|
M.P.FERNANDEZ,等: "QoS provisioning across a DiffServ domain using policy-based management", 《 GLOBECOM"01. IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE (CAT. NO.01CH37270)》, 6 August 2022 (2022-08-06) * |
葛敬国等: "基于剩余路径跳数的动态优先调度实现区分服务EF PHB", 《微电子学与计算机》, no. 07, 25 July 2002 (2002-07-25) * |
郑枫等: "区分服务网络域内定性服务的动态配置", 《计算机工程》, no. 06, 20 June 2001 (2001-06-20) * |
Also Published As
Publication number | Publication date |
---|---|
CN112600762B (en) | 2023-01-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10122637B2 (en) | Advertising network layer reachability information specifying a quality of service for an identified network flow | |
US6996062B1 (en) | Policy-based weighted random early detection method for avoiding congestion in internet traffic | |
US8861358B2 (en) | Applying router quality of service on a cable modem interface on a per-service-flow basis | |
US7263063B2 (en) | Per hop behavior for differentiated services in mobile ad hoc wireless networks | |
US8130650B2 (en) | Retro flow control for arriving traffic in computer networks | |
US20190014053A1 (en) | Network Flow Control Method And Network Device | |
JP4619584B2 (en) | Method for scheduling packets at a router in a packet switched network | |
US20050201373A1 (en) | Packet output-controlling device, packet transmission apparatus | |
US7428216B2 (en) | Method and apparatus for policy and admission control in packet-based communication systems | |
JP2003087316A (en) | Method of transmitting data | |
JP2002111742A (en) | Method for marking packet of data transmission flow and marker device performing this method | |
US20090323525A1 (en) | Priority aware policer and method of priority aware policing | |
CN113726671A (en) | Network congestion control method and related product | |
CN112600762B (en) | Dynamic flow distribution method for accelerating EF service forwarding | |
Bodamer | A new scheduling mechanism to provide relative differentiation for real-time IP traffic | |
JP2005295524A (en) | Packet output-controlling device, packet shaper, and packet repeater | |
JP4500701B2 (en) | Packet transfer device | |
Andrew et al. | MaxNet: Theory and implementation | |
Le Boudec et al. | Asymmetric best effort service for packet networks | |
Phan et al. | FICC-DiffServ: A new QoS architecture supporting resources discovery, admission and congestion controls | |
Karimi et al. | Qos mapping method based on DSCP/IP in LTE and EDCA AC/MAC in wifi network | |
Shinohara et al. | Queue management scheme stabilizing buffer utilization in the IP router | |
Gaidioz et al. | Differentiated fairness: Service model and implementation | |
Büchli et al. | On the Efficiency of Voice over Integrated Services using Guaranteed Service | |
Nahm et al. | Enhanced service differentiation for layered video multicast in differentiated service networks |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |