CN113316261A - Multi-dimensional flow comprehensive control system and flow ordering method - Google Patents

Multi-dimensional flow comprehensive control system and flow ordering method Download PDF

Info

Publication number
CN113316261A
CN113316261A CN202110868579.3A CN202110868579A CN113316261A CN 113316261 A CN113316261 A CN 113316261A CN 202110868579 A CN202110868579 A CN 202110868579A CN 113316261 A CN113316261 A CN 113316261A
Authority
CN
China
Prior art keywords
user
service
bandwidth
plane
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.)
Granted
Application number
CN202110868579.3A
Other languages
Chinese (zh)
Other versions
CN113316261B (en
Inventor
谢永强
汲锡林
许波
李忠博
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Network Engineering Institute of Systems Engineering Academy of Military Sciences
Original Assignee
Institute of Network Engineering Institute of Systems Engineering Academy of Military Sciences
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Institute of Network Engineering Institute of Systems Engineering Academy of Military Sciences filed Critical Institute of Network Engineering Institute of Systems Engineering Academy of Military Sciences
Priority to CN202110868579.3A priority Critical patent/CN113316261B/en
Publication of CN113316261A publication Critical patent/CN113316261A/en
Application granted granted Critical
Publication of CN113316261B publication Critical patent/CN113316261B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation
    • H04W72/044—Wireless resource allocation based on the type of the allocated resource
    • H04W72/0453—Resources in frequency domain, e.g. a carrier in FDMA
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control

Abstract

The embodiment of the invention provides a multi-dimensional flow comprehensive control system and a flow ordering method, relates to the technical field of multi-dimensional flow control, and aims to solve the problems that a flow control system with global height is lacked and bandwidth allocation depends on subjective lack of objective basis of managers. The method comprises the following steps: the first step is as follows: defining related parameters; the second step is that: dividing plane distribution guarantee dimensions; the third step: dividing service classification guarantee dimensions; the fourth step: dividing bandwidth allocation guarantee dimensions; the fifth step: determining a plane shunting strategy, and a sixth step: and determining a service classification guarantee strategy.

Description

Multi-dimensional flow comprehensive control system and flow ordering method
Technical Field
The invention relates to the technical field of multidimensional flow control, in particular to a multidimensional flow comprehensive control system and a flow ordering method.
Background
At present, network application enters the full media era, various new services and new applications emerge endlessly, the development of various services of users depends on the support provided by the network, the viscosity of the users to the network is continuously enhanced, meanwhile, the services also occupy a large amount of network bandwidth resources, and the network construction increasingly faces bandwidth expansion pressure, among various services, unnecessary, non-urgent, even non-official large bandwidth demand services occupy a large amount of bandwidth resources, but important, core and urgent services and users cannot be effectively guaranteed bandwidth resources, the mismatching between the importance of the secured object and the allocation mode of the secured resources causes the disorder of network traffic and the waste of bandwidth resources, and the control on the user traffic is urgently needed from the purposes of providing differentiated services and refined bandwidth security for users.
The flow control scheme in the prior art is usually based on a certain technical scheme, such as DSCP, RSVP, policy routing, etc., or a flow control product (system) of one manufacturer, so that the advantages of various technologies or products of multiple manufacturers cannot be fully exerted, a flow control system cannot be formed, and microscopic local optimization and global flow disordering are easily caused.
In addition, the existing flow control scheme can allocate and use bandwidth resources based on service types, users and the like, but cannot realize reasonable allocation and mapping service requirements to limited bandwidth resource guarantee capacity scientifically and reasonably. The bandwidth resource allocation depends on the experience of management personnel and subjective analysis, and the subjective analysis easily causes the problem that the bandwidth resource cannot be quickly and flexibly adjusted according to the quick change of the traffic situation and the service requirement.
Disclosure of Invention
In order to solve the technical problems, the invention provides a multi-dimensional flow comprehensive control system and a flow ordering method, and aims to solve the problems that a flow control system with overall height is missing and bandwidth allocation depends on subjective lack of objective basis of managers.
A multidimensional traffic comprehensive management and control ordering method comprises the following steps:
the first step is as follows: defining related parameters;
the second step is that: dividing plane distribution guarantee dimensions;
the plane refers to an independent IP routing plane and comprises a physical network or a virtual network; dividing the IP routing plane into an HQS plane and an LQS plane;
the third step: dividing service classification guarantee dimensions;
mapping the user service type to a plurality of groups based on a group model, wherein the user service type corresponds to a certain field, applying the field marking strategy to each router input port, and adding a field mark to each router input port message; the HQS plane and the LQS plane are managed by a switch device according to the field flags;
the fourth step: dividing bandwidth allocation guarantee dimensions;
setting flow control equipment at the front end of each router, and controlling the flow and distributing bandwidth resources to each user service according to a certain proportion;
the fifth step: determining a plane offload policy for bandwidth, comprising: shunting bandwidth according to task importance, shunting bandwidth according to user service importance, shunting bandwidth according to multi-dimensional service, and shunting according to network bandwidth occupation and flow distribution conditions;
and a sixth step: and establishing a mapping relation table between the service types and the guarantee types, wherein each service guarantee type corresponds to one value or a plurality of values, and when the network is congested, adjusting the bandwidth according to the mapping relation table.
The first step comprises: defining users, user importance, user service importance, user service types, tasks and task importance;
the user refers to a person or a set of persons who directly use various types of applications, and the user is represented by an IP address or a set of IP addresses;
the user importance refers to a set of priority degrees when a user or a user group uses network bandwidth resources;
the user service refers to a set of information systems or services such as data, audio or video which are directly used by a user through an IP network, and the user service is determined according to a TCP/UDP port number;
the user service importance refers to a set of priority degrees of network bandwidth resources occupied by user services;
the user service type is divided into 32 types according to different expected combinations according to four dimensions of time delay sensitivity, packet loss sensitivity, jitter sensitivity and throughput;
the task refers to a role which is assumed by a user to achieve a certain purpose and is represented as a series of work which is carried out by the user by using user service, and the task is a set of specific tasks;
the task importance refers to a set of priorities of work related to the task while occupying bandwidth resources.
The user service is divided into a plurality of types according to time delay sensitivity, packet loss sensitivity, jitter sensitivity and throughput; service type adoption
Figure 63885DEST_PATH_IMAGE001
It is shown that,
Figure 59654DEST_PATH_IMAGE002
representing the number of the service types, and representing sensitivity or insensitivity by adopting 0 and 1 for delay sensitivity, packet loss sensitivity and jitter sensitivity, and representing insensitivity by adopting 0 and representing sensitivity by adopting 0; dividing the throughput into four continuous threshold segments from small to large, wherein the first threshold segment is 00, the second threshold segment is 01, the third threshold segment is 10 and the fourth threshold segment is 11; and establishing a user service type table.
The shunting the bandwidth according to the task importance comprises the following steps:
the task set is as follows:
Figure 901708DEST_PATH_IMAGE003
；
sorting by task importance:
Figure 608502DEST_PATH_IMAGE004
defining the total bandwidth as the sum of the egress bandwidths of all routers, expressed as:
Figure 485191DEST_PATH_IMAGE005
estimating the bandwidth of each task, and recording as:
Figure 120703DEST_PATH_IMAGE006
according to the sequencing relation, the bandwidth of each user is accumulated and summed one by one to obtain:
Figure 184474DEST_PATH_IMAGE007
when is coming into contact with
Figure 697888DEST_PATH_IMAGE008
Greater than the total bandwidth of the network
Figure 429084DEST_PATH_IMAGE009
70% of, the task
Figure 235497DEST_PATH_IMAGE010
Then the HQS plane is shunted, task
Figure 52143DEST_PATH_IMAGE011
Shunting to the LQS plane.
The shunting the bandwidth according to the importance of the user comprises the following steps: the user set is set as follows:
Figure 100739DEST_PATH_IMAGE012
；
ordering by user importance:
Figure 420862DEST_PATH_IMAGE013
defining the total network bandwidth as the sum of the exit bandwidths of all the access routers as follows:
Figure 913023DEST_PATH_IMAGE014
and estimates the bandwidth of each user, recorded as:
Figure 967698DEST_PATH_IMAGE015
bandwidth to each user in a sorted relationshipAre accumulated and summed one by one to obtain
Figure 305139DEST_PATH_IMAGE016
When is coming into contact with
Figure 257264DEST_PATH_IMAGE017
Greater than the total bandwidth of the network
Figure 389168DEST_PATH_IMAGE018
70% of (1), user
Figure 931139DEST_PATH_IMAGE019
Then the HQS plane is shunted, the user
Figure 72271DEST_PATH_IMAGE020
Shunting to the LQS plane.
The shunting the bandwidth according to the importance of the user service comprises the following steps:
setting the service set as:
Figure 616253DEST_PATH_IMAGE021
normalizing and ordering according to task importance priority:
Figure 450217DEST_PATH_IMAGE022
defining the total network bandwidth as the sum of the exit bandwidths of all the access routers as follows:
Figure 213905DEST_PATH_IMAGE023
and estimating the bandwidth of each user service, and recording as:
Figure 158727DEST_PATH_IMAGE024
in order of sequenceThe system carries out accumulation and summation to each user service bandwidth one by one to obtain:
Figure 560146DEST_PATH_IMAGE025
when is coming into contact with
Figure 565011DEST_PATH_IMAGE026
Greater than the total bandwidth of the network
Figure 81574DEST_PATH_IMAGE027
70% of the time, per user service
Figure 564508DEST_PATH_IMAGE028
Then the HQS plane is shunted, the user traffic
Figure 348662DEST_PATH_IMAGE029
Shunting to the LQS plane.
The shunting the bandwidth according to the multi-dimensional service comprises the following steps:
firstly, identifying a multidimensional service flow;
tasks, applications, and users are represented as a set of services:
Figure 524429DEST_PATH_IMAGE030
，T m on behalf of a particular task or tasks, the user may,
Figure 528288DEST_PATH_IMAGE031
on behalf of a particular application, it is possible to use,
Figure 814913DEST_PATH_IMAGE032
representing a particular user;
sequencing the service sets, and firstly setting a multi-dimensional service flow set
Figure 185064DEST_PATH_IMAGE033
Then, the importance of a certain service flow is defined as:
Figure 531732DEST_PATH_IMAGE034
normalizing the priority of the service flow to obtain:
Figure 491729DEST_PATH_IMAGE035
Figure 113203DEST_PATH_IMAGE036
for the whole of the identifiable traffic flows,
Figure 606370DEST_PATH_IMAGE037
is a specific service flow;
then, sequencing the results after the service flow priority normalization:
Figure 858360DEST_PATH_IMAGE038
Figure 571232DEST_PATH_IMAGE039
is to the service flow
Figure 199660DEST_PATH_IMAGE040
Reordering by importance;
defining the total bandwidth of the network as the sum of the exit bandwidths of all the access routers
Figure 815842DEST_PATH_IMAGE041
Estimating the bandwidth of each user service, and recording as follows:
Figure 255045DEST_PATH_IMAGE042
according to the ordering relation, the service bandwidths of all users are accumulated and summed one by one to obtain
Figure 704481DEST_PATH_IMAGE043
When is coming into contact with
Figure 385867DEST_PATH_IMAGE044
Greater than the total bandwidth of the network
Figure 338779DEST_PATH_IMAGE045
70%, multidimensional traffic flow
Figure 948883DEST_PATH_IMAGE046
The HQS plane is split and,
multidimensional traffic flow
Figure 620036DEST_PATH_IMAGE047
Shunting to the LQS plane.
A multidimensional flow comprehensive control system adopts the multidimensional flow comprehensive control ordering method, and comprises the following steps: an access layer, an HQS plane, an LQS plane and a traffic sensing and scheduling service device;
the access layer is connected with a user, the HQS plane is provided with a plurality of routers and a plurality of flow control devices, and the LQS plane is provided with a plurality of routers and a plurality of flow control devices;
the method comprises the following steps that an HQS plane bears fixed application of a flow model, and an LQS plane bears uncertain application of the flow model;
and the router routes the service flow to the HQS plane and the LQS plane respectively according to the task, the user and the user service level.
The embodiment of the invention provides a multi-dimensional comprehensive flow control system and a flow ordering method, and provides a task-driven multi-granularity comprehensive flow control system and a bandwidth allocation strategy, which provide a feasible scheme for establishing a whole network flow control system for a network management department, and provide a resource allocation method for reasonably utilizing limited bandwidth resources according to the importance and urgency of guarantee of users, tasks and services, so that the whole network flow ordering performance is greatly improved.
Drawings
Fig. 1 is a schematic diagram of a multidimensional traffic comprehensive management and control architecture according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a PHB value priority of a delay and jitter sensitive service according to an embodiment of the present invention;
fig. 3 is a schematic diagram of policy mapping according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
The embodiment of the invention provides a multidimensional flow comprehensive control ordering method, which comprises the following steps of firstly defining relevant parameters:
(1) user' s
A user refers to a person who directly uses various types of applications or a set of a series of person sets (user groups), and the user is represented by an IP address or a set of IP addresses.
Figure 125621DEST_PATH_IMAGE048
Figure 198619DEST_PATH_IMAGE049
A specific user is expressed as
Figure 979624DEST_PATH_IMAGE050
Wherein
Figure 138073DEST_PATH_IMAGE051
；
(2) Importance of user
User importance is the set of priorities, typically specified by an administrator, of users (or groups of users) when using network bandwidth resources.
Figure 161262DEST_PATH_IMAGE052
The importance of a specific user is expressed as
Figure 88767DEST_PATH_IMAGE053
Wherein
Figure 306252DEST_PATH_IMAGE054
(3) User service
The user service refers to a set of various information systems or services such as data, audio or video which are directly used by a specific user through an IP network, and specific user service is determined according to a TCP/UDP port number.
Figure 951997DEST_PATH_IMAGE055
The specific user service is expressed as
Figure 516227DEST_PATH_IMAGE056
Wherein
Figure 563818DEST_PATH_IMAGE057
(4) Importance of user traffic
The set is the priority degree of the network bandwidth resources occupied by the user service, and is generally specified by management personnel.
Figure 686626DEST_PATH_IMAGE058
The service importance of a specific user is
Figure 819667DEST_PATH_IMAGE059
Wherein
Figure 919079DEST_PATH_IMAGE060
(5) Type of user service
The user service types refer to four dimensions of time delay sensitivity, packet loss sensitivity, jitter sensitivity and throughput, and the user services in the IP bearer network are divided into 32 types according to different expected combinations. The first threshold segment is 00 and may be defined as small, the second threshold segment is 01 and may be defined as small, the third threshold segment is 10 and may be defined as large, and the fourth threshold segment is 11 and may be defined as large.
Figure 821176DEST_PATH_IMAGE061
TABLE 1 subscriber service types
Figure 380464DEST_PATH_IMAGE062
(6) Task
A specific task is the responsibility the user undertakes to achieve a certain purpose, which is represented by the user performing a series of tasks using the user service. A task is a collection of specific tasks
Figure 735222DEST_PATH_IMAGE063
A specific task is expressed as
Figure 900974DEST_PATH_IMAGE064
Wherein
Figure 657578DEST_PATH_IMAGE065
(7) Importance of task
The set of priorities of task-related work in occupying bandwidth resources is typically specified by a manager.
Figure 122188DEST_PATH_IMAGE066
The importance of a specific task is expressed as
Figure 229822DEST_PATH_IMAGE067
Wherein
Figure 936615DEST_PATH_IMAGE068
。
Referring to fig. 1, an embodiment of the present invention further provides a multidimensional traffic comprehensive management and control system, where the management and control system applies the above-mentioned multidimensional traffic comprehensive management and control ordering method, and the following description is further described in detail with reference to the accompanying drawings.
(1) Plane split guarantee dimension
The plane refers to an independent IP routing plane, and may be an independent physical network, or a virtual network that occupies independent bandwidth resources and is virtualized on the physical network. According to the quality requirement of service guarantee, a plurality of planes can be divided, and the invention adopts two plane division methods: the hqs (high Quality service) plane bandwidth resource is abundant compared with the service bandwidth requirement which needs to be guaranteed, mainly bears the application with high service Quality requirement and relatively fixed traffic model, and the lqs (low Quality service) plane bandwidth resource is tense compared with the service bandwidth requirement which needs to be guaranteed, mainly bears the application with low service Quality requirement and uncertain traffic model. The access router (namely, the plane splitter) routes the service flow to the HQS plane and the LQS plane respectively through an ACL + strategy routing mode according to whether the service quality of tasks, users and users needs to be guaranteed.
(2) Service classification guarantee dimension
Based on an Assured Forwarding (AF) packet model suggested in a PHB (Per-Hop-Behavior) -oriented group determined by RFC2597, 32 user service types specified in table 1 are mapped to AF packets, that is, each user service type corresponds to a DSCP field, and a DSCP marking policy is applied to an ingress port of a router, where DSCP marking is filled for each ingress packet. The HQS plane and the LQS plane route switching equipment perform management control according to the DSCP mark.
(3) Bandwidth allocation guarantee dimension
The traffic control device is deployed at the front end of each router in the network, and is used for controlling traffic entering an access layer, a convergence layer or a core layer from a certain direction, so that the traffic control device is used for controlling the use of bandwidth resources by user services from a traffic source and reasonably allocating the bandwidth resources to each user service according to the principle of priority guarantee of important user services.
The flow ordering refers to that network bandwidth resources are scientifically and reasonably distributed and used through various strategies or methods, and the important distribution principle is that important tasks, important users and important user services can preferentially occupy the bandwidth resources, and the indexes of time delay, jitter, packet loss, throughput and the like are optimized as far as possible, so that high-quality transmission service is obtained as far as possible.
Plane shunting strategy
(1) Plane shunting according to task importance
Set up task
Figure 547725DEST_PATH_IMAGE069
Task importance prioritization:
Figure 714395DEST_PATH_IMAGE070
defining the total bandwidth of the network as the sum of the exit bandwidths of all the access routers
Figure 43746DEST_PATH_IMAGE071
And estimating the bandwidth of each task, recording
Figure 557160DEST_PATH_IMAGE072
According to the sorting relation, the bandwidth of each user is accumulated and summed one by one
Figure 22776DEST_PATH_IMAGE073
When is coming into contact with
Figure 94769DEST_PATH_IMAGE074
Greater than the total bandwidth of the network
Figure 176994DEST_PATH_IMAGE075
70% of, the task
Figure 960011DEST_PATH_IMAGE076
Then the HQS plane is shunted, task
Figure 545713DEST_PATH_IMAGE078
Shunting to the LQS plane.
(2) Plane splitting by user importance
Is provided with a user set
Figure 54186DEST_PATH_IMAGE079
User importance prioritization:
Figure 826970DEST_PATH_IMAGE080
defining the total bandwidth of the network as the sum of the exit bandwidths of all the access routers
Figure 676328DEST_PATH_IMAGE075
And estimating the bandwidth of each user, recording
Figure 382116DEST_PATH_IMAGE081
According to the sorting relation, the bandwidth of each user is accumulated and summed one by one
Figure 530331DEST_PATH_IMAGE082
When is coming into contact with
Figure 587149DEST_PATH_IMAGE083
Greater than the total bandwidth of the network
Figure 977548DEST_PATH_IMAGE075
70% of (1), user
Figure 554154DEST_PATH_IMAGE084
Then the HQS plane is shunted, the user
Figure 122538DEST_PATH_IMAGE085
Shunting to the LQS plane.
(3) Plane shunting according to user service importance
Provisioning service sets
Figure 653270DEST_PATH_IMAGE086
And (3) ordering after task importance priority normalization:
Figure 598093DEST_PATH_IMAGE087
defining the total bandwidth of the network as the sum of the exit bandwidths of all the access routers
Figure 747314DEST_PATH_IMAGE075
And estimating bandwidth of each user service, recording
Figure 502912DEST_PATH_IMAGE088
According to the sorting relation, the service bandwidths of all users are accumulated and summed one by one
Figure 3163DEST_PATH_IMAGE089
When is coming into contact with
Figure 735365DEST_PATH_IMAGE043
Greater than the total bandwidth of the network
Figure 270251DEST_PATH_IMAGE075
70% of time, user traffic
Figure 931171DEST_PATH_IMAGE090
Then the HQS plane is shunted, the user traffic
Figure 653139DEST_PATH_IMAGE091
Shunting to the LQS plane.
(4) Planar splitting for multidimensional importance
1) Multi-dimensional traffic flow identification
Task-application-user, i.e.
Figure 451681DEST_PATH_IMAGE092
2) Multidimensional traffic flow prioritization
Provisioning of multidimensional service flow sets
Figure 841074DEST_PATH_IMAGE093
Defining the importance of a certain service flow as follows:
Figure 672895DEST_PATH_IMAGE094
and (3) traffic flow priority normalization:
Figure 147739DEST_PATH_IMAGE095
（
Figure 487322DEST_PATH_IMAGE096
for all identifiable traffic flows)
And (3) ordering after the priority normalization of the service flows:
Figure 747533DEST_PATH_IMAGE097
defining the total bandwidth of the network as the sum of the exit bandwidths of all the access routers
Figure 265102DEST_PATH_IMAGE075
And estimating bandwidth of each user service, recording
Figure 745019DEST_PATH_IMAGE098
According to the sorting relation, the service bandwidths of all users are accumulated and summed one by one
Figure 373446DEST_PATH_IMAGE099
When is coming into contact with
Figure 222584DEST_PATH_IMAGE100
Greater than the total bandwidth of the network
Figure 911055DEST_PATH_IMAGE075
70%, multidimensional traffic flow
Figure 609758DEST_PATH_IMAGE101
Then the HQS plane, multi-dimensional traffic is split
Figure 776297DEST_PATH_IMAGE102
Shunting to the LQS plane.
(5) Dynamic plane shunt adjustment strategy
The method comprises the steps of selecting a task importance, user service importance or multi-dimensional importance plane shunting strategy as required, and dynamically adjusting the plane shunting strategy according to the current network bandwidth occupation and flow distribution conditions.
Service classification guarantee strategy
(1) Basic strategy for mapping service type and guarantee type
The principles are that delay sensitive and jitter sensitive traffic types try to use high priority PHB behavior to reduce queuing time in the routing switch device, and packet loss sensitive and low throughput traffic types try to use lowest drop precedence selected PHB behavior to reduce strategic packet loss due to queuing. The basic mapping relationship between the service type and the provisioning type is shown in table 2. One PHB value and one DSCP value can meet the requirement, and certainly, two columns can be adopted according to the requirement, namely one column is the PHB value, and the other column is the DSCP value.
TABLE 2 service type and Provisioning type proposed mapping relationship
Figure 745521DEST_PATH_IMAGE103
(2) The service type and the guarantee type map the dynamic adjustment policy, see fig. 2.
When the network has the guarantee quality reduction, the service type and the guarantee type mapping are dynamically adjusted according to the following criteria.
1) The service sensitive to time delay and jitter is transversely adjusted, and the forwarding success rate is reduced by improving the discarding probability;
2) and for the service types sensitive to the throughput, longitudinal adjustment is carried out, and the forwarding priority is reduced.
Multi-parameter bandwidth allocation policy
And the flow control strategy information is issued to the flow control equipment through the flow control system, the flow monitoring mode is adjusted in real time, and when the flow control strategy is started, the flow control strategy is started.
(1) Traffic flow prioritization procedure
1) Flow identification unit
Task-application-user, i.e.
Figure 604893DEST_PATH_IMAGE104
2) Traffic flow prioritization
Figure 545821DEST_PATH_IMAGE105
And (3) traffic flow priority normalization:
Figure 781631DEST_PATH_IMAGE106
（
Figure 605361DEST_PATH_IMAGE107
for all identifiable traffic flows)
And (3) ordering after the priority normalization of the service flows:
Figure 901213DEST_PATH_IMAGE109
formula 2-1
(2) Traffic flow bandwidth policy mapping procedure
1) Traffic flow policy
The service flow strategies are classified into four categories:
Figure 308930DEST_PATH_IMAGE110
wherein:
Figure 817271DEST_PATH_IMAGE111
securing a range for a bandwidth
Figure 761088DEST_PATH_IMAGE112
I.e. guaranteed bandwidth is less than the set bandwidth
Figure 962262DEST_PATH_IMAGE075
Figure 125784DEST_PATH_IMAGE113
For fixed bandwidth guarantees
Figure 437816DEST_PATH_IMAGE114
I.e. guaranteed bandwidth equal to fixed set bandwidth
Figure 236139DEST_PATH_IMAGE115
Figure 608214DEST_PATH_IMAGE116
Securing a range for a bandwidth
Figure 990523DEST_PATH_IMAGE117
I.e. guarantee bandwidth not less than set bandwidth
Figure 106247DEST_PATH_IMAGE115
Is not more than
Figure 759076DEST_PATH_IMAGE075
Figure 302053DEST_PATH_IMAGE118
Securing a range for a bandwidth
Figure 903149DEST_PATH_IMAGE119
I.e. guaranteed bandwidth is greater than the set bandwidth
Figure 822563DEST_PATH_IMAGE115
Service flow strategy type guarantee sequencing:
Figure 64320DEST_PATH_IMAGE120
2) process for classifying priority similarity of service flows
The collection of the change rate of each point after the service flow priority normalization:
Figure 778198DEST_PATH_IMAGE121
ranking of Change Rate to find the first 3 points
Figure 135099DEST_PATH_IMAGE122
The service flow priority after normalization is sorted into 4 types:
Figure 592625DEST_PATH_IMAGE123
explanation: that is, points with similar slopes are regarded as one type, and points with large changes in slopes are regarded as the joint portions of the classification, as shown in fig. 3.
And realizing the mapping of the strategy class:
Figure 954467DEST_PATH_IMAGE124
formula 2-2.
3) Demand bandwidth adjustment
According to the formula 2-1,
Figure 104826DEST_PATH_IMAGE125
in the listed order, the bandwidth requirement (user application or administrator plan) for each traffic flow is:
Figure 686373DEST_PATH_IMAGE126
where the bandwidth requirement is the minimum bandwidth required.
Setting the total bandwidth of an outlet of a flow control device as
Figure 947590DEST_PATH_IMAGE127
If it is
Figure 429518DEST_PATH_IMAGE128
Then:
strategy 1:
Figure 750778DEST_PATH_IMAGE129
i.e. to unify the proportion of each class of traffic by sacrificing a certain bandwidth requirement.
Strategy 2: find out
Figure DEST_PATH_IMAGE130
So that:
Figure 613430DEST_PATH_IMAGE131
will be assembled
Figure 429070DEST_PATH_IMAGE133
All the medium elements are assigned to be 0, that is, all the guarantee services with the minimum importance are adopted
Figure DEST_PATH_IMAGE134
And (5) policy guarantee.
4) Bandwidth policy setting
If the method is based on the formula 2-2,
Figure 323427DEST_PATH_IMAGE135
then for the traffic flow
Figure 550009DEST_PATH_IMAGE136
The flow control equipment executes the strategy:
Figure 339105DEST_PATH_IMAGE137
guarantee bandwidth less than bandwidth
Figure DEST_PATH_IMAGE138
If the method is based on the formula 2-2,
Figure 253709DEST_PATH_IMAGE139
then for the traffic flow
Figure 693917DEST_PATH_IMAGE136
The flow control equipment executes the strategy:
Figure DEST_PATH_IMAGE140
guarantee bandwidth equal to fixed bandwidth
Figure 170029DEST_PATH_IMAGE141
If the method is based on the formula 2-2,
Figure DEST_PATH_IMAGE142
then for the traffic flow
Figure 10203DEST_PATH_IMAGE143
The flow control equipment executes the strategy:
Figure 902066DEST_PATH_IMAGE145
i.e. guarantee bandwidth not less than bandwidth
Figure DEST_PATH_IMAGE146
，
Not more than
Figure 39524DEST_PATH_IMAGE147
Explanation is: with the increase of the priority of the service flow, the interval proportion of the bandwidth guarantee is larger.
If the method is based on the formula 2-2,
Figure 607909DEST_PATH_IMAGE148
then aim atTraffic flow
Figure 637176DEST_PATH_IMAGE149
The flow detection and control equipment executes the strategy:
Figure 581998DEST_PATH_IMAGE150
guarantee bandwidth is greater than the set bandwidth
Figure 243137DEST_PATH_IMAGE151
。
The embodiment of the invention provides a multi-dimensional comprehensive flow control system and a flow ordering method, and provides a task-driven multi-granularity comprehensive flow control system and a bandwidth allocation strategy, which provide a feasible scheme for establishing a whole network flow control system for a network management department, and provide a resource allocation method for reasonably utilizing limited bandwidth resources according to the importance and urgency of guarantee of users, tasks and services, so that the whole network flow ordering performance is greatly improved.
The embodiment of the present invention further provides a multidimensional flow comprehensive control system, where the control system adopts the above multidimensional flow comprehensive control ordering method, and the control system includes: an access layer, an HQS plane, an LQS plane and a traffic sensing and scheduling service device; the flow sensing and scheduling service device comprises a flow controller and a flow monitor, wherein the flow controller is used for controlling flow, and the flow monitor is used for monitoring flow.
The access layer is connected with a user, the HQS plane is provided with a plurality of routers and a plurality of flow control devices, and the LQS plane is provided with a plurality of routers and a plurality of flow control devices;
the method comprises the following steps that an HQS plane bears fixed application of a flow model, and an LQS plane bears uncertain application of the flow model;
and the router routes the service flow to the HQS plane and the LQS plane respectively according to the task, the user and the user service level.
It will be evident to those skilled in the art that the embodiments of the present invention are not limited to the details of the foregoing illustrative embodiments, and that the embodiments of the present invention are capable of being embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the embodiments being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. Several units, modules or means recited in the system, apparatus or terminal claims may also be implemented by one and the same unit, module or means in software or hardware.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the embodiments of the present invention and not for limiting, and although the embodiments of the present invention are described in detail with reference to the above preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made on the technical solutions of the embodiments of the present invention without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. A multidimensional flow comprehensive control ordering method is characterized by comprising the following steps:
the first step is as follows: defining related parameters;
the second step is that: dividing plane distribution guarantee dimensions;
the plane refers to an independent IP routing plane and comprises a physical network or a virtual network; dividing the IP routing plane into an HQS plane and an LQS plane;
the third step: dividing service classification guarantee dimensions;
mapping the user service type to a plurality of groups based on a group model, wherein the user service type corresponds to a certain field, applying the field marking strategy to each router input port, and adding a field mark to each router input port message; the HQS plane and the LQS plane are managed by a switch device according to the field flags;
the fourth step: dividing bandwidth allocation guarantee dimensions;
setting flow control equipment at the front end of each router, and controlling the flow and distributing bandwidth resources to each user service according to a certain proportion;
the fifth step: determining a plane offload policy for bandwidth, comprising: shunting bandwidth according to task importance, shunting bandwidth according to user service importance, shunting bandwidth according to multi-dimensional service, and shunting according to network bandwidth occupation and flow distribution conditions;
and a sixth step: and establishing a mapping relation table between the service types and the guarantee types, wherein each service guarantee type corresponds to one value or a plurality of values, and when the network is congested, adjusting the bandwidth according to the mapping relation table.
2. The method for ordering comprehensive management and control of multidimensional traffic according to claim 1, wherein the first step comprises: defining users, user importance, user service importance, user service types, tasks and task importance;
the user refers to a person or a set of persons who directly use various types of applications, and the user is represented by an IP address or a set of IP addresses;
the user importance refers to a set of priority degrees when a user or a user group uses network bandwidth resources;
the user service refers to a set of information systems or services such as data, audio or video which are directly used by a user through an IP network, and the user service is determined according to a TCP/UDP port number;
the user service importance refers to a set of priority degrees of network bandwidth resources occupied by user services;
the user service type is divided into 32 types according to different expected combinations according to four dimensions of time delay sensitivity, packet loss sensitivity, jitter sensitivity and throughput;
the task refers to a role which is assumed by a user to achieve a certain purpose and is represented as a series of work which is carried out by the user by using user service, and the task is a set of specific tasks;
the task importance refers to a set of priorities of work related to the task while occupying bandwidth resources.
3. The method according to claim 2, wherein the user traffic is divided into a plurality of types according to delay sensitivity, packet loss sensitivity, jitter sensitivity, and throughput; service type adoptionS i It is shown that,irepresenting the number of the service types, and representing sensitivity or insensitivity by adopting 0 and 1 for delay sensitivity, packet loss sensitivity and jitter sensitivity, and representing insensitivity by adopting 0 and representing sensitivity by adopting 0; dividing the throughput into four continuous threshold segments from small to large, wherein the first threshold segment is 00, the second threshold segment is 01, the third threshold segment is 10 and the fourth threshold segment is 11; and establishing a user service type table.
4. The method according to claim 1, wherein the shunting bandwidth according to task importance comprises:
the task set is as follows:
Figure 500485DEST_PATH_IMAGE002
sorting by task importance:
Figure 214363DEST_PATH_IMAGE004
defining the total bandwidth as the sum of the egress bandwidths of all routers, expressed as:B S estimating the bandwidth of each task, and recording as:
Figure 56417DEST_PATH_IMAGE006
according to the ordering relation, the bandwidth of each user is processedAnd accumulating and summing one by one to obtain:
Figure 393505DEST_PATH_IMAGE008
when is coming into contact with
Figure 739036DEST_PATH_IMAGE008
Greater than the total bandwidth of the networkB S 70% of, the task
Figure 358236DEST_PATH_IMAGE010
The HQS plane is split and,
task
Figure 953166DEST_PATH_IMAGE012
Shunting to the LQS plane.
5. The method according to claim 1, wherein the splitting the bandwidth according to the importance of the user comprises: the user set is set as follows:
Figure 948803DEST_PATH_IMAGE014
；
ordering by user importance:
Figure 679999DEST_PATH_IMAGE016
defining the total network bandwidth as the sum of the exit bandwidths of all the access routers as follows:
Figure 735680DEST_PATH_IMAGE018
and estimates the bandwidth of each user, recorded as:
Figure 555256DEST_PATH_IMAGE020
(ii) a According to the ordering relation, the bandwidth of each user is accumulated and summed one by one to obtainTo
Figure 354584DEST_PATH_IMAGE022
When is coming into contact with
Figure 940287DEST_PATH_IMAGE023
Greater than the total bandwidth of the network
Figure 291502DEST_PATH_IMAGE017
70% of (1), user
Figure 595445DEST_PATH_IMAGE025
The HQS plane is split and,
user' s
Figure 667306DEST_PATH_IMAGE027
Shunting to the LQS plane.
6. The method according to claim 1, wherein the splitting the bandwidth according to the importance of the user service comprises:
setting the service set as:
Figure 373094DEST_PATH_IMAGE029
normalizing and ordering according to task importance priority:
Figure 502068DEST_PATH_IMAGE031
defining the total network bandwidth as the sum of the exit bandwidths of all the access routers as follows:
Figure 293307DEST_PATH_IMAGE017
and estimating the bandwidth of each user service, and recording as:
Figure 168859DEST_PATH_IMAGE033
and accumulating and summing the service bandwidths of the users one by one according to the ordering relation to obtain:
Figure 729153DEST_PATH_IMAGE021
when is coming into contact with
Figure 422171DEST_PATH_IMAGE021
Greater than the total bandwidth of the network
Figure 435127DEST_PATH_IMAGE017
70% of the time, per user service
Figure 379949DEST_PATH_IMAGE035
The HQS plane is split and,
user service
Figure 797680DEST_PATH_IMAGE037
Shunting to the LQS plane.
7. The ordering method for comprehensive management and control of multidimensional traffic according to claim 1, wherein the shunting bandwidth according to multidimensional service comprises:
firstly, identifying a multidimensional service flow;
tasks, applications, and users are represented as a set of services:
Figure 802545DEST_PATH_IMAGE039
，T m on behalf of a particular task or tasks, the user may,
Figure DEST_PATH_IMAGE040
on behalf of a particular application, it is possible to use,
Figure 99534DEST_PATH_IMAGE041
representing a particular user;
sequencing the service sets, and firstly setting a multi-dimensional service flow set
Figure 582468DEST_PATH_IMAGE043
Then, the importance of a certain service flow is defined as:
Figure 851775DEST_PATH_IMAGE045
normalizing the priority of the service flow to obtain:
Figure 27542DEST_PATH_IMAGE047
，
Figure DEST_PATH_IMAGE048
for the whole of the identifiable traffic flows,
Figure 340056DEST_PATH_IMAGE049
is a specific service flow;
then, sequencing the results after the service flow priority normalization:
Figure 626680DEST_PATH_IMAGE051
Figure 750494DEST_PATH_IMAGE053
is to the service flow
Figure 97162DEST_PATH_IMAGE055
Reordering by importance;
defining the total bandwidth of the network as the sum of the exit bandwidths of all the access routers
Figure 572006DEST_PATH_IMAGE017
Estimating the bandwidth of each user service, and recording as follows:
Figure 662321DEST_PATH_IMAGE057
，
according to the ordering relation, the service bandwidths of all users are accumulated and summed one by one to obtain
Figure 906221DEST_PATH_IMAGE059
When is coming into contact with
Figure 158211DEST_PATH_IMAGE060
Greater than the total bandwidth of the network
Figure DEST_PATH_IMAGE061
70%, multidimensional traffic flow
Figure DEST_PATH_IMAGE063
Then the HQS plane, multi-dimensional traffic is split
Figure DEST_PATH_IMAGE065
Shunting to the LQS plane.
8. A multidimensional traffic comprehensive management and control system, wherein the management and control system adopts the multidimensional traffic comprehensive management and control ordering method according to any one of claims 1 to 7, and the management and control system comprises: an access layer, an HQS plane, an LQS plane and a traffic sensing and scheduling service device;
the access layer is connected with a user, the HQS plane is provided with a plurality of routers and a plurality of flow control devices, and the LQS plane is provided with a plurality of routers and a plurality of flow control devices;
the method comprises the following steps that an HQS plane bears fixed application of a flow model, and an LQS plane bears uncertain application of the flow model;
and the router routes the service flow to the HQS plane and the LQS plane respectively according to the task, the user and the user service level.
CN202110868579.3A 2021-07-30 2021-07-30 Multi-dimensional flow comprehensive control system and flow ordering method Active CN113316261B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110868579.3A CN113316261B (en) 2021-07-30 2021-07-30 Multi-dimensional flow comprehensive control system and flow ordering method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110868579.3A CN113316261B (en) 2021-07-30 2021-07-30 Multi-dimensional flow comprehensive control system and flow ordering method

Publications (2)

Publication Number Publication Date
CN113316261A true CN113316261A (en) 2021-08-27
CN113316261B CN113316261B (en) 2021-10-29

Family

ID=77382178

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110868579.3A Active CN113316261B (en) 2021-07-30 2021-07-30 Multi-dimensional flow comprehensive control system and flow ordering method

Country Status (1)

Country Link
CN (1) CN113316261B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103841052A (en) * 2012-11-27 2014-06-04 中国科学院声学研究所 Bandwidth resource distribution system and method
US20150180791A1 (en) * 2013-12-20 2015-06-25 Avaya, Inc. Adaptive modification of class of service for supporting bandwidth over-allocation
US20150236962A1 (en) * 2014-02-14 2015-08-20 Exinda Networks PTY, Ltd. of Australia Method and system for using dynamic bandwidth detection to drive quality of service control refinement
CN108924058A (en) * 2018-08-30 2018-11-30 中国联合网络通信集团有限公司 Service traffics transmission method and device
CN109547517A (en) * 2017-09-22 2019-03-29 贵州白山云科技股份有限公司 A kind of bandwidth scheduling method and apparatus
CN110213175A (en) * 2019-06-08 2019-09-06 西安电子科技大学 A kind of intelligent managing and control system and management-control method towards knowledge definition network
CN108880904B (en) * 2018-07-05 2020-08-28 中国人民解放军国防科技大学 64-level service quality guarantee method with user and service attribute fusion
CN111917662A (en) * 2019-05-07 2020-11-10 华为技术有限公司 Bandwidth reservation method and related equipment
US20210184894A1 (en) * 2017-12-13 2021-06-17 Telefonaktiebolaget Lm Ericsson (Publ) ACTN Virtual Network Augmentation for Resource Sharing

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103841052A (en) * 2012-11-27 2014-06-04 中国科学院声学研究所 Bandwidth resource distribution system and method
US20150180791A1 (en) * 2013-12-20 2015-06-25 Avaya, Inc. Adaptive modification of class of service for supporting bandwidth over-allocation
US20150236962A1 (en) * 2014-02-14 2015-08-20 Exinda Networks PTY, Ltd. of Australia Method and system for using dynamic bandwidth detection to drive quality of service control refinement
CN109547517A (en) * 2017-09-22 2019-03-29 贵州白山云科技股份有限公司 A kind of bandwidth scheduling method and apparatus
US20210184894A1 (en) * 2017-12-13 2021-06-17 Telefonaktiebolaget Lm Ericsson (Publ) ACTN Virtual Network Augmentation for Resource Sharing
CN108880904B (en) * 2018-07-05 2020-08-28 中国人民解放军国防科技大学 64-level service quality guarantee method with user and service attribute fusion
CN108924058A (en) * 2018-08-30 2018-11-30 中国联合网络通信集团有限公司 Service traffics transmission method and device
CN111917662A (en) * 2019-05-07 2020-11-10 华为技术有限公司 Bandwidth reservation method and related equipment
CN110213175A (en) * 2019-06-08 2019-09-06 西安电子科技大学 A kind of intelligent managing and control system and management-control method towards knowledge definition network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
郭金博: "网络流量管理初探", 《信息与电脑(理论版)》 *
魏永涛等: "《一种新网络体系结构下的流量分配路由机制》", 《小型微型计算机系统》 *

Also Published As

Publication number Publication date
CN113316261B (en) 2021-10-29

Similar Documents

Publication Publication Date Title
CN107770096B (en) SDN/NFV network dynamic resource allocation method based on load balancing
US7272155B2 (en) Central policy manager
US7480304B2 (en) Predictive congestion management in a data communications switch using traffic and system statistics
US9559956B2 (en) Sharing bandwidth among multiple users of network applications
US11595315B2 (en) Quality of service in virtual service networks
US8254252B2 (en) Bandwidth control apparatus
US8630171B2 (en) Policing virtual connections
JP2009543485A (en) Method, apparatus and computer program for controlling distribution of network traffic
EP3318009B1 (en) Model management in a dynamic qos environment
EP3318026B1 (en) Model management in a dynamic qos environment
US7245635B2 (en) System and method for resizing the physical link bandwidth based on utilization thereof
CN109005126B (en) Data stream processing method, device and computer readable storage medium
EP1509008A1 (en) Resource time adaptive internet protocol routing system and method
CN111371690A (en) Flow regulation and control method and device, network equipment and computer readable storage medium
CN109922003B (en) Data sending method, system and related components
CN113316261B (en) Multi-dimensional flow comprehensive control system and flow ordering method
US20090113071A1 (en) Methods and Systems for Providing Efficient Provisioning of Data Flows
CN115580568B (en) Method and system for realizing network service quality guarantee based on IPv6 flow label
US8953450B2 (en) Method and system for ingress multicast load balancing
US8953449B2 (en) Virtual subport data traffic management
CN110753007B (en) QoS-based flow strategy configuration method and device
Elbasheer et al. Video Streaming Adaptive QoS Routing with Resource Reservation (VQoSRR) Model for SDN Networks. Electronics 2022, 11, 1252
CN111585905B (en) Optimal concurrent connection number prediction method suitable for bidirectional coflow scheduling method
US20230138522A1 (en) Queue Bandwidth Estimation for Management of Shared Buffers and Allowing Visibility of Shared Buffer Status
KR20030034871A (en) Apparatus and Method for Buffer Scheduling in Communication System

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