CN103023811B - The switching method and device of queue scheduling strategy - Google Patents
The switching method and device of queue scheduling strategy Download PDFInfo
- Publication number
- CN103023811B CN103023811B CN201110283028.7A CN201110283028A CN103023811B CN 103023811 B CN103023811 B CN 103023811B CN 201110283028 A CN201110283028 A CN 201110283028A CN 103023811 B CN103023811 B CN 103023811B
- Authority
- CN
- China
- Prior art keywords
- queue scheduling
- scheduling strategy
- queue
- switching
- strategy
- 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
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
Abstract
The invention discloses a kind of switching method and device of queue scheduling strategy, this method comprises the following steps：Whether the effective time for detecting current queue scheduling strategy is overtime；In the case of the effective time time-out of current queue scheduling strategy, the switching of queue scheduling strategy is carried out according to default queue scheduling switching mode, wherein, queue scheduling switching mode includes one below：Direct switching mode, serial switching mode, parallel switching mode.The utilization rate and performance of system resource are improved by the present invention.
Description
Technical field
The present invention relates to the communications field, more particularly to a kind of switching method and device of queue scheduling strategy.
Background technology
Generally use the mode of statistic multiplexing come shared resource to reach higher resource utilization in field of data communication.
But meet that different user is wanted to service quality (Quality of Service, referred to as QoS) using limited resource
A kind of effective method must just be provided to solve the problems, such as resource-sharing conflict by asking.In order to solve this problem, at present
There is multiple technologies application.
At present, the service quality of client traffic is substantially to be ensured by following application process：In data access
During to device resource (for example, port), limited using accesses control list (Access Control List, referred to as ACL)
Or allow access of the client to equipment, wherein, accesses control list can be configured based on the time, can be special according to packet
Sign carries out the access of classification control customer service.In order to ensure the priority of client traffic, it is also necessary to by the preferential of client traffic
Level is mapped to the priority of device interior scheduling, so that next step is scheduled to different service queues.Also, provided in equipment
Inside source, in order to avoid concurrency conflict, packet is needed to carry out flow control, it is necessary to carry out speed limit, shaping, team to packet
Row scheduling, discarding etc..Meanwhile conflict can be detected before enqueue is entered, and carry out packet by the strategy of configuration and lose
Abandon.Afterwards, packet will enter queue scheduling process part, be forwarded by corresponding priority.
On the basis of basic queue scheduling, stratification queue scheduling technology also begins to apply.This technology is by client traffic
Different users is divided into according to flow point class, and the user with same characteristic features is formed user's group.Based on user and user's group
Business can be scheduled in a manner of queue.Meanwhile carry the tunnel of customer service and port and also have respective scheduling
Device.This multi-level dispatching technique, specific user can be distinguished and be scheduled so that traffic scheduling is more flexible, has
Effect.
But configured because existing dispatching technique and hiberarchy scheduling technology are not based on the time so that queue
Scheduling lacks flexibility, it is impossible to meets the needs of the customer service scheduling of having time feature.Under certain conditions, network can be provided
Cause to waste in source.For example, user requires that its business has higher priority within some specific period, and it is special at some
Fixed period priority can be with relatively low.In addition, if certain business is configured to forbid accessing in a certain period in ACL,
But relative queue scheduling is due to no time response, the also scheduling bandwidth for its reservation, then, in this case
Queue scheduling does not just utilize Internet resources effectively.It can be seen that in the related art, dispatching technique can only be in acl section realization pair
The access period control of user, queue scheduling lack flexibility, and use and network management to user bring inconvenience.
The content of the invention
The present invention is intended to provide a kind of handover scheme of queue scheduling strategy, at least to solve to dispatch in above-mentioned correlation technique
Technology can not meet the problem of needs of the customer service scheduling with time feature.
To achieve these goals, according to an aspect of the invention, there is provided a kind of switching side of queue scheduling strategy
Method.
According to the switching method of the queue scheduling strategy of the present invention, comprise the following steps：Detect current queue scheduling strategy
Effective time it is whether overtime；In the case of the effective time time-out of current queue scheduling strategy, adjusted according to default queue
The switching that switching mode carries out queue scheduling strategy is spent, wherein, queue scheduling switching mode includes one below：Directly switching side
Formula, serial switching mode, parallel switching mode.
Preferably, detect current queue scheduling strategy effective time whether time-out before, this method also includes：To participate in
The object configuration queue scheduling strategy of traffic scheduling, and effective time is configured for the queue scheduling strategy, wherein, the queue scheduling
Strategy comprises at least one below：Queue identity, queue scheduling algorithm, dispatch weight, drop policy.
Preferably, include to participate in the object of traffic scheduling configuration queue scheduling strategy：According to flow point Class Type to from
The business packet of service port is classified, and by sorted business packet in the way of user or user's group with specific industry
Port binding be engaged in together；Queue scheduling strategy is configured to participate in the object of traffic scheduling, wherein, participate in the object of traffic scheduling
Including at least one below：Service port, tunnel, pseudo-wire, user, user's group.
Preferably, the switching of queue scheduling strategy is carried out according to default queue scheduling switching mode to be included：Judge current
Whether the queue scheduling algorithm of queue scheduling strategy is consistent with the queue scheduling algorithm for the queue scheduling strategy to be switched to；If one
Cause, then the switching of queue scheduling strategy is carried out using direct switching mode；Otherwise, using serial switching mode or parallel switching side
Formula carries out the switching of queue scheduling strategy.
Preferably, the switching of queue scheduling strategy is carried out using direct switching mode to be included：Use the queue to be switched to
Scheduling strategy replaces current queue scheduling strategy, wherein, in the queue scheduling strategy to be switched to and current queue scheduling strategy
The value of parameter is different.
Preferably, the switching of queue scheduling strategy is carried out using serial switching mode to be included：Current queue scheduling strategy
Queue scheduling algorithm no longer receives new packet, and the queue scheduling algorithm for the queue scheduling strategy to be switched to receives new number
According to bag；Current queue scheduling strategy fails, and by according to the insertion of default priority orders and the queue scheduling plan to be switched to
In priority query corresponding to queue scheduling algorithm slightly.
Preferably, the switching of queue scheduling strategy is carried out using parallel switching mode to be included：Current queue scheduling strategy
Queue scheduling algorithm no longer receives new packet, and the queue scheduling algorithm for the queue scheduling strategy to be switched to receives new number
According to bag；Current queue scheduling strategy is survival；Until the queue to be dispatched of queue scheduling strategy to be switched to is all
After data dispatch is completed, current queue scheduling strategy fails again.
Preferably, detect current queue scheduling strategy effective time whether time-out after, this method also includes：Current
In the case that the effective time of queue scheduling strategy has not timed out, whether the effective time for continuing to detect current queue scheduling strategy surpasses
When.
Preferably, after the switching that queue scheduling strategy is carried out according to default queue scheduling switching mode, this method is also
Including：If handoff failure, recover using switching before queue scheduling strategy, and extend the queue scheduling strategy it is effective when
Between.
To achieve these goals, according to another aspect of the present invention, a kind of switching of queue scheduling strategy is additionally provided
Device.
According to the switching device of the queue scheduling strategy of the present invention, including：Detection module, for detecting current queue scheduling
Whether the effective time of strategy is overtime；Handover module, for detection module detect current queue scheduling strategy it is effective when
Between in the case of time-out, the switching of queue scheduling strategy is carried out according to default queue scheduling switching mode, wherein, queue scheduling
Switching mode includes one below：Direct switching mode, serial switching mode, parallel switching mode.
By the present invention, using the effective time of detection current queue scheduling strategy, in the effective time time-out to team
The mode that row scheduling strategy switches over so that the customer service dispatch contact of queue scheduling and special time gets up, and solves
Dispatching technique can not meet the problem of needs of the customer service scheduling with time feature in correlation technique, improve system money
The utilization rate and performance in source.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, forms the part of the application, this hair
Bright schematic description and description is used to explain the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings：
Fig. 1 is the flow chart of the switching method of queue scheduling strategy according to embodiments of the present invention；
Fig. 2 is the structured flowchart of the switching device of queue scheduling strategy according to embodiments of the present invention；
Fig. 3 is the implementation process figure of according to embodiments of the present invention two array dispatching method；
Fig. 4 is the schematic diagram of according to embodiments of the present invention three queue scheduling device.
Embodiment
Describe the present invention in detail below with reference to accompanying drawing and in conjunction with the embodiments.It should be noted that do not conflicting
In the case of, the feature in embodiment and embodiment in the application can be mutually combined.
According to embodiments of the present invention, there is provided a kind of switching method of queue scheduling strategy.Fig. 1 is implemented according to the present invention
The flow chart of the switching method of the queue scheduling strategy of example, as shown in figure 1, comprising the following steps：
Whether step S102, the effective time for detecting current queue scheduling strategy are overtime；
Step S104, in the case of the effective time time-out of current queue scheduling strategy, according to default queue scheduling
Switching mode carries out the switching of queue scheduling strategy, wherein, queue scheduling switching mode includes one below：Directly switching side
Formula, serial switching mode, parallel switching mode.
By above-mentioned steps, using the effective time of detection current queue scheduling strategy, in the effective time time-out pair
The mode that queue scheduling strategy switches over so that the customer service dispatch contact of queue scheduling and special time gets up, and solves
Dispatching technique can not meet the problem of needs of the customer service scheduling with time feature in correlation technique, improve system
The utilization rate and performance of resource.
Preferably, before step S102, queue scheduling strategy is configured to participate in the object of traffic scheduling, and be the queue
Scheduling strategy configures effective time, wherein, the queue scheduling strategy comprises at least one below：Queue identity, queue scheduling are calculated
Method, dispatch weight, drop policy.For example, the object for participating in traffic scheduling can be, but not limited to：Port, tunnel, pseudo-wire, user,
User's group.This method can improve the flexibility of system.
In implementation process, the dispatch weight in queue scheduling strategy can be from dispatching algorithm weight and bandwidth scheduling weight
In select one.For example, the selection scheduling algorithm weights when carrying out flow queue scheduling, bandwidth scheduling weight is selected in queue scheduling.
Preferably, can include to participate in the object of traffic scheduling configuration queue scheduling strategy：According to flow point Class Type pair
Business packet from service port is classified, and by sorted business packet in the way of user or user's group with it is specific
Service port bind together (for example, classified according to flow point Class Type to the business packet from service port, according to point
Class creates user, and sets user's group according to the user of establishment, and user's group is tied into port)；To participate in pair of traffic scheduling
As configuring queue scheduling strategy, wherein, the object for participating in traffic scheduling comprises at least one below：Service port, tunnel, puppet
Line, user, user's group.Wherein, service port can be ethernet port or aggregation port etc..This method causes ACL processing units
The stream of classification and the stream of queue scheduling part classifying, which combine, realizes time-based scheduling controlling, improves system
Disposal ability.
For example, flow point Class Type can be：Internal layer VLAN (Virtual Local Area Network, referred to as
For VLAN), outside VLAN, source IP address, place IP address, source medium access control (Media Access Control, referred to as
MAC) address, place MAC Address, differential operation code point (Differential Services Code Point, are referred to as
DSCP), 802.1p (that is, IEEE 802.1p agreements, 802.1p protocol headers include 3 precedence fields), EXP be (i.e.,
Experiment, be the precedence field in MPLS headings) etc. information.
Preferably, in step S104, the switching of queue scheduling strategy is carried out according to default queue scheduling switching mode
It can include：Judge the queue scheduling algorithm of current queue scheduling strategy and the queue scheduling for the queue scheduling strategy to be switched to
Whether algorithm is consistent；If consistent, direct switching mode can be used, which to carry out the switching of queue scheduling strategy, (needs what is illustrated
It is one of method that directly switching can simply select, in implementation process, according to circumstances user can also select parallel or string
Row switching mode)；Otherwise, serial switching mode or parallel switching mode can only be used to carry out the switching of queue scheduling strategy.Should
Method can select different switching modes according to different situations, improve the flexibility of system.
Preferably, in step S104, the switching of queue scheduling strategy is carried out using direct switching mode to be included：Make
With the queue scheduling policy replacement current queue scheduling strategy to be switched to, wherein, the queue scheduling strategy to be switched to is with working as
The value of parameter is different in preceding queue scheduling strategy.This method can improve the accuracy of system.
For example, the queue scheduling strategy to be switched to use with current queue scheduling strategy identical queue scheduling algorithm,
And the parameter of the object of participation traffic scheduling and type are also identical in two strategies, such as, adjusted in two strategies containing bandwidth
Weight is spent, the bandwidth scheduling weight in the queue scheduling strategy to be simply switched to is 40%, in current queue scheduling strategy
Bandwidth scheduling weight is 50%.
Preferably, in step S104, the switching of queue scheduling strategy is carried out using serial switching mode to be included：When
The queue scheduling algorithm of preceding queue scheduling strategy no longer receives new packet, and the queue for the queue scheduling strategy to be switched to is adjusted
Spend algorithm and receive new packet；Current queue scheduling strategy fails, and will be inserted according to default priority orders with to cut
In priority query corresponding to the queue scheduling algorithm for the queue scheduling strategy changed to.This method is only a kind of within the same time
Queue scheduling strategy comes into force, and has ensured the orderly progress of system.
Preferably, in step S104, the switching of queue scheduling strategy is carried out using parallel switching mode to be included：When
The queue scheduling algorithm of preceding queue scheduling strategy no longer receives new packet, and the queue for the queue scheduling strategy to be switched to is adjusted
Spend algorithm and receive new packet；Current queue scheduling strategy is survival；Until the queue scheduling strategy to be switched to is wanted
After all data dispatch of the queue of scheduling are completed, current queue scheduling strategy fails again.This method switching initial stage, two
Strategy can come into force, and ensure the validity and integrality of system.
Preferably, after step s 102, in the case where the effective time of current queue scheduling strategy has not timed out, continue
Whether the effective time for detecting current queue scheduling strategy is overtime.This method is simple and practical, workable.
Preferably, after step s 104, if handoff failure, recover using the queue scheduling strategy before switching, and
Extend the effective time of the queue scheduling strategy.This method ensure that the stability of system, the performance of system is improved.
Corresponding to the above method, the embodiment of the present invention additionally provides a kind of switching device of queue scheduling strategy.Fig. 2 is root
According to the structured flowchart of the switching device of the queue scheduling strategy of the embodiment of the present invention, as shown in Fig. 2 the device includes：Detect mould
Whether block 22, the effective time for detecting current queue scheduling strategy are overtime；Handover module 24, coupled to detection module 22,
In the case of effective time time-out for detecting current queue scheduling strategy in detection module 22, adjusted according to default queue
The switching that switching mode carries out queue scheduling strategy is spent, wherein, queue scheduling switching mode includes one below：Directly switching side
Formula, serial switching mode, parallel switching mode.
By said apparatus, detection module 22 detects the effective time of current queue scheduling strategy, and handover module 24 is at this
Queue scheduling strategy is switched over during effective time time-out so that the customer service dispatch contact of queue scheduling and special time
Get up, solve the problems, such as that dispatching technique can not meet the needs of the customer service scheduling with time feature in correlation technique,
Improve the utilization rate and performance of system resource.
The implementation to technical scheme is described in further detail below in conjunction with the accompanying drawings：
Embodiment one
It can not be configured, can not be met with time spy based on the time in order to overcome queue scheduling of the prior art
The shortcomings that customer service scheduling of point needs, solves the problems, such as network resources waste present in prior art, the present embodiment carries
A kind of array dispatching method is supplied.
In implementation process, the array dispatching method can include：First, in interface, the industry distinguished according to packet
Business creates user, different users is formed into user's group, and user's group is tied into interface；Secondly, configuration schedules queue and its
Corresponding scheduling strategy, it is each scheduling queue configuration schedules effective time.Wherein, user here refers to there is specific flow point class
A kind of flow point class of feature, user's group refer to the set of one group of user, and scheduling effective time refers to the scheduling strategy effective time
Scope.
Preferably, during queue scheduling, the effective time of queue scheduling strategy is detected.If current queue
Scheduling strategy effective time mistake, then into queue scheduling strategy switching state, carry out queue scheduling strategy switching.Adjusted in queue
Tactful switching state is spent, directly switching, serial switching or parallel switching can be carried out according to the switchover policy of setting.
Preferably, the setting of switchover policy can be according to the dispatching algorithm of former scheduling strategy and the algorithm of new scheduling strategy
Whether unanimously determine, if for example, consistent, can directly be switched, otherwise, carry out serially switching or parallel switching.
For example, under direct switching mode, original scheduling strategy is replaced using the scheduling strategy newly to come into force.Serially cutting
Change under mode, the queue of current scheduling algorithm no longer receives new packet, and new data packets will enter the tune of new scheduling strategy
Queue is spent, the queue of original failure dispatching algorithm enters priority query corresponding with new algorithm.Under parallel switching mode, when
The queue of preceding dispatching algorithm no longer receives new packet, and new data packets will enter the scheduling queue of new scheduling strategy, and former
Have that scheduling strategy is survival, until all data dispatch of the queue of scheduling are completed, original scheduling strategy failure, i.e. herein
Period, original scheduling strategy and the scheduling strategy newly to come into force while come into force.
Preferably, queue scheduling strategy switches successfully, and currently valid scheduling strategy is revised as new scheduling strategy.If
Handoff failure, reporting system queue scheduling handoff failure are dispatched, and recovers original scheduling strategy, original scheduling strategy is effective
Time is extended by the time specified.
It should be noted that when the effective time mistake of queue scheduling strategy, other efficient scheduling plans can also be selected
Slightly.In the case of in the absence of efficient scheduling strategy, it can also select to give tacit consent to scheduling strategy.
By the present embodiment, time progress queue scheduling strategy switching can be based on so that the scheduling of queue is more flexible,
Time-based differentiation queue scheduling demand is met, saves bandwidth, in the case of avoiding hiberarchy scheduling, forbids accessing
User improves bandwidth availability ratio to the occupancy of scheduling bandwidth.
Embodiment two
Fig. 3 is the implementation process figure of according to embodiments of the present invention two array dispatching method, as shown in figure 3, the present embodiment
In array dispatching method may comprise steps of：
Step S301, in ethernet port or other service ports (for example, aggregation port etc.), according to the flow point of business packet
Class Type is classified business packet, for example, flow point Class Type can be：Inner VLAN, outside VLAN, source IP address, place IP
The information such as address, source MAC, place MAC Address, DSCP, 802.1p, EXP.Based on this classification, service-user is created, also
It is the client with a kind of type of service.Multiple users are by can form user's group.Multiple user's groups can be tied to specifically
Service port.
Step S302, User Network Interface (User-to-Network Interface, referred to as UNI) side, for user
Business, internal system can be mapped to by its characteristic type different spikelets per panicles (Per-Hop Behavior, referred to as
PHB) type.The business of user is scheduled based on PHB types.For the business of a user, the bandwidth scheduling specified is distributed
Weight.It is also required to configure the dispatch weight of its bandwidth for a user's group.In Network-Node Interface (Network Node
Interface, referred to as NNI) side, the tunnel of pseudo-wire and carrying pseudo-line needs the dispatch weight of nominated bandwidth.Every one-level is joined
With the object of queue scheduling, it is required for configuring its scheduling strategy, wherein, queue scheduling strategy can include：Dispatching algorithm, scheduling
Algorithm weights or bandwidth scheduling weight, drop policy etc..
Step S303, it is the queue scheduling strategy configuration schedules entry-into-force time.Each traffic scheduling object is (for example, port, tunnel
Road, pseudo-wire, user, user's group) multiple queue scheduling strategies can be configured simultaneously.As long as the entry-into-force time of these scheduling strategies
Do not clash can.Each scheduling strategy is effective only within its entry-into-force time.
Step S304, whether queue scheduling control process periodically detects current scheduling strategy effective.If exceed
Entry-into-force time, then the scheduling strategy up to the entry-into-force time is chosen in other scheduling strategies of each traffic scheduling object configuration.
Policy replacement is carried out according to the replacement policy mode of configuration simultaneously.If queue scheduling strategy entry-into-force time mistake, according to strategy
Switching mode goes to step S305, step S306, step S307 respectively；Otherwise continue to detect.
Step S305, during using direct switching mode replacement policy, directly replaced using the scheduling strategy newly to come into force original
Scheduling strategy.Such case is all identical suitable for two tactful parameters and type, simply the different situation of the value of parameter.
For example the dispatch weight of former strategy is 30%, the dispatch weight of new strategy is 60%, and at this moment directly weight parameter can be replaced
Past.
Step S306, during using serial switchover policy, the queue of current scheduling algorithm no longer receives new packet, new number
The scheduling queue of new scheduling strategy will be entered according to bag.The queue of original failure dispatching algorithm enters corresponding with new algorithm preferential
Level queue.The only scheduling strategy that this mode comes into force simultaneously, failure scheduling strategy adjust the data-moving of queue to new
Degree strategy.
Step S307, during using parallel switchover policy, the queue of current scheduling algorithm no longer receives new packet, new number
The scheduling queue of new scheduling strategy will be entered according to bag.Original scheduling strategy is survival, until all numbers of the queue of scheduling
Completed according to scheduling, original scheduling strategy failure.This mode has two scheduling queues to come into force before switching is completed.
Step S308, if tactful handoff failure, reporting system queue scheduling handoff failure, and recover original scheduling plan
Slightly, original scheduling strategy entry-into-force time is extended by the time specified.Go to step S304.
Embodiment three
Fig. 4 is the schematic diagram of according to embodiments of the present invention three queue scheduling device, as shown in figure 4, the queue scheduling fills
Put including：Enqueue control unit, queue scheduling control unit, queue scheduling strategy storage unit and queue scheduling unit.With
Under each unit is described in detail.
Enqueue control unit, for controlling Congestion Avoidance, control data to enter what which dispatch control unit was controlled
Scheduling queue.For example, when using serial switching or parallel switching mode switching queue scheduling strategy, enqueue control unit is born
Duty control data stream turns to the queue of the queue scheduling policy control newly to come into force.Wherein, queue scheduling strategy comprises at least queue
Mark, queue scheduling algorithm, dispatch weight, drop policy.
Queue scheduling control unit, for from queue scheduling strategy storage unit obtain queue scheduling strategy and its it is effective when
Between data, for regularly detecting queue scheduling strategy validity；Queue scheduling unit is created during system initialization, or needs to adopt
During with parallel switching queue scheduling mode, the queue scheduling unit newly to come into force is created；Timing detects the queue scheduling of current application
The effective time of the queue scheduling strategy of unit, judge its validity；Control enqueue control unit transmits data to specified
Queue；Perform the switching of queue scheduling strategy.Under parallel switching queue scheduling strategy mode, nullify and failed and completed switching
Queue scheduling unit.
Queue scheduling strategy storage unit, for being responsible for the storage and retrieval of queue scheduling strategy and effective time data.
That is, queue scheduling strategy storage unit also preserves the effective time of queue scheduling strategy in addition to preserving queue scheduling strategy.
Queue scheduling unit, dispatched for specific queuing data.That is, application queue scheduling strategy, takes out from queue
Data are transmitted.For example, queue scheduling unit performs specific queue scheduling algorithm, when using direct switching mode progress team
During row switching, queue scheduling unit directly replaces failed queue scheduling algorithm weight using new dispatching algorithm weight configuration
Configuration.
In summary, the present embodiments relate to a kind of time-based array dispatching method in field of data communication and
System, stream and the stream of queue scheduling part classifying that ACL process parts are classified are combined and carry out time-based control, solution
Dispatching technique can not meet to have the problem of needs of customer service scheduling of time feature in correlation technique of having determined, and improve and be
The utilization rate and performance for resource of uniting.
Obviously, those skilled in the art should be understood that above-mentioned each module of the invention or each step can be with general
Computing device realize that they can be concentrated on single computing device, or be distributed in multiple computing devices and formed
Network on, alternatively, they can be realized with the program code that computing device can perform, so as to being stored in
Performed in storage device by computing device, and in some cases, can be with different from shown by order execution herein
Or the step of description, they are either fabricated to each integrated circuit modules respectively or by the multiple modules or step in them
Suddenly single integrated circuit module is fabricated to realize.So, the present invention is not restricted to any specific hardware and software combination.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies
Change, equivalent substitution, improvement etc., should be included in the scope of the protection.
Claims (6)
1. a kind of switching method of queue scheduling strategy, it is characterised in that comprise the following steps：
Whether the effective time for detecting current queue scheduling strategy is overtime；
In the case of the effective time time-out of the current queue scheduling strategy, entered according to default queue scheduling switching mode
The switching of row queue scheduling strategy, wherein, the queue scheduling switching mode includes one below：It is direct switching mode, serial
Switching mode, parallel switching mode；
Wherein, the switching for carrying out queue scheduling strategy according to the default queue scheduling switching mode includes：Judge described work as
Whether the queue scheduling algorithm of preceding queue scheduling strategy is consistent with the queue scheduling algorithm for the queue scheduling strategy to be switched to；If
Unanimously, then the switching of queue scheduling strategy is carried out using the directly switching mode；Otherwise, using the serial switching mode or
The parallel switching mode carries out the switching of queue scheduling strategy；
Wherein, the switching for carrying out queue scheduling strategy using the directly switching mode includes：Use the team to be switched to
Row scheduling strategy replaces the current queue scheduling strategy, wherein, the queue scheduling strategy to be switched to it is described current
The value of parameter is different in queue scheduling strategy；
Wherein, the switching for carrying out queue scheduling strategy using the serial switching mode includes：The current queue scheduling strategy
Queue scheduling algorithm no longer receive new packet, the queue scheduling algorithm of the queue scheduling strategy to be switched to receives
The new packet；The current queue scheduling strategy failure, and will insert according to default priority orders and be wanted with described
In priority query corresponding to the queue scheduling algorithm for the queue scheduling strategy being switched to；
Wherein, the switching for carrying out queue scheduling strategy using the parallel switching mode includes：The current queue scheduling strategy
Queue scheduling algorithm no longer receive new packet, the queue scheduling algorithm of the queue scheduling strategy to be switched to receives
The new packet；The current queue scheduling strategy is survival；Until the queue scheduling strategy institute to be switched to
After all data dispatch of the queue to be dispatched are completed, the current queue scheduling strategy fails again.
2. according to the method for claim 1, it is characterised in that detecting the effective time of the current queue scheduling strategy is
Before no time-out, in addition to：
Queue scheduling strategy is configured to participate in the object of traffic scheduling, and effective time is configured for the queue scheduling strategy, wherein,
The queue scheduling strategy comprises at least one below：Queue identity, queue scheduling algorithm, dispatch weight, drop policy.
3. according to the method for claim 2, it is characterised in that the object for the participation traffic scheduling configures the queue
Scheduling strategy includes：
The business packet from service port is classified according to flow point Class Type, and by sorted business packet according to user or
The mode of person's user's group is bound together with the specific service port；
Object for the participation traffic scheduling configures the queue scheduling strategy, wherein, the object for participating in traffic scheduling
Including at least one below：The service port, tunnel, pseudo-wire, the user, the user's group.
4. according to the method for claim 1, it is characterised in that detecting the effective time of the current queue scheduling strategy is
After no time-out, in addition to：
In the case where the effective time of the current queue scheduling strategy has not timed out, continue to detect the current queue scheduling plan
Whether effective time slightly is overtime.
5. method according to any one of claim 1 to 4, it is characterised in that cut according to the default queue scheduling
After the mode of changing carries out the switching of queue scheduling strategy, in addition to：
If handoff failure, recover using switching before queue scheduling strategy, and extend the queue scheduling strategy it is effective when
Between.
A kind of 6. switching device of queue scheduling strategy, it is characterised in that including：
Whether detection module, the effective time for detecting current queue scheduling strategy are overtime；
Handover module, the situation of the effective time time-out for detecting the current queue scheduling strategy in the detection module
Under, the switching of queue scheduling strategy is carried out according to default queue scheduling switching mode, wherein, the queue scheduling switching mode
Including one below：Direct switching mode, serial switching mode, parallel switching mode；
Wherein, the handover module is additionally operable to, wherein, judge the queue scheduling algorithm of the current queue scheduling strategy with to cut
Whether the queue scheduling algorithm for the queue scheduling strategy changed to is consistent；If consistent, team is carried out using the directly switching mode
The switching of row scheduling strategy；Otherwise, queue scheduling strategy is carried out using the serial switching mode or the parallel switching mode
Switching；
Wherein, the switching for carrying out queue scheduling strategy using the directly switching mode includes：Use the team to be switched to
Row scheduling strategy replaces the current queue scheduling strategy, wherein, the queue scheduling strategy to be switched to it is described current
The value of parameter is different in queue scheduling strategy；
Wherein, the switching for carrying out queue scheduling strategy using the serial switching mode includes：The current queue scheduling strategy
Queue scheduling algorithm no longer receive new packet, the queue scheduling algorithm of the queue scheduling strategy to be switched to receives
The new packet；The current queue scheduling strategy failure, and will insert according to default priority orders and be wanted with described
In priority query corresponding to the queue scheduling algorithm for the queue scheduling strategy being switched to；
Wherein, the switching for carrying out queue scheduling strategy using the parallel switching mode includes：The current queue scheduling strategy
Queue scheduling algorithm no longer receive new packet, the queue scheduling algorithm of the queue scheduling strategy to be switched to receives
The new packet；The current queue scheduling strategy is survival；Until the queue scheduling strategy institute to be switched to
After all data dispatch of the queue to be dispatched are completed, the current queue scheduling strategy fails again.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110283028.7A CN103023811B (en) | 2011-09-21 | 2011-09-21 | The switching method and device of queue scheduling strategy |
| PCT/CN2011/082116 WO2012151893A1 (en) | 2011-09-21 | 2011-11-11 | Method and device for switching queue scheduling policies |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110283028.7A CN103023811B (en) | 2011-09-21 | 2011-09-21 | The switching method and device of queue scheduling strategy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103023811A CN103023811A (en) | 2013-04-03 |
| CN103023811B true CN103023811B (en) | 2018-04-10 |
Family
ID=47138694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110283028.7A Active CN103023811B (en) | 2011-09-21 | 2011-09-21 | The switching method and device of queue scheduling strategy |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN103023811B (en) |
| WO (1) | WO2012151893A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111600738B (en) * | 2020-03-30 | 2022-12-09 | 福建天泉教育科技有限公司 | Method for optimizing timeout processing and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6157654A (en) * | 1997-06-24 | 2000-12-05 | Alcatel Networks Corporation | Adaptive service weight assignments for ATM scheduling |
| US6757897B1 (en) * | 2000-02-29 | 2004-06-29 | Cisco Technology, Inc. | Apparatus and methods for scheduling and performing tasks |
| EP1545090A1 (en) * | 2003-12-19 | 2005-06-22 | Broadcom Corporation | Transmit scheduling method for multi-layer network interface card (NIC) operation |
| CN101120612A (en) * | 2005-02-16 | 2008-02-06 | 松下电器产业株式会社 | Radio base station, control apparatus, and wireless communication method |
| CN101123541A (en) * | 2007-09-29 | 2008-02-13 | 华中科技大学 | A construction method applied to policy model of legal monitoring system |
| CN101222284A (en) * | 2008-02-03 | 2008-07-16 | 北京创毅视讯科技有限公司 | Scheduling system and method for authorized management information in mobile multimedia broadcast |
| CN101547079A (en) * | 2008-03-24 | 2009-09-30 | 大唐移动通信设备有限公司 | Method and equipment for discontinuously receiving data and system and equipment for discontinuously dispatching data |
| CN101742497A (en) * | 2009-12-24 | 2010-06-16 | 中兴通讯股份有限公司 | Method for realizing access authentication and client |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101610198B (en) * | 2008-06-17 | 2012-03-28 | 大唐移动通信设备有限公司 | Method and device for scheduling packet services |
-
2011
- 2011-09-21 CN CN201110283028.7A patent/CN103023811B/en active Active
- 2011-11-11 WO PCT/CN2011/082116 patent/WO2012151893A1/en active Application Filing
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6157654A (en) * | 1997-06-24 | 2000-12-05 | Alcatel Networks Corporation | Adaptive service weight assignments for ATM scheduling |
| US6757897B1 (en) * | 2000-02-29 | 2004-06-29 | Cisco Technology, Inc. | Apparatus and methods for scheduling and performing tasks |
| EP1545090A1 (en) * | 2003-12-19 | 2005-06-22 | Broadcom Corporation | Transmit scheduling method for multi-layer network interface card (NIC) operation |
| CN101120612A (en) * | 2005-02-16 | 2008-02-06 | 松下电器产业株式会社 | Radio base station, control apparatus, and wireless communication method |
| CN101123541A (en) * | 2007-09-29 | 2008-02-13 | 华中科技大学 | A construction method applied to policy model of legal monitoring system |
| CN101222284A (en) * | 2008-02-03 | 2008-07-16 | 北京创毅视讯科技有限公司 | Scheduling system and method for authorized management information in mobile multimedia broadcast |
| CN101547079A (en) * | 2008-03-24 | 2009-09-30 | 大唐移动通信设备有限公司 | Method and equipment for discontinuously receiving data and system and equipment for discontinuously dispatching data |
| CN101742497A (en) * | 2009-12-24 | 2010-06-16 | 中兴通讯股份有限公司 | Method for realizing access authentication and client |
Non-Patent Citations (1)
| Title |
|---|
| 基于P2P技术的视频点播系统中的数据调度策略;额尔敦巴日斯等;《内蒙古民族大学学报》;20080315(第02期);第19-20页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103023811A (en) | 2013-04-03 |
| WO2012151893A1 (en) | 2012-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10148492B2 (en) | Data center bridging network configuration and management | |
| US11588733B2 (en) | Slice-based routing | |
| US10044627B2 (en) | QoS on a virtual interface over multi-path transport | |
| EP2984784B1 (en) | System and method for providing a software defined protocol stack | |
| CA2963580C (en) | Data forwarding method, device, and system in software-defined networking | |
| CN103959724B (en) | Mix networking path selection and load balance | |
| US7903552B2 (en) | Directional and priority based flow control mechanism between nodes | |
| CN102740503B (en) | Data link layer protocol unit, and relative apparatus and method | |
| CN105122747A (en) | Control device and control method in software defined network (sdn) | |
| CN102075437B (en) | Communication method, gateway and network | |
| CN103534989A (en) | Priority based flow control in a distributed fabric protocol (DFP) switching network architecture | |
| CN101841487A (en) | Configuration method for aggregating link service flow and packet switching device | |
| CN102132511A (en) | Virtual switch quality of service for virtual machines | |
| WO2018036516A1 (en) | Method and system for traffic management, packet switching device and user device | |
| CN113746751A (en) | Communication method and device | |
| CN102546395B (en) | Business scheduling method and device based on L2VPN networks | |
| CN106330710B (en) | Data stream scheduling method and device | |
| US20160007366A1 (en) | Method and device for transmitting enhanced transmission selection standard configuration information | |
| CN110447207A (en) | The system and method for reaction equation Path selection | |
| CN103312628A (en) | Scheduling method and device for aggregated links in packet switched network | |
| CN103023811B (en) | The switching method and device of queue scheduling strategy | |
| US9166815B2 (en) | Communication apparatus and communication controlling method in ring network | |
| Polishchuk et al. | Improving TCP-friendliness and Fairness for mHIP | |
| CN103973590B (en) | A kind of QoS dispatching methods and device | |
| CN104092622B (en) | A kind of message processing method and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |