CN106330710B - Data stream scheduling method and device - Google Patents
Data stream scheduling method and device Download PDFInfo
- Publication number
- CN106330710B CN106330710B CN201510379415.9A CN201510379415A CN106330710B CN 106330710 B CN106330710 B CN 106330710B CN 201510379415 A CN201510379415 A CN 201510379415A CN 106330710 B CN106330710 B CN 106330710B
- Authority
- CN
- China
- Prior art keywords
- scheduling
- level
- enabled
- node
- data stream
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
Abstract
The invention discloses a kind of data stream scheduling method, the data stream scheduling method includes: to determine that default port needs enabled scheduling level according to current scheduling parameter;Make the corresponding scheduling node of scheduling level that can determine that, and establishes the linking relationship between different scheduling level scheduling nodes;For enabled each scheduling node configuration schedules strategy;When the default port receives data flow, according to the linking relationship between the scheduling strategy and each scheduling node of the configuration of each scheduling node, data flow layering is scheduled between enabled each scheduling node.The invention also discloses a kind of data stream scheduling devices.The present invention can be improved the efficiency of scheduling data flow.
Description
Technical field
The present invention relates to field of communication technology more particularly to data stream scheduling methods and device.
Background technique
With the rapid development of the communication technology, so as to bandwidth, delay, the voice of jitter-sensitive, video and data etc.
Business more and more passes through network transmission.In order to provide data transmission performance different promise and guarantee, widely
QoS (Quality of Service, service quality) technology has been used to guarantee network transmission quality.However, as user advises
The expansion of mould, type of business increase, it is desirable that the network equipment can not only further refine discriminating service flow, but also can
Scheduling is managed collectively and is layered for connection objects such as multiple users, multiple business and various flow rates.This is for tradition
QoS technology for, be difficult to realize.In order to achieve the purpose that layering scheduling, HQoS (Hierarchical Quality
Of Service is layered QoS) technology comes into being.But existing HQoS technology scheduling level is fixed, it cannot be according to user
Demand carries out flexible layering scheduling, and the efficiency for dispatching data flow is lower.
Summary of the invention
The main purpose of the present invention is to provide a kind of data stream scheduling method and devices, it is intended to improve scheduling data flow
Efficiency.
To achieve the above object, the present invention provides a kind of data stream scheduling method, and the data stream scheduling method includes:
Determine that default port needs enabled scheduling level according to current scheduling parameter;
Make the corresponding scheduling node of scheduling level that can determine that, and establishes the link between different scheduling level scheduling nodes
Relationship;
For enabled each scheduling node configuration schedules strategy;
When the default port receives data flow, the scheduling strategy and each scheduling node that are configured according to each scheduling node
Between linking relationship, data flow layering is scheduled between enabled each scheduling node.
Preferably, the default port includes a generic port and two generic ports, when the default port is a generic port,
The default port includes 5 grades of scheduling levels;When the default port is two generic port, the default port includes 3 grades of tune
Spend level.
Preferably, described to determine that the enabled scheduling level of default port needs includes: according to current scheduling parameter
Always using the minimum and highest scheduling level of the default Port Scheduling priority as the dispatch layer for needing to enable
Grade, and according to current scheduling parameter, determine what the default port needed to enable, except the default Port Scheduling priority is minimum
With other scheduling levels except highest scheduling level.
Preferably, described to include: for enabled each scheduling node configuration schedules strategy
For enabled each scheduling node configuration schedules algorithm, traffic shaping strategy and/or congestion avoidance policy.
It is preferably, described to be determined before presetting the scheduling level that port needs to enable according to current scheduling parameter, further includes:
Receive the scheduling parameter that user is inputted based on scheduling parameter set interface;
Using the scheduling parameter received as current scheduling parameter.
In addition, to achieve the above object, the present invention also provides a kind of data stream scheduling device, the data stream scheduling device
Include:
Determining module, for determining that default port needs enabled scheduling level according to current scheduling parameter;
Enabled module for making the corresponding scheduling node for the scheduling level that can determine that, and establishes different scheduling level scheduling
Linking relationship between node;
Configuration module, for each scheduling node configuration schedules strategy to enable;
Scheduler module, the scheduling plan for being configured according to each scheduling node when the default port receives data flow
Linking relationship between summary and each scheduling node is scheduled data flow layering between enabled each scheduling node.
Preferably, the default port includes a generic port and two generic ports, when the default port is a generic port,
The default port includes 5 grades of scheduling levels;When the default port is two generic port, the default port includes 3 grades of tune
Spend level.
Preferably, the determining module is also used to the default minimum and highest scheduling of Port Scheduling priority always
Level determines what the default port needed to enable, removes institute as the scheduling level for needing to enable, and according to current scheduling parameter
State other scheduling levels except the minimum and highest scheduling level of default Port Scheduling priority.
Preferably, the configuration module is also used to as enabled each scheduling node configuration schedules algorithm, traffic shaping strategy
And/or congestion avoidance policy.
Preferably, the data stream scheduling device further includes setup module, is arranged for receiving user based on scheduling parameter
The scheduling parameter of interface input;And using the scheduling parameter received as current scheduling parameter.
For the present invention on the basis of existing HQoS technology, flexible choice needs enabled scheduling level, and enables selection
The corresponding scheduling node of level is dispatched, each scheduling node is according to the linking relationship phase between the scheduling strategy and scheduling node of configuration
The layering scheduling of data flow is realized in mutually cooperation.Solve the problems, such as that level is dispatched in existing HQoS technology fixes, and more meets user
Demand, and unnecessary process resource waste is avoided, so that the present invention can be improved the efficiency of scheduling data flow.
Detailed description of the invention
Fig. 1 is the flow diagram of data stream scheduling method preferred embodiment of the present invention;
Fig. 2 is the networking schematic diagram in data stream scheduling method preferred embodiment of the present invention;
Fig. 3 is a kind of scheduling level schematic diagram in data stream scheduling method preferred embodiment of the present invention;
Fig. 4 is another scheduling level schematic diagram in data stream scheduling method preferred embodiment of the present invention;
Fig. 5 is another scheduling level schematic diagram in data stream scheduling method preferred embodiment of the present invention;
Fig. 6 is the functional block diagram of data stream scheduling device preferred embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The present invention provides a kind of data stream scheduling method, referring to Fig.1, in the preferable implementation of data stream scheduling method of the present invention
In example, the data stream scheduling method includes:
S10 determines that default port needs enabled scheduling level according to current scheduling parameter;
S20 makes the corresponding scheduling node of scheduling level that can determine that, and establishes between different scheduling level scheduling nodes
Linking relationship;
Data stream scheduling method provided in this embodiment can be applied to MPLS (Multi-Protocol Label
Switching, multiprotocol label switching) network, for example, in an mpls network, when certain port of certain network node receives number
When according to stream, layering scheduling is carried out to the data flow according to preconfigured scheduling level and corresponding scheduling strategy, efficiently
Stream compression is sent out.It is illustrated below with data stream scheduling method provided by the invention applied to MPLS network.
It should be noted that MPLS is the IP high-speed backbone network exchange standard of a new generation, by IETF (Internet
Engineering Task Force, internet engineering task group) it proposes.MPLS independently of the second layer and third layer protocol,
Data forwarding is carried out using Label (label): when data flow enters network node, the label of regular length is distributed for it, and
By label together with data stream, in entire repeating process, network node is forwarded according only to label.Wherein, number
Occur on LSP (label switched path) according to transmission, LSP is the network node on each path along source to terminal
Sequence label.
The data stream scheduling method that the present embodiment proposes divides different scheduling levels according to the characteristic of MPLS in advance, and
Determine that default port needs enabled scheduling level according to current scheduling parameter;And make to can determine that according to current scheduling parameter
The corresponding scheduling node of level is dispatched, and establishes the linking relationship between different scheduling level scheduling nodes.Wherein, the scheduling
Parameter includes but is not limited to need enabled scheduling hierarchical information, scheduling node information, scheduling strategy information and different scheduling
Linking relationship information between level scheduling node.
S30, for enabled each scheduling node configuration schedules strategy;
After scheduling node is enabled, need for enabled each scheduling node configuration schedules strategy.Wherein, different scheduling levels
Or identical scheduling strategy can be configured between the scheduling node of identical scheduling level, different scheduling plans can also be configured
Slightly, it is configured with specific reference to the scheduling strategy information in scheduling parameter.
S40, when the default port receives data flow, according to the scheduling strategy of each scheduling node configuration and each scheduling
Linking relationship between node is scheduled data flow layering between enabled each scheduling node.
It should be noted that each scheduling node for different scheduling levels is mounted with corresponding scheduling respectively in the present embodiment
The scheduler of level realizes that the layering of data flow is dispatched by the scheduler of different scheduling levels.Specifically, according between scheduler
Scheduling hierarchical relationship (linking relationship between difference scheduling level scheduling nodes) and corresponding scheduling strategy, higher level's scheduling
Device controls the total bandwidth of one group of junior's scheduler, and according to the promise of the rank of junior's scheduler and weight distribution junior scheduler
Bandwidth and peak bandwidth realize the layering scheduling of data flow.
The data stream scheduling method that the present embodiment proposes, on the basis of existing HQoS technology, flexible choice needs enabled
Scheduling level, and the corresponding scheduling node of scheduling level of enabled selection, each scheduling node according to the scheduling strategy of configuration with
And the linking relationship between scheduling node cooperates, and realizes the layering scheduling of data flow.It solves and is dispatched in existing HQoS technology
The fixed problem of level, more meets user demand, and avoids unnecessary process resource waste, so that the present invention can be improved
Dispatch the efficiency of data flow.
Further, in the preferred embodiment of data stream scheduling method of the present invention, the default port includes a kind of end
Mouth and two generic ports, when the default port is a generic port, the default port includes 5 grades of scheduling levels;When described pre-
If port is two generic port, the default port includes 3 grades of scheduling levels.
It should be noted that in an mpls network, including two generic ports, referring to Fig. 2, by the network node phase in present networks
Port even is known as NNI (network node port), and the port that present networks are connected with other networks is known as (the user network end UNI
Mouthful).The present embodiment divides the scheduling level of different series for different types of port according to the network structure of MPLS.
Specifically, when the default port is a generic port (NNI), the default port includes 5 grades of scheduling levels, i.e.,
There are 5 grades of scheduling levels for NNI division, is respectively as follows: port, VLAN subport, tunnel, pseudo-wire and queue, respectively dispatches level
Including one or more scheduling nodes.For example, referring to Fig. 3, for NNI, the scheduling node of all 5 grades of scheduling levels is enabled;
Referring to Fig. 4, for NNI, enable port, tunnel, pseudo-wire and queue scheduling level scheduling node.
When the default port is two generic ports (UNI), the default port includes 3 grades of scheduling levels, that is, is directed to UNI
Division has 3 grades of scheduling levels, be respectively as follows: port, CIP (Client Computer's IP address, client address) with
And queue, each level of dispatching includes one or more scheduling nodes.For example, referring to Fig. 5, for UNI, all 3 grades of tune are enabled
Spend the scheduling node of level.
The present embodiment divides different scheduling levels, accordingly by being directed to different types of port to meet different nets
Network network environment.
Further, described true according to current scheduling parameter in the preferred embodiment of data stream scheduling method of the present invention
Surely presetting the enabled scheduling level of port needs includes:
Always using the minimum and highest scheduling level of the default Port Scheduling priority as the dispatch layer for needing to enable
Grade, and according to current scheduling parameter, determine what the default port needed to enable, except the default Port Scheduling priority is minimum
With other scheduling levels except highest scheduling level.
It should be noted that even if being dispatched using common QoS, the scheduling of Port Scheduling level and queue scheduling level
Be it is enabled always, in the present embodiment, also need the scheduling of the scheduling of enable port always level and queue scheduling level, i.e.,
Always by the minimum scheduling level (Port Scheduling level) of the default Port Scheduling priority and the highest tune of dispatching priority
Level (queue scheduling level) is spent as the scheduling level for needing to enable.Aforementioned always by Port Scheduling level and queue scheduling
On the basis of level is as enabled scheduling level is needed, according to current scheduling parameter, it is enabled to determine that the default port needs
, other scheduling levels in addition to the minimum and highest scheduling level of the default Port Scheduling priority.
For example, NNI is directed to, always using Port Scheduling level and queue scheduling level as the dispatch layer for needing to enable
Grade, and according to current scheduling parameter, dynamic select vlan sub-interface dispatches level, tunnel scheduling level and pseudo-wire and dispatches level
As the scheduling level for needing to enable;It is always that Port Scheduling level and queue scheduling level is enabled as needing for UNI
Scheduling level, and according to current scheduling parameter, whether dynamic select using CIP scheduling level as the dispatch layer for needing to enable
Grade.
Further, described for enabled each scheduling node in the preferred embodiment of data stream scheduling method of the present invention
Configuration schedules strategy includes:
For enabled each scheduling node configuration schedules algorithm, traffic shaping strategy and/or congestion avoidance policy.
In the present embodiment, the scheduling strategy information in scheduling parameter includes but is not limited to dispatching algorithm, traffic shaping strategy
And congestion avoidance policy, described includes configuring for each scheduling node enabled for enabled each scheduling node configuration schedules strategy
Dispatching algorithm, traffic shaping strategy and/or congestion avoidance policy.
Specifically, for the scheduling node of different scheduling levels, the dispatching algorithm that can be configured includes but is not limited to SP
(Strict Priority, stringent priority), RR (Round Robin, round-robin scheduling), DRR (Deficit Round
Robin, Deficit Round Robin), WRR (Weighted Round Robin, weighted round-robin scheduling), WDRR (Weighted
Deficit Round Robin, deficit weighted round robin scheduling) and DWRR (Dynamic Weighted Round Robin,
One of dynamic weighting round-robin scheduling).For example, referring to Fig. 3, the scheduling node of level can be dispatched for vlan sub-interface
Null_vlan configures SP algorithm, and the scheduling node null_lsp for dispatching level for tunnel configures SP algorithm, dispatches level for tunnel
Scheduling node Lsp_1 configure DWRR algorithm.
In order to avoid the data flow that the default port transfers is more than the bearing capacity of network, it is preferable that the present embodiment
Traffic shaping strategy is configured in the scheduling node of each scheduling level, carries out traffic shaping.Traffic shaping strategy is using token bucket
Assessment and peak load shifting are carried out to data stream, peak-data stream is cached in shaping queue, delay retransmits, and guarantees as far as possible
Message (data flow is made of message) is not lost.
Congestion avoidance policy is only configured in the scheduling node of queue scheduling level, and Congestion Avoidance is a kind of flow-control mechanism,
Its service condition cached by monitoring queue, when congestion has the tendency that aggravation, active dropping packets, by adjusting data flow
Amount is to release network over loading.The method that traditional congestion avoidance policy uses tail drop (Tail-Drop), when the length of queue
After reaching a certain maximum value, the message of all new arrivals will be all dropped, and Tail-Drop will cause TCP global synchronization phenomenon.
Preferably, to avoid TCP global synchronization phenomenon, the present embodiment uses WRED (Weighted Random Early
Detection, Weighted random early detection) carry out Congestion Avoidance.Loss priority of the WRED based on message is each queue
Upper and lower bound is set, the message in queue is handled as follows:
(1), when the length of queue is less than lower limit, not dropping packets;
(2), when the length of queue is more than the upper limit, the message of all arrivals is abandoned;
(3), when the length of queue is between upper and lower bound, the message of random drop arrival.Queue is longer, discarding
Probability is higher, but has a maximum drop probability.
Further, described true according to current scheduling parameter in the preferred embodiment of data stream scheduling method of the present invention
Surely default port needs before enabled scheduling level, further includes:
Receive the scheduling parameter that user is inputted based on scheduling parameter set interface;
Using the scheduling parameter received as current scheduling parameter.
Specifically, CLI (the Command Line that user can be provided by network node according to actual needs
Interface, Command Line Interface) and/or GUI (Graphical User Interface, graphic user interface) scheduling parameter
Set interface carries out the setting of default Port Scheduling parameter.For example, when judgement needs larger scheduling dynamics to be just able to satisfy data flow
When dispatching requirement, user can enable the scheduling level of more series by the way that corresponding scheduling parameter is arranged, to meet data flow tune
Degree demand, it is ensured that the normal transmitting data flow of network node.
Further, user can also according to actual needs be updated the scheduling parameter of setting, so that network node
Adapt to continually changing network environment.
The present invention also provides a kind of data stream scheduling devices, referring to Fig. 6, in the preferable reality of data stream scheduling device of the present invention
It applies in example, the data stream scheduling device includes:
Determining module 10, for determining that default port needs enabled scheduling level according to current scheduling parameter;
Enabled module 20, for making the corresponding scheduling node for the scheduling level that can determine that, and establishes different scheduling level tune
Spend the linking relationship between node;
Data stream scheduling device provided in this embodiment can be applied to MPLS (Multi-Protocol Label
Switching, multiprotocol label switching) network, for example, in an mpls network, data stream scheduling device is built in network node
Operation, when certain port of the network node receives data flow, data stream scheduling device is according to preconfigured scheduling level
And corresponding scheduling strategy carries out layering scheduling to the data flow, efficiently sends out stream compression.Below with this hair
The data stream scheduling device of bright offer is illustrated applied to MPLS network.
It should be noted that MPLS is the IP high-speed backbone network exchange standard of a new generation, by IETF (Internet
Engineering Task Force, internet engineering task group) it proposes.MPLS independently of the second layer and third layer protocol,
Data forwarding is carried out using Label (label): when data flow enters network node, the label of regular length is distributed for it, and
By label together with data stream, in entire repeating process, network node is forwarded according only to label.Wherein, number
Occur on LSP (label switched path) according to transmission, LSP is the network node on each path along source to terminal
Sequence label.
The data stream scheduling device that the present embodiment proposes divides different scheduling levels according to the characteristic of MPLS in advance, and
Determine that default port needs enabled scheduling level according to current scheduling parameter by determining module 10;And it is pressed by enabled module 20
According to the corresponding scheduling node for the scheduling level that current scheduling parameter makes can determine that, and establish between different scheduling level scheduling nodes
Linking relationship.Wherein, the scheduling parameter include but is not limited to need enabled scheduling hierarchical information, scheduling node information,
Linking relationship information between scheduling strategy information and different scheduling level scheduling nodes.
Configuration module 30, for each scheduling node configuration schedules strategy to enable;
After scheduling node is enabled, needing to configure module 30 is enabled each scheduling node configuration schedules strategy.Wherein, no
With identical scheduling strategy can be configured between the scheduling node of scheduling level or identical scheduling level, difference can also be configured
Scheduling strategy, specifically configured by configuration module 30 according to the scheduling strategy information in scheduling parameter.
Scheduler module 40, the scheduling for being configured according to each scheduling node when the default port receives data flow
Linking relationship between tactful and each scheduling node adjusts data flow layering between enabled each scheduling node
Degree.
It should be noted that scheduler module 40 is that each scheduling node of different scheduling levels distinguishes carry in the present embodiment
There is the scheduler of corresponding scheduling level, realizes that the layering of data flow is dispatched by the scheduler of different scheduling levels.Specifically, according to
Scheduling hierarchical relationship (linking relationship between difference scheduling level scheduling nodes) between scheduler and dispatch plan accordingly
Slightly, higher level's scheduler controls the total bandwidth of one group of junior's scheduler, and according to the rank of junior's scheduler and weight distribution junior
The guaranteed bandwidth and peak bandwidth of scheduler realize the layering scheduling of data flow.
The data stream scheduling device that the present embodiment proposes, on the basis of existing HQoS technology, flexible choice needs enabled
Scheduling level, and the corresponding scheduling node of scheduling level of enabled selection, each scheduling node according to the scheduling strategy of configuration with
And the linking relationship between scheduling node cooperates, and realizes the layering scheduling of data flow.It solves and is dispatched in existing HQoS technology
The fixed problem of level, more meets user demand, and avoids unnecessary process resource waste, so that the present invention can be improved
Dispatch the efficiency of data flow.
Further, in the preferred embodiment of data stream scheduling device of the present invention, the default port includes a kind of end
Mouth and two generic ports, when the default port is a generic port, the default port includes 5 grades of scheduling levels;When described pre-
If port is two generic port, the default port includes 3 grades of scheduling levels.
It should be noted that in an mpls network, including two generic ports, referring to Fig. 2, by the network node phase in present networks
Port even is known as NNI (network node port), and the port that present networks are connected with other networks is known as (the user network end UNI
Mouthful).In the present embodiment, the data stream scheduling device further includes division module, for the network structure according to MPLS, for difference
The port of type divides the scheduling level of different series.
Specifically, when the default port is a generic port (NNI), the default port includes 5 grades of scheduling levels, i.e.,
Division module has 5 grades of scheduling levels for NNI division, is respectively as follows: port, VLAN subport, tunnel, pseudo-wire and queue, respectively
Dispatching level includes one or more scheduling nodes.For example, referring to Fig. 3, for NNI, enabled module 20 enables division module
The scheduling node of all 5 grades of scheduling levels of division;Referring to Fig. 4, for NNI, enabled module 20 enables division module division
Port, tunnel, pseudo-wire and queue scheduling level scheduling node.
When the default port is two generic ports (UNI), the default port includes 3 grades of scheduling levels, i.e. division mould
Block has 3 grades of scheduling levels for UNI division, is respectively as follows: port, CIP (Client Computer's IP address, client
Hold address) and queue, each level of dispatching includes one or more scheduling nodes.Mould is enabled for UNI for example, referring to Fig. 5
Block 20 enables the scheduling node of all 3 grades of scheduling levels of division module division.
The present embodiment divides different scheduling levels, accordingly by being directed to different types of port to meet different nets
Network network environment.
Further, in the preferred embodiment of data stream scheduling device of the present invention, the determining module 10 is also used to begin
The minimum and highest scheduling level of the default Port Scheduling priority is as the scheduling level for needing to enable at last, and according to working as
Preceding scheduling parameter determines what the default port needed to enable, removes the default minimum and highest tune of Port Scheduling priority
Spend other scheduling levels except level.
It should be noted that even if being dispatched using common QoS, the scheduling of Port Scheduling level and queue scheduling level
Be it is enabled always, in the present embodiment, also need the scheduling of the scheduling of enable port always level and queue scheduling level, i.e.,
Determining module 10 is always by the minimum scheduling level (Port Scheduling level) of the default Port Scheduling priority and dispatching priority
Highest scheduling level (queue scheduling level) of grade is as the scheduling level for needing to enable.Aforementioned always by Port Scheduling level
With queue scheduling level on the basis of the scheduling level enabled as needs, determining module 10 is determined according to current scheduling parameter
What the default port needed to enable, it is other in addition to the minimum and highest scheduling level of the default Port Scheduling priority
Dispatch level.
For example, being directed to NNI, determining module 10 is enabled as needing using Port Scheduling level and queue scheduling level always
Scheduling level, and according to current scheduling parameter, dynamic select vlan sub-interface dispatches level, tunnel scheduling level and pseudo-wire
Level is dispatched as the scheduling level for needing to enable;For UNI, determining module 10 is always by Port Scheduling level and queue tune
Spend level as needing enabled scheduling level, and according to current scheduling parameter, dynamic select whether using CIP dispatch level as
Need enabled scheduling level.
Further, in the preferred embodiment of data stream scheduling device of the present invention, the configuration module 30 be also used to for
Enabled each scheduling node configuration schedules algorithm, traffic shaping strategy and/or congestion avoidance policy.
In the present embodiment, the scheduling strategy information in scheduling parameter includes but is not limited to dispatching algorithm, traffic shaping strategy
And congestion avoidance policy, the configuration module 30 are also used to as enabled each scheduling node configuration schedules algorithm, traffic shaping
Strategy and/or congestion avoidance policy.
Specifically, for it is different scheduling levels scheduling nodes, the dispatching algorithm that configuration module 30 can configure include but
It is not limited to SP (Strict Priority, stringent priority), RR (Round Robin, round-robin scheduling), DRR (Deficit
Round Robin, Deficit Round Robin), WRR (Weighted Round Robin, weighted round-robin scheduling), WDRR
(Weighted Deficit Round Robin, deficit weighted round robin scheduling) and DWRR (Dynamic Weighted
One of Round Robin, dynamic weighting round-robin scheduling).For example, referring to Fig. 3, configuration module 30 can be vlan sub-interface
The scheduling node null_vlan for dispatching level configures SP algorithm, and the scheduling node null_lsp for dispatching level for tunnel configures SP
Algorithm, the scheduling node Lsp_1 for dispatching level for tunnel configure DWRR algorithm.
In order to avoid the data flow that the default port transfers is more than the bearing capacity of network, it is preferable that configuration module
30 configure traffic shaping strategy in each scheduling node for dispatching level, carry out traffic shaping.Traffic shaping strategy is using token
Bucket carries out assessment and peak load shifting to data stream, peak-data stream is cached in shaping queue, delay retransmits, and protects as far as possible
Card message (data flow is made of message) is not lost.
Congestion avoidance policy is only configured in the scheduling node of queue scheduling level, and Congestion Avoidance is a kind of flow-control mechanism,
Its service condition cached by monitoring queue, when congestion has the tendency that aggravation, active dropping packets, by adjusting data flow
Amount is to release network over loading.The method that traditional congestion avoidance policy uses tail drop (Tail-Drop), when the length of queue
After reaching a certain maximum value, the message of all new arrivals will be all dropped, and Tail-Drop will cause TCP global synchronization phenomenon.
Preferably, to avoid TCP global synchronization phenomenon, the present embodiment uses WRED (Weighted Random Early
Detection, Weighted random early detection) carry out Congestion Avoidance.Loss priority of the WRED based on message is each queue
Upper and lower bound is set, the message in queue is handled as follows:
(1), when the length of queue is less than lower limit, not dropping packets;
(2), when the length of queue is more than the upper limit, the message of all arrivals is abandoned;
(3), when the length of queue is between upper and lower bound, the message of random drop arrival.Queue is longer, discarding
Probability is higher, but has a maximum drop probability.
Further, in the preferred embodiment of data stream scheduling device of the present invention, the data stream scheduling device is also wrapped
Setup module is included, the scheduling parameter inputted for receiving user based on scheduling parameter set interface;And described in receiving
Scheduling parameter is as current scheduling parameter.
Specifically, the CLI (Command that user can be provided by setup module based on network node according to actual needs
Line Interface, Command Line Interface) and/or GUI (Graphical User Interface, graphic user interface) scheduling
Parameter setting interface carries out the setting of default Port Scheduling parameter.For example, when judgement needs larger scheduling dynamics to be just able to satisfy number
When according to stream dispatching requirement, user can enable the scheduling level of more series by the way that corresponding scheduling parameter is arranged, to meet data
Flow dispatching requirement, it is ensured that the normal transmitting data flow of network node.
Further, user can also according to actual needs be updated the scheduling parameter of setting, so that network node
Adapt to continually changing network environment.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of data stream scheduling method, which is characterized in that the data stream scheduling method includes:
Determine that default port needs enabled scheduling level according to current scheduling parameter, wherein the scheduling parameter includes at least
It needs between enabled scheduling hierarchical information, scheduling node information, scheduling strategy information and different scheduling level scheduling nodes
Linking relationship information;
Make the corresponding scheduling node of scheduling level that can determine that, and the link established between different scheduling level scheduling nodes is closed
System;
For enabled each scheduling node configuration schedules strategy;
When the default port receives data flow, according between the scheduling strategy and each scheduling node of the configuration of each scheduling node
Linking relationship, data flow layering is scheduled between enabled each scheduling node.
2. data stream scheduling method as described in claim 1, which is characterized in that the default port includes a generic port and two
Generic port, when the default port is a generic port, the default port includes 5 grades of scheduling levels;When the default port
When for two generic ports, the default port includes 3 grades of scheduling levels.
3. data stream scheduling method as claimed in claim 2, which is characterized in that described determine according to current scheduling parameter is preset
Port needs the enabled scheduling level to include:
Always using the minimum and highest scheduling level of the default Port Scheduling priority as the scheduling level for needing to enable, and
According to current scheduling parameter, it is enabled to determine that the default port needs, except the default Port Scheduling priority is minimum and most
Other scheduling levels except high scheduling level.
4. data stream scheduling method as described in any one of claims 1-3, which is characterized in that described for enabled each scheduling section
Putting configuration schedules strategy includes:
For enabled each scheduling node configuration schedules algorithm, traffic shaping strategy and/or congestion avoidance policy.
5. data stream scheduling method as described in any one of claims 1-3, which is characterized in that described according to current scheduling parameter
Before determining that default port needs enabled scheduling level, further includes:
Receive the scheduling parameter that user is inputted based on scheduling parameter set interface;
Using the scheduling parameter received as current scheduling parameter.
6. a kind of data stream scheduling device, which is characterized in that the data stream scheduling device includes:
Determining module, for determining that default port needs enabled scheduling level according to current scheduling parameter, wherein the scheduling
Parameter, which includes at least, needs enabled scheduling hierarchical information, scheduling node information, scheduling strategy information and different scheduling levels
Linking relationship information between scheduling node;
Enabled module, for making the corresponding scheduling node for the scheduling level that can determine that, and establishes different scheduling level scheduling nodes
Between linking relationship;
Configuration module, for each scheduling node configuration schedules strategy to enable;
Scheduler module, for when the default port receives data flow, according to the scheduling strategy of each scheduling node configuration and
Linking relationship between each scheduling node is scheduled data flow layering between enabled each scheduling node.
7. data stream scheduling device as claimed in claim 6, which is characterized in that the default port includes a generic port and two
Generic port, when the default port is a generic port, the default port includes 5 grades of scheduling levels;When the default port
When for two generic ports, the default port includes 3 grades of scheduling levels.
8. data stream scheduling device as claimed in claim 7, which is characterized in that the determining module is also used to always will be described
The minimum and highest scheduling level of default Port Scheduling priority is joined as the scheduling level for needing to enable according to current scheduling
It is enabled to determine that the default port needs for number, except the minimum and highest scheduling level of the default Port Scheduling priority it
Outer other scheduling levels.
9. such as the described in any item data stream scheduling devices of claim 6-8, which is characterized in that the configuration module be also used to for
Enabled each scheduling node configuration schedules algorithm, traffic shaping strategy and/or congestion avoidance policy.
10. such as the described in any item data stream scheduling devices of claim 6-8, which is characterized in that the data stream scheduling device
It further include setup module, the scheduling parameter inputted for receiving user based on scheduling parameter set interface;And it will receive
The scheduling parameter is as current scheduling parameter.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510379415.9A CN106330710B (en) | 2015-07-01 | 2015-07-01 | Data stream scheduling method and device |
PCT/CN2016/075256 WO2016177058A1 (en) | 2015-07-01 | 2016-03-01 | Method and apparatus for data stream scheduling |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510379415.9A CN106330710B (en) | 2015-07-01 | 2015-07-01 | Data stream scheduling method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106330710A CN106330710A (en) | 2017-01-11 |
CN106330710B true CN106330710B (en) | 2019-09-13 |
Family
ID=57217486
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510379415.9A Active CN106330710B (en) | 2015-07-01 | 2015-07-01 | Data stream scheduling method and device |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106330710B (en) |
WO (1) | WO2016177058A1 (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109840145B (en) * | 2019-01-08 | 2021-09-07 | 苏州盛科通信股份有限公司 | Multistage scheduling method, device, network chip and storage medium |
CN113810314A (en) * | 2020-06-15 | 2021-12-17 | 中兴通讯股份有限公司 | Data scheduling method, device and storage medium |
CN113949675A (en) * | 2020-07-16 | 2022-01-18 | 华为技术有限公司 | Queue scheduling method, device and system |
CN115118677A (en) * | 2022-06-24 | 2022-09-27 | 无锡中微亿芯有限公司 | Routing node scheduling method of network on chip in FPGA |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7379467B1 (en) * | 2003-05-08 | 2008-05-27 | Cypress Semiconductor Corporation | Scheduling store-forwarding of back-to-back multi-channel packet fragments |
CN101478475A (en) * | 2008-11-21 | 2009-07-08 | 中兴通讯股份有限公司 | Method for implementing HQoS technique in T-MPLS network |
CN102546395A (en) * | 2011-12-14 | 2012-07-04 | 中兴通讯股份有限公司 | Service scheduling method and service scheduling device based on Layer 2 Virtual Private Networks (L2VPN) |
-
2015
- 2015-07-01 CN CN201510379415.9A patent/CN106330710B/en active Active
-
2016
- 2016-03-01 WO PCT/CN2016/075256 patent/WO2016177058A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7379467B1 (en) * | 2003-05-08 | 2008-05-27 | Cypress Semiconductor Corporation | Scheduling store-forwarding of back-to-back multi-channel packet fragments |
CN101478475A (en) * | 2008-11-21 | 2009-07-08 | 中兴通讯股份有限公司 | Method for implementing HQoS technique in T-MPLS network |
CN102546395A (en) * | 2011-12-14 | 2012-07-04 | 中兴通讯股份有限公司 | Service scheduling method and service scheduling device based on Layer 2 Virtual Private Networks (L2VPN) |
Also Published As
Publication number | Publication date |
---|---|
WO2016177058A1 (en) | 2016-11-10 |
CN106330710A (en) | 2017-01-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101843239B1 (en) | System and method for providing a software defined protocol stack | |
US10044627B2 (en) | QoS on a virtual interface over multi-path transport | |
US8705363B2 (en) | Packet scheduling method and apparatus | |
CN101692648B (en) | Method and system for queue scheduling | |
EP2302843A1 (en) | Method and device for packet scheduling | |
CN108353029A (en) | For managing the method and system for calculating the data service in network | |
CN106330710B (en) | Data stream scheduling method and device | |
CN105490962B (en) | A kind of QoS management methods based on OpenFlow networks | |
CN109391563B (en) | Resource reservation method and related equipment | |
JP4163044B2 (en) | BAND CONTROL METHOD AND BAND CONTROL DEVICE THEREOF | |
CN110557333A (en) | method and system for controlling and guaranteeing quality of service of software defined network | |
CN102413051B (en) | Method and device for scheduling quality of service (QOS) | |
CN112005528B (en) | Data exchange method, data exchange node and data center network | |
CN101478486B (en) | Method, equipment and system for switch network data scheduling | |
CN110247859B (en) | Data sending method and device based on AIRT-ROS node | |
CN105099947A (en) | Space network access method and device | |
CN100531203C (en) | Method for ensuring service quality in multi protocol label switching protocol | |
CN110417656A (en) | A kind of QoS guarantee system and method based on SDN network | |
TW200303670A (en) | Inverse multiplexing of managed traffic flows over a multi-star network | |
KR101587379B1 (en) | Method of dynamic control for queue size and apparatus thereof | |
CN103326962B (en) | Diversified service switching method | |
US9166924B2 (en) | Packet scheduling method and apparatus considering virtual port | |
CN103023811B (en) | The switching method and device of queue scheduling strategy | |
Joung et al. | Scalable Flow Isolation with Work Conserving Stateless Core Fair Queuing for Deterministic Networking | |
Chen et al. | Improve the problem of insufficient bandwidth and path congestion in the cloud computing environment by using self-adjustment scheduler |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20190807 Address after: Yuhuatai District of Nanjing City, Jiangsu province 210012 Bauhinia Road No. 68 Applicant after: Nanjing Zhongxing New Software Co., Ltd. Address before: 518057 Nanshan District Guangdong high tech Industrial Park, South Road, science and technology, ZTE building, Ministry of Justice Applicant before: ZTE Corporation |
|
GR01 | Patent grant | ||
GR01 | Patent grant |