CN102694740B - A kind of array dispatching method and device - Google Patents
A kind of array dispatching method and device Download PDFInfo
- Publication number
- CN102694740B CN102694740B CN201210217512.4A CN201210217512A CN102694740B CN 102694740 B CN102694740 B CN 102694740B CN 201210217512 A CN201210217512 A CN 201210217512A CN 102694740 B CN102694740 B CN 102694740B
- Authority
- CN
- China
- Prior art keywords
- scheduler object
- state
- scheduling
- priority
- scheduler
- 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
Abstract
This application discloses a kind of array dispatching method, the method comprises: each scheduler object is assigned to corresponding scheduling queue by the first priority of status according to each described scheduler object; When detecting that the flow of arbitrary described scheduler object is greater than CIR and is less than MIR, this scheduler object is switched to corresponding scheduling queue by the second priority of status according to this scheduler object; When detecting that the flow of arbitrary described scheduler object is not less than MIR, this scheduler object is switched out current place scheduling queue; According to the absolute priority of each scheduling queue, each scheduler object on scheduling queue is dispatched, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue.Based on same inventive concept, also propose a kind of device, different scheduling strategies can be realized by the different scheduling parameter of configuration, improve dispatching flexibility.
Description
Technical field
The application relates to communication technical field, particularly a kind of array dispatching method and device.
Background technology
In data communication network network, because upstream and downstream device bandwidth is inconsistent, easily occur congested in data transmission procedure, in order to ensure that the QoS demand of different business in network just needs to dispatch in equipment exit the flow of different business.
The relevant dispatching method that the dispatching algorithm usually used at present has SP, RR algorithm and derives from based on this.SP can realize priority scheduling, but Low Priority Queuing may be caused to be died of hunger by high-priority queue.RR can realize the priority that proportionally still cannot ensure queue to multiple queue scheduling.In order to realize scheduling strategy more flexibly, SP and RR combines by some realizations, and same priority carries out RR scheduling with regard to queue, and SP scheduling is carried out in the queue of different priorities.This needs for the multi-service flow scheduling of reality the scheduling strategy flexibly that provides to a certain extent, but still can not meet some complicated and diversified dispatching requirement completely.In fact dispatching requirement is completely different and changeable in different applications, although can meet the flexibility of specific application shortage for often kind of application scenarios customizes specific dispatching algorithm.
Summary of the invention
In view of this, the application provides a kind of array dispatching method and device, can realize different scheduling strategies, improve dispatching flexibility by the different scheduling parameter of configuration.
For solving the problems of the technologies described above, technical scheme of the present invention is achieved in that
A kind of array dispatching method, be each scheduler object configuration schedules parameter according to scheduling strategy, wherein, described scheduling parameter comprises promised information rate, maximum rate MIR, priority of status and state quota, described priority of status comprises the first priority of status and the second priority of status, and described method comprises:
Each scheduler object is assigned to corresponding scheduling queue by the first priority of status according to each described scheduler object; When detecting that the flow of arbitrary described scheduler object is greater than CIR and is less than MIR, this scheduler object is switched to corresponding scheduling queue by the second priority of status according to this scheduler object; When detecting that the flow of arbitrary described scheduler object is not less than MIR, this scheduler object is switched out current place scheduling queue, wherein, the queue number of each scheduling queue is corresponding with the value of the priority of status of scheduler object;
According to the absolute priority of each scheduling queue, each scheduler object on scheduling queue is dispatched, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue.
A kind of device, this device comprises: dispensing unit, detecting unit, allocation units and scheduling unit;
Described dispensing unit, for according to scheduling strategy being each scheduler object configuration schedules parameter, wherein, described scheduling parameter comprises promised information rate, maximum rate MIR, priority of status and state quota, and described priority of status comprises the first priority of status and the second priority of status;
Described detecting unit, for detecting the flow of each scheduler object;
Described allocation units, each scheduler object is assigned to corresponding scheduling queue by the first priority of status for each described scheduler object configured according to described dispensing unit; When described detecting unit detects that the flow of arbitrary described scheduler object is greater than CIR and is less than MIR, this scheduler object is switched to corresponding scheduling queue by the second priority of status of this scheduler object configured according to described dispensing unit; When described detecting unit detects that the flow of arbitrary described scheduler object is not less than MIR, this scheduler object is switched out current place scheduling queue, wherein, the queue number of each scheduling queue is corresponding with the value of the priority of status of scheduler object;
Described scheduling unit, for the absolute priority according to each scheduling queue, each scheduler object that allocation units described on scheduling queue distribute is dispatched, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue.
In sum, the each scheduler object configuration schedules parameter of the application by according to scheduling strategy being scheduling, the scheduling parameter of configuration comprises double priority level, and according to the relation of the current flow of scheduler object and CIR and MIR, determine this scheduler object according to that priority assign to corresponding scheduling queue, each scheduler object on scheduling queue is dispatched according to the absolute priority of each scheduling queue when dispatching, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue, different scheduling strategies can be realized by the different scheduling parameter of configuration, improve dispatching flexibility.
Accompanying drawing explanation
Fig. 1 is array dispatching method schematic flow sheet in the embodiment of the present invention;
Fig. 2 is be the schematic diagram of each scheduler object allocation schedule queue according to the scheduling parameter of each scheduler object;
Fig. 3 be the flow rate of scheduler object S1 when being greater than CIR each scheduler object at the distribution condition schematic diagram of scheduling queue;
Fig. 4 be the flow rate of scheduler object S2 when being greater than CIR each scheduler object at the distribution condition schematic diagram of scheduling queue;
Fig. 5 be the flow rate of scheduler object S1 when being greater than MIR each scheduler object at the distribution condition schematic diagram of scheduling queue;
Fig. 6 is the structural representation for the device of scheduling queue in the specific embodiment of the invention.
Embodiment
For making object of the present invention, technical scheme and advantage clearly understand, to develop simultaneously embodiment referring to accompanying drawing, scheme of the present invention is described in further detail.
A kind of array dispatching method is proposed in the embodiment of the present invention, for each scheduler object configuration schedules parameter, wherein, described scheduling parameter comprises committed rate (CIR), maximum rate (MIR), priority of status and state quota, described priority of status comprises the first priority of status and the second priority of status, and described state quota comprises the first state quota and the second state quota.When the flow of arbitrary scheduler object is not more than CIR, determine that this scheduler object is in the first state, when the flow of arbitrary scheduler object is greater than CIR and is less than MIR, determine that this scheduler object is in the second state, when the flow of arbitrary scheduler object is not less than MIR, determine that this scheduler object is in the third state.When arbitrary scheduler object is in the first state, the scheduling parameter of its correspondence is the first priority of status and the first state quota, when arbitrary scheduler object is in the second state, the scheduling parameter of its correspondence is the second priority of status and the second state quota, when arbitrary scheduler object is in the third state, owing to can not be assigned on scheduling queue, therefore do not configure corresponding priority and state quota.
Be array dispatching method schematic flow sheet in the embodiment of the present invention see Fig. 1, Fig. 1.Concrete steps are:
Step 101, each scheduler object is assigned to corresponding scheduling queue by the first priority of status according to each scheduler object; When detecting that the flow of arbitrary described scheduler object is greater than CIR and is less than MIR, this scheduler object is switched to corresponding scheduling queue by the second priority of status according to this scheduler object; When detecting that the flow of arbitrary described scheduler object is not less than MIR, this scheduler object is switched out current place scheduling queue, wherein, the queue number of each scheduling queue is corresponding with the value of the priority of status of scheduler object.
When starting allocation schedule object in this step, the flow of each scheduler object is bound to be less than CIR, therefore the first state is all in, because the queue number of each scheduling queue is corresponding with the value of the priority of status of scheduler object, if queue number is 1-10, so the first priority of scheduler object and the scope that arranges of the second priority are 1-10, therefore, distribute each scheduler object to corresponding scheduling queue according to the first priority of status of each scheduler object.The scheduler object scheduling of the two quota of double priority level is adopted in the embodiment of the present invention, therefore when arbitrary scheduler object is in the third state, this scheduler object is switched out scheduling queue, even if without any scheduler object schedulable on scheduling queue, the scheduler object switching out scheduling queue also can not be dispatched.According to actual needs, the also scheduler object scheduling of configurable multipriority polygamy volume, specific implementation and double priority level pair quota is similar, is no longer described in detail here.
Step 102, dispatches according to the flow of absolute priority to the scheduler object on scheduling queue of each scheduling queue, carries out robin scheduling to the multiple scheduler objects on same scheduling queue according to the state quota of correspondence by RR algorithm.
In this step, the absolute priority of each scheduling queue can pre-set, queue number as scheduling queue is less, the absolute priority of this scheduling queue is larger, or the queue number of scheduling queue is larger, the absolute priority of this scheduling queue is larger, and the priority of each scheduler object also needs to configure accordingly according to the priority of scheduling queue.
Carry out robin scheduling according to the state quota of correspondence by RR algorithm to the multiple scheduler objects on same scheduling queue to comprise: when the arbitrary scheduler object in described multiple scheduler object is in the first state, corresponding state quota is the first state quota; When in the second state, corresponding state quota is the second state quota, the state quota corresponding according to the current status of described multiple scheduler object carries out robin scheduling to each scheduler object on same scheduling queue by RR algorithm, wherein, when the flow of arbitrary scheduler object is not more than CIR, this scheduler object is in the first state, and when the flow of scheduler object is greater than CIR and is less than MIR, this scheduler object is in the second state.
When this scheduler object being switched to corresponding scheduling queue according to the second priority of status of this scheduler object in step 101, the method comprises further:
Second state is carried out to this scheduler object and becomes the first state time-out timing, when described time-out time then, if this scheduler object is in the first state, this scheduler object is switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, time-out timing is re-started to this scheduler object; If this scheduler object is in the third state, then this scheduler object is switched out scheduling queue, wherein, when the flow of scheduler object is not less than MIR, this scheduler object is in the third state.
When this scheduler object being switched out scheduling queue in step 101, the method comprises further:
The third state is carried out to this scheduler object and becomes the second state time-out timing, when described time-out time then, if this scheduler object is in the first state, this scheduler object is switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, this scheduler object is switched to corresponding scheduling queue according to its second priority of status; If this scheduler object is in the third state, then time-out timing is re-started to this scheduler object.
The wherein setting of time-out timing, the mode by token bucket during specific implementation realizes, and the flow current according to scheduler object knows that this scheduler object second state becomes the time of the first state needs or has the third state to become the time of the second state needs.
When the scheduling parameter change of described each scheduler object configuration, the scheduling parameter according to each scheduler object after change is again the queue of each described scheduler object allocation schedule and dispatches.Therefore, when scheduling strategy changes, new scheduling strategy can be met by amendment scheduling parameter, save cost greatly.
The parameter how configuring each scheduler object according to scheduling strategy is illustrated below by two examples:
Scheduling strategy one, has the service traffics of three not appositions, is F1, F2 and F3 respectively, requires now that F1 at least ensures 2M, but restriction F1 is no more than 5M at most, and F2 at least ensures 3M, and the maximum bandwidth of F2 is unrestricted, and F3 at least ensures 3M, and the maximum bandwidth of F3 is unrestricted.When F1, F2, F3 are when ensureing in bandwidth, priority F1 and F2 is identical and quota ratio higher than F3, F1 and F2 is 2:3.When F1, F2, F3 exceed its ensure bandwidth time, priority F3 higher than F2 higher than F1.
According to above-mentioned strategy, assuming that scheduling queue number N is 6, assuming that interface bandwidth 10M, creating three scheduler objects is S1, S2 and S3 respectively, and the scheduling parameter arranging each scheduler object is as follows:
S1：
CIR=2M, MIR=5M, the first priority of status is the 0, second priority of status be the 5, first state quota be 2K, K is constant factor, and the second state quota is that xK, x represent Arbitrary Digit.
S2：
CIR=3M, MIR=10M, the first priority of status is the 0, second priority of status be the 4, first state quota is 3K, and the second state quota is xK.
S2：
CIR=3M, MIR=10M, the first priority of status is the 2, second priority of status be the 3, first state quota is xK, and the second state quota is xK.
According to above-mentioned setting, just scheduling strategy can be realized, identical from first priority of status of known F1 with F2 of scheduling strategy of supposition, therefore according to priority when the first state, the state quota ratio of F1 and F2 need be configured, service traffics are only had, any quota ratio during other priority.
Scheduling strategy two, assuming that for a certain reason, present requirement is at F1, F2, F3 becoming F1 higher than F2 higher than F3 for the priority exceeded when ensureing bandwidth, and it is identical higher than F1 that the priority exceeded when it ensures bandwidth becomes F3 and F2, then only need the parameter of adjustment S1, S2 and S3 as follows:
S1：
CIR=2M, MIR=5M, the first priority of status is the 0, second priority of status be the 5, first state quota is xK, and the second state quota is xK.
S2：
CIR=3M, MIR=10M, the first priority of status is the 1, second priority of status be the 3, first state quota is xK, and the second state quota is 3K.
S3：
CIR=3M, MIR=10M, the first priority of status is the 2, second priority of status be the 3, first state quota is xK, and the second state quota is 3K.
As seen from the above, when scheduling strategy changes, only need to adjust these six parameters and just can meet new scheduling strategy demand.
Below for scheduling strategy one, by reference to the accompanying drawings, how the detailed description specific embodiment of the invention realizes queue scheduling.
During initialization, scheduler object S1, S2 and S3 are in the first state, therefore according to the value 0,0,2 of priority during each scheduler object the first state, scheduler object S1 and scheduler object S2 being assigned to queue number is on the scheduling queue of 0, scheduler object S3 being assigned to queue number is on the scheduling queue of 2, be be the schematic diagram of each scheduler object allocation schedule queue according to the scheduling parameter of each scheduler object see Fig. 2, Fig. 2.According to the absolute priority of each scheduling queue scheduling individual queue in Fig. 2, therefore first dispatch to be queue number be 0 scheduling queue on scheduler object, this scheduling queue has two scheduler objects, and carries out RR robin scheduling according to the ratio of 2K:3K.
When detecting that the flow rate of scheduler object S1 is greater than 2M and is less than 5M, that determines scheduler object S1 is in the second state, when scheduler object S1 is in the second state, its second priority of status is 5, therefore scheduler object S is switched to the scheduling queue that queue number is 5, when being greater than CIR see Fig. 3, Fig. 3 flow rate that is scheduler object S1, each scheduler object is at the distribution condition schematic diagram of scheduling queue.From the second state, the first state is become to scheduler object S1 simultaneously and carries out time-out timing, when time-out time then, if scheduler object S1 is in the first state, scheduler object S1 is switched to corresponding scheduling queue according to its first priority of status; If scheduler object S1 is in the second state, time-out timing is re-started to scheduler object S1; If scheduler object S1 is in the third state, then scheduler object S1 is switched out scheduling queue, wherein, when the flow of scheduler object is not less than MIR, scheduler object S1 is in the third state.According to the absolute priority of each scheduling queue scheduling individual queue in Fig. 3, therefore first dispatch to be queue number be 0 scheduling queue on scheduler object, this scheduling queue has 1 scheduler object, then the scheduler object S2 on this scheduling queue is dispatched.
When after certain hour, when detecting that the flow of scheduler object S2 is greater than 3M and when being less than 10M, determine that scheduler object S2 is in the second state, when being in the second state according to scheduler object S2, its second priority of status is 4, therefore scheduler object S2 is switched to the scheduling queue that queue number is 4, when being greater than CIR see Fig. 4, Fig. 4 flow rate that is scheduler object S2, each scheduler object is at the distribution condition schematic diagram of scheduling queue.From the second state, the first state is become to scheduler object S2 simultaneously and carries out time-out timing, when time-out time then, if scheduler object S2 is in the first state, scheduler object S2 is switched to corresponding scheduling queue according to its first priority of status; If scheduler object S2 is in the second state, time-out timing is re-started to scheduler object S2; If scheduler object S2 is in the third state, then scheduler object S2 is switched out scheduling queue.According to the absolute priority of each scheduling queue scheduling individual queue in Fig. 4, therefore first dispatch to be queue number be 2 scheduling queue on scheduler object, this scheduling queue has 1 scheduler object, then the scheduler object S3 on this scheduling queue is dispatched.
Assuming that certain moment scheduler object S1 flowed exceed 5M, then S1 is switched out scheduling queue, and when being greater than MIR see Fig. 5, Fig. 5 flow rate that is scheduler object S1, each scheduler object is at the distribution condition schematic diagram of scheduling queue.According to the absolute priority of each scheduling queue scheduling individual queue in Fig. 5, therefore first dispatch to be queue number be 2 scheduling queue on scheduler object, this scheduling queue has 1 scheduler object, then the scheduler object S3 on this scheduling queue is dispatched.Because scheduling S1 is switched out scheduling queue, even if now scheduling queue S2 and scheduling queue S3 does not have flow schedulable, also can not scheduler object S1.From the third state, the second state is become to scheduler object S1 simultaneously and carries out time-out timing, when time-out time then, if scheduler object S1 is in the first state, scheduler object S1 is switched to corresponding scheduling queue according to its first priority of status; If scheduler object S1 is in the second state, be then switched to corresponding scheduling queue according to second priority of status of scheduler object S1; If scheduler object S1 is in the third state, time-out timing is re-started to scheduler object S1.
During specific implementation, in order to easy to the differentiation of each state, the first state can be set to green, the second state is set to yellow, and the third state is set to redness.Also can be realized by other forms or color according to actual needs.
Based on same inventive concept in the specific embodiment of the invention, a kind of device is also proposed.See the structural representation that Fig. 6, Fig. 6 are for the device of scheduling queue in the specific embodiment of the invention.This device comprises: dispensing unit 601, detecting unit 602, allocation units 603 and scheduling unit 604.
Dispensing unit 601, for according to scheduling strategy being each scheduler object configuration schedules parameter, wherein, described scheduling parameter comprises CIR, MIR, priority of status and state quota, and described priority of status comprises the first priority of status and the second priority of status.
Detecting unit 602, for detecting the flow of each scheduler object.
Allocation units 603, each scheduler object is assigned to corresponding scheduling queue by the first priority of status for each described scheduler object configured according to dispensing unit 601; When detecting unit 602 detects that the flow of arbitrary described scheduler object is greater than CIR and is less than MIR, this scheduler object is switched to corresponding scheduling queue by the second priority of status according to this scheduler object of dispensing unit 601 configuration; When detecting unit 602 detects that the flow of arbitrary described scheduler object is not less than MIR, this scheduler object is switched out current place scheduling queue, wherein, the queue number of each scheduling queue is corresponding with the value of the priority of status of scheduler object.
Scheduling unit 604, for the absolute priority according to each scheduling queue, each scheduler object that the allocation units 603 on scheduling queue distribute is dispatched, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue.
Preferably,
Dispensing unit 601, state quota described in the scheduling parameter of configuration comprises the first state quota and the second state quota.
Scheduling unit 604, for when the arbitrary scheduler object in described multiple scheduler object is in the first state, corresponding state quota is the first state quota; When in the second state, corresponding state quota is the second state quota, the state quota corresponding according to the current status of described multiple scheduler object carries out robin scheduling to each scheduler object on same scheduling queue by RR algorithm, wherein, when the flow of arbitrary scheduler object is not more than CIR, this scheduler object is in the first state, and when the flow of scheduler object is greater than CIR and is less than MIR, this scheduler object is in the second state.
Preferably, this device comprises further: timing unit 605.
Timing unit 605, for when this scheduler object to be switched to corresponding scheduling queue according to the second priority of status of this scheduler object by allocation units 603, to carry out the second state to this scheduler object and becomes the first state time-out timing.
Allocation units 603, are further used for working as timing unit 605 timing time-out time then, if this scheduler object is in the first state, this scheduler object are switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, triggers timing unit 605 and time-out timing is re-started to this scheduler object; If this scheduler object is in the third state, then this scheduler object is switched out scheduling queue, wherein, when the flow of scheduler object is not less than MIR, this scheduler object is in the third state.
Preferably,
Timing unit 605, is further used for when this scheduler object is switched out scheduling queue by allocation units 603, carries out the third state become the second state time-out timing to this scheduler object.
Allocation units 603, are further used for working as timing unit 605 timing time-out time then, if this scheduler object is in the first state, this scheduler object are switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, this scheduler object is switched to corresponding scheduling queue according to its second priority of status; If this scheduler object is in the third state, then triggers timing unit 605 and time-out timing is re-started to this scheduler object.
Preferably, this device comprises further: determining unit 606.
Determining unit 606, for determining whether the scheduling parameter that described each scheduler object that dispensing unit configures configures changes.
Allocation units 603, be further used for, when determining unit 606 determines the scheduling parameter change of described each scheduler object configuration, being again the queue of each described scheduler object allocation schedule and triggering scheduling unit 604 to dispatch according to the scheduling parameter of each scheduler object after change.
The unit of above-described embodiment can be integrated in one, and also can be separated deployment; A unit can be merged into, also can split into multiple subelement further.
In sum, be each scheduler object configuration schedules parameter of scheduling according to scheduling strategy in the specific embodiment of the invention, the scheduling parameter of configuration comprises double priority level, and according to the relation of the current flow of scheduler object and CIR and MIR, determine this scheduler object according to that priority assign to corresponding scheduling queue, each scheduler object on scheduling queue is dispatched according to the absolute priority of each scheduling queue when dispatching, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue, different scheduling strategies can be realized by the different scheduling parameter of configuration, improve dispatching flexibility.
When scheduling strategy changes, only need to adjust these six parameters and just can meet new scheduling strategy demand, the parameter only revising configuration just can meet new scheduling strategy, not only increases the flexibility of scheduling, has greatly saved lower deployment cost simultaneously.
The above, be only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1. an array dispatching method, it is characterized in that, be each scheduler object configuration schedules parameter according to scheduling strategy, wherein, described scheduling parameter comprises promised information rate, maximum rate MIR, priority of status and state quota, described priority of status comprises the first priority of status and the second priority of status, and described method comprises:
Each scheduler object is assigned to corresponding scheduling queue by the first priority of status according to each described scheduler object; When detecting that the flow of arbitrary described scheduler object is greater than CIR and is less than MIR, this scheduler object is switched to corresponding scheduling queue by the second priority of status according to this scheduler object; When detecting that the flow of arbitrary described scheduler object is not less than MIR, this scheduler object is switched out current place scheduling queue, wherein, the queue number of each scheduling queue is corresponding with the value of the priority of status of scheduler object;
According to the absolute priority of each scheduling queue, each scheduler object on scheduling queue is dispatched, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue;
Wherein, described state quota comprises the first state quota and the second state quota;
Describedly by RR algorithm, robin scheduling carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue comprise:
When the arbitrary scheduler object in described multiple scheduler object is in the first state, corresponding state quota is the first state quota; When in the second state, corresponding state quota is the second state quota, the state quota corresponding according to the current status of described multiple scheduler object carries out robin scheduling to each scheduler object on same scheduling queue by RR algorithm, wherein, when the flow of arbitrary scheduler object is not more than CIR, this scheduler object is in the first state, and when the flow of arbitrary scheduler object is greater than CIR and is less than MIR, this scheduler object is in the second state.
2. method according to claim 1, is characterized in that, when this scheduler object is switched to corresponding scheduling queue by described the second priority of status according to this scheduler object, described method comprises further:
Second state is carried out to this scheduler object and becomes the first state time-out timing, when described time-out time then, if this scheduler object is in the first state, this scheduler object is switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, time-out timing is re-started to this scheduler object; If this scheduler object is in the third state, then this scheduler object is switched out scheduling queue, wherein, when the flow of scheduler object is not less than MIR, this scheduler object is in the third state.
3. method according to claim 2, is characterized in that, described when this scheduler object is switched out scheduling queue, described method comprises further:
The third state is carried out to this scheduler object and becomes the second state time-out timing, when described time-out time then, if this scheduler object is in the first state, this scheduler object is switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, this scheduler object is switched to corresponding scheduling queue according to its second priority of status; If this scheduler object is in the third state, then time-out timing is re-started to this scheduler object.
4. according to the method described in claim 1-3 any one, it is characterized in that, described method comprises further:
When the scheduling parameter change of described each scheduler object configuration, the scheduling parameter according to each scheduler object after change is again the queue of each described scheduler object allocation schedule and dispatches.
5. a device, is characterized in that, this device comprises: dispensing unit, detecting unit, allocation units and scheduling unit;
Described dispensing unit, for according to scheduling strategy being each scheduler object configuration schedules parameter, wherein, described scheduling parameter comprises promised information rate, maximum rate MIR, priority of status and state quota, and described priority of status comprises the first priority of status and the second priority of status;
Described detecting unit, for detecting the flow of each scheduler object;
Described allocation units, each scheduler object is assigned to corresponding scheduling queue by the first priority of status for each described scheduler object configured according to described dispensing unit; When described detecting unit detects that the flow of arbitrary described scheduler object is greater than CIR and is less than MIR, this scheduler object is switched to corresponding scheduling queue by the second priority of status of this scheduler object configured according to described dispensing unit; When described detecting unit detects that the flow of arbitrary described scheduler object is not less than MIR, this scheduler object is switched out current place scheduling queue, wherein, the queue number of each scheduling queue is corresponding with the value of the priority of status of scheduler object;
Described scheduling unit, for the absolute priority according to each scheduling queue, each scheduler object that allocation units described on scheduling queue distribute is dispatched, by RR algorithm, robin scheduling is carried out according to the state quota of correspondence to the multiple scheduler objects on same scheduling queue
Wherein,
Described dispensing unit, state quota described in the scheduling parameter of configuration comprises the first state quota and the second state quota;
Described scheduling unit, for when the arbitrary scheduler object in described multiple scheduler object is in the first state, corresponding state quota is the first state quota; When in the second state, corresponding state quota is the second state quota, the state quota corresponding according to the current status of described multiple scheduler object carries out robin scheduling to each scheduler object on same scheduling queue by RR algorithm, wherein, when the flow of arbitrary scheduler object is not more than CIR, this scheduler object is in the first state, and when the flow of scheduler object is greater than CIR and is less than MIR, this scheduler object is in the second state.
6. device according to claim 5, is characterized in that, described device comprises further: timing unit;
Described timing unit, for when this scheduler object to be switched to corresponding scheduling queue according to the second priority of status of this scheduler object by described allocation units, to carry out the second state to this scheduler object and becomes the first state time-out timing;
Described allocation units, are further used for working as described timing unit timing time-out time then, if this scheduler object is in the first state, this scheduler object are switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, triggers described timing unit and time-out timing is re-started to this scheduler object; If this scheduler object is in the third state, then this scheduler object is switched out scheduling queue, wherein, when the flow of scheduler object is not less than MIR, this scheduler object is in the third state.
7. device according to claim 6, is characterized in that,
Described timing unit, is further used for when this scheduler object is switched out scheduling queue by described allocation units, carries out the third state become the second state time-out timing to this scheduler object;
Described allocation units, are further used for working as described timing unit timing time-out time then, if this scheduler object is in the first state, this scheduler object are switched to corresponding scheduling queue according to its first priority of status; If this scheduler object is in the second state, this scheduler object is switched to corresponding scheduling queue according to its second priority of status; If this scheduler object is in the third state, then triggers described timing unit and time-out timing is re-started to this scheduler object.
8. according to the device described in claim 5-7 any one, it is characterized in that, described device comprises further: determining unit;
Described determining unit, for determining whether the scheduling parameter that described each scheduler object that described dispensing unit configures configures changes;
Described allocation units, be further used for, when described determining unit determines the scheduling parameter change of described each scheduler object configuration, being again the queue of each described scheduler object allocation schedule and triggering described scheduling unit to dispatch according to the scheduling parameter of each scheduler object after change.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210217512.4A CN102694740B (en) | 2012-06-28 | 2012-06-28 | A kind of array dispatching method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210217512.4A CN102694740B (en) | 2012-06-28 | 2012-06-28 | A kind of array dispatching method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102694740A CN102694740A (en) | 2012-09-26 |
| CN102694740B true CN102694740B (en) | 2015-10-28 |
Family
ID=46860029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210217512.4A Active CN102694740B (en) | 2012-06-28 | 2012-06-28 | A kind of array dispatching method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102694740B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104917692B (en) * | 2015-06-26 | 2018-05-11 | 新华三技术有限公司 | A kind of method and apparatus for distributing token |
| CN109408208B (en) * | 2018-09-21 | 2022-04-01 | 深圳华大北斗科技股份有限公司 | Multitasking method, device and system of navigation chip and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6862265B1 (en) * | 2000-04-13 | 2005-03-01 | Advanced Micro Devices, Inc. | Weighted fair queuing approximation in a network switch using weighted round robin and token bucket filter |
| US7499398B2 (en) * | 2003-04-16 | 2009-03-03 | International Business Machines Corporation | Method and system for oversubscribing bandwidth in a communications network |
| CN101692648A (en) * | 2009-08-14 | 2010-04-07 | 中兴通讯股份有限公司 | Method and system for queue scheduling |
| CN101938403A (en) * | 2009-06-30 | 2011-01-05 | 中国电信股份有限公司 | Assurance method of multi-user and multi-service quality of service and service access control point |
| CN102036131A (en) * | 2009-09-28 | 2011-04-27 | 中兴通讯股份有限公司 | End-to-end QoS (Quality of Service) control method and device based on GPON (Gigabit Passive Optical Network) system |
-
2012
- 2012-06-28 CN CN201210217512.4A patent/CN102694740B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6862265B1 (en) * | 2000-04-13 | 2005-03-01 | Advanced Micro Devices, Inc. | Weighted fair queuing approximation in a network switch using weighted round robin and token bucket filter |
| US7499398B2 (en) * | 2003-04-16 | 2009-03-03 | International Business Machines Corporation | Method and system for oversubscribing bandwidth in a communications network |
| CN101938403A (en) * | 2009-06-30 | 2011-01-05 | 中国电信股份有限公司 | Assurance method of multi-user and multi-service quality of service and service access control point |
| CN101692648A (en) * | 2009-08-14 | 2010-04-07 | 中兴通讯股份有限公司 | Method and system for queue scheduling |
| CN102036131A (en) * | 2009-09-28 | 2011-04-27 | 中兴通讯股份有限公司 | End-to-end QoS (Quality of Service) control method and device based on GPON (Gigabit Passive Optical Network) system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102694740A (en) | 2012-09-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107925587B (en) | Method and apparatus for network slicing | |
| CN101834786B (en) | Queue scheduling method and device | |
| CN102448147B (en) | Method and device for accessing wireless service | |
| CN104378309A (en) | Method, system and related equipment for achieving QoS in Open Flow network | |
| CN102130823A (en) | Method and network apparatus for communicating data | |
| CN102572962A (en) | Data transmitting method and device for multimode base station | |
| CN101009655A (en) | Traffic scheduling method and device | |
| CN103516622A (en) | Flexible queues in a network switch | |
| CN108809707A (en) | A kind of TSN dispatching methods towards real-time application demand | |
| CN1989738A (en) | Propagation of minimum guaranteed scheduling rates | |
| US11929911B2 (en) | Shaping outgoing traffic of network packets in a network management system | |
| US20070248007A1 (en) | Broadband access network capacity management | |
| CN104348751B (en) | Virtual output queue authorization management method and device | |
| CN102594663A (en) | Queue scheduling method and device | |
| CN101984414B (en) | Method and device for scheduling central processing unit (CPU) resource | |
| CN102868638A (en) | Method and system for dynamically regulating bandwidth | |
| CN101594307A (en) | Dispatching method and system based on multi-queue | |
| CN105573193A (en) | Network assembly and method for prioritization of real time messages in a convergent network | |
| CN102546395B (en) | Business scheduling method and device based on L2VPN networks | |
| CN102694740B (en) | A kind of array dispatching method and device | |
| CN109617806A (en) | A kind of data traffic dispatching method and device | |
| CN103281258B (en) | A kind of data transmission method for uplink and device | |
| CN102769566B (en) | A kind of Multi-stage scheduling system configuration, variation and device | |
| CN105812283B (en) | Array dispatching method 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |