CN106559354A - A kind of method and device for preventing CPU packet congestions - Google Patents
A kind of method and device for preventing CPU packet congestions Download PDFInfo
- Publication number
- CN106559354A CN106559354A CN201510629688.4A CN201510629688A CN106559354A CN 106559354 A CN106559354 A CN 106559354A CN 201510629688 A CN201510629688 A CN 201510629688A CN 106559354 A CN106559354 A CN 106559354A
- Authority
- CN
- China
- Prior art keywords
- message
- queue
- cpu
- fifo
- message queue
- 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.)
- Pending
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
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/6295—Queue scheduling characterised by scheduling criteria using multiple queues, one for each individual QoS, connection, flow or priority
-
- 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
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/625—Queue scheduling characterised by scheduling criteria for service slots or service orders
- H04L47/6275—Queue scheduling characterised by scheduling criteria for service slots or service orders based on priority
Abstract
The invention discloses a kind of method and device for preventing CPU packet congestions, is related to communication technical field, its method is comprised the following steps:Message scheduler sets up the message queue of different priorities according to the priority of all kinds of messages for receiving;Priority and bandwidth demand of the message scheduler according to each message queue, it is determined that for the scheduling time weight of schedules message queue;Service priority of the message scheduler according to every kind of type of message, it is determined that the priority dispatched by CPU for message queue;Buffer status of the message scheduler according to the FIFO of the scheduling time weight, CPU dispatching priorities and the message queue that will be dispatched, the message queue of round-robin scheduling different priorities is to CPU.Present invention can assure that the message of high priority can be processed in time, the message of low priority will not congestion, for control is joined the team the speed of message, can to greatest extent using idle cpu resource.
Description
Technical field
The present invention relates to communication technical field, more particularly to one kind prevent CPU (Central Processing
Unit, central processing unit) packet congestion method and device.
Background technology
For the communication system equipment in net, always need to allow CPU on equipment carry out protocol massages or
The process of other control messages.Such as, broadcasting packet, SNMP (Simple Network Management
Protocol, Simple Network Management Protocol) message, OAM (Operation Administration and
Maintenance, operation, management, maintenance) message etc..Common equipment link model is that CPU passes through
(such as PCI, (Peripheral Component Interconnect, external equipment interconnection are total for system bus devices
Line) bus) mounting special chip, similar to exchanger chip, network processor chip,
FPGA (Field-Programmable Gate Array, field programmable gate array) etc., and special chip is logical
COM1 is crossed, great majority are connected with CPU for Ethernet interface, by messages transmitted to CPU.But, sometimes can
Occur as the speed of CPU software process does not catch up with the speed of message up sending, and the message of a certain type it is a large amount of on
Report CPU and cause other messages to cannot get the phenomenon that CPU is processed.At this time, need to try every possible means to release CPU
The congestion state of message so that all types of transmitted to CPU messages can have the chance processed by CPU.
For in real work for a common example, it is assumed that there is the message of 3 priority types can be same
When transmitted to CPU, then the message for resetting 0~2 priority is the message of 3 kinds of target types.Then report for this 3 kinds
Text simultaneously transmitted to CPU, then when taking message to final CPU and being processed, CPU is always from a total buffering
Message is taken out in area.And this relief area does not differentiate between type of message, when only CPU is taken out message,
Just can know that this is a what kind of message.It is like a flight data recorder, when the message of a certain type
When sending on a large amount of, and other message up sending speed are not so good as it, (such as, differing an order of magnitude), then CPU
Very low will be become to the hit rate of other messages, some Protocol latency time-out have been may result in, occurring in that cannot
The situation of normal work.
In order to release such case, need to limit such message up sending, speed limit is carried out to which.It is common
Way is that speed limit is carried out to queue, one queue its speed for reporting message of setting, it is assumed that for 1000
It is per second.If beyond this speed, abandoned using hardware mechanisms or software mechanism.
But, such method, it wastes the resource of CPU, and it matches somebody with somebody the speed of message up sending in the dust,
That is, when CPU is idle, CPU has the ability to process the message more than the speed, and uploading rate is
Cannot change, this is just in the resource for virtually wasting CPU.
The content of the invention
It is an object of the invention to provide a kind of method and device for preventing CPU packet congestions, solves existing skill
So that all types of transmitted to CPU messages cannot be by CPU process under the congestion state of CPU messages in art
Problem.
According to an aspect of the invention, there is provided a kind of method for preventing CPU packet congestions, including following step
Suddenly:
Message scheduler sets up the message queue of different priorities according to the priority of all kinds of messages for receiving;
Priority and bandwidth demand of the message scheduler according to each message queue, it is determined that being used for schedules message queue
Scheduling time weight;
Service priority of the message scheduler according to every kind of type of message, it is determined that being dispatched by CPU for message queue
Priority;
Message scheduler is according to the scheduling time weight, CPU dispatching priorities and the message team that will be dispatched
The buffer status of the FIFO (First Input First Output, FIFO relief area) of row, circulation
The message queue of scheduling different priorities is to CPU.
Preferably, during the message queue of message scheduler root scheduling different priorities is to CPU, to scheduled
Message queue carry out flow-control.
Preferably, it is described flow-control is carried out to scheduled message queue to include:
The caching message amount of the FIFO of the message queue of each priority is detected;
When the caching message amount for detecting FIFO reaches caching Second Threshold, by FIFO will be entered
Message carry out stream control shaping controlling the message into FIFO message queues;
When the caching message amount for detecting FIFO reaches three threshold value of caching, all entrance FIFO are abandoned
The message of message queue;
When the caching message amount of FIFO is detected less than caching first threshold, recover to receive all entrance
The message of FIFO message queues.
Preferably, the first threshold is less than the Second Threshold, and the Second Threshold is less than the 3rd threshold value.
Preferably, priority and bandwidth demand of the message scheduler according to each message queue, it is determined that being used for
The scheduling time weight of schedules message queue includes:
According to the priority and bandwidth demand of each message queue, it is that each message queue distributes for schedules message
The timeslice of queue;
According to the ratio of the timeslice distributed between the different message queues, the scheduling time power of message queue is determined
Weight.
Preferably, the message scheduler is according to the scheduling time weight, CPU dispatching priorities and will
The buffer status of the FIFO of the message queue of scheduling, the message queue of round-robin scheduling different priorities give CPU bags
Include:
Detected by the buffer status of the FIFO of the message queue to dispatching, judge whether to need
Message to be dispatched;
When the message for judging to there is no needs scheduling, then this dispatcher can be postponed and dispatch excellent to next CPU
The queue of first level;
When the message for judging to there is needs scheduling, the message queue is scheduled to according to the scheduling time weight
CPU。
According to a further aspect in the invention, there is provided a kind of device for preventing CPU packet congestions, including:
Message queue module is set up, for message scheduler according to the priority of all kinds of messages for receiving, is set up not
The message queue of same priority;
Determine scheduling time weight and priority block, for message scheduler according to the preferential of each message queue
Level and bandwidth demand, it is determined that for the scheduling time weight of schedules message queue, and according to every kind of type of message
Service priority, it is determined that the priority dispatched by CPU for message queue;
Schedules message Queue module, for message scheduler according to the scheduling time weight, CPU dispatching priorities
The buffer status of the FIFO of level and the message queue that will be dispatched, the message team of round-robin scheduling different priorities
Arrange to CPU.
Preferably, the schedules message Queue module includes flow controlling unit, for adjusting in message scheduler root
During the message queue of degree different priorities is to CPU, flow-control is carried out to scheduled message queue.
Preferably, the flow controlling unit includes:
Detection sub-unit, for examining to the caching message amount of the FIFO of the message queue of each priority
Survey;
Flow-control subelement, for when the caching message amount for detecting FIFO reach caching Second Threshold when,
Carry out flowing control shaping control into the message of FIFO message queues by the message to FIFO will be entered, work as inspection
When measuring the caching message amount of FIFO and reaching three threshold value of caching, all entrance FIFO message queues are abandoned
Message, and when the caching message amount of FIFO is detected less than caching first threshold, recover to receive all
Into the message of FIFO message queues.
Preferably, the first threshold is less than the Second Threshold, and the Second Threshold is less than the 3rd threshold value.
Compared with prior art, the beneficial effects of the present invention is:
Binding time piece robin scheduling of the present invention and weighted priority queue, and use Qos (Quality of
Service, service quality) method dispatched carry out the method for packet loss reaching and prevent CPU from leaping high and
Make full use of cpu resource, it is therefore prevented that certain form of message cannot report target cpu.
Description of the drawings
Fig. 1 is a kind of method flow diagram for preventing CPU packet congestions provided in an embodiment of the present invention;
Fig. 2 is a kind of schematic device for preventing CPU packet congestions provided in an embodiment of the present invention;
Fig. 3 is the schematic diagram of the scheduler of three queues provided in an embodiment of the present invention;
Fig. 4 is the schematic diagram of the scheduling behavior that C buckets provided in an embodiment of the present invention are space-time;
Fig. 5 is the first signal that flow-control is carried out to scheduled message queue provided in an embodiment of the present invention
Figure;
Fig. 6 is the second signal that flow-control is carried out to scheduled message queue provided in an embodiment of the present invention
Figure;
Fig. 7 is the 3rd signal that flow-control is carried out to scheduled message queue provided in an embodiment of the present invention
Figure.
Specific embodiment
Below in conjunction with accompanying drawing to a preferred embodiment of the present invention will be described in detail, it will be appreciated that described below
Preferred embodiment be merely to illustrate and explain the present invention, be not intended to limit the present invention.
Fig. 1 is a kind of method flow diagram for preventing CPU packet congestions provided in an embodiment of the present invention, such as Fig. 1
It is shown, comprise the following steps:
Step S101:Message scheduler sets up different priorities according to the priority of all kinds of messages for receiving
Message queue;
Step S102:Priority and bandwidth demand of the message scheduler according to each message queue, it is determined that being used for
The scheduling time weight of schedules message queue;
Step S103:Service priority of the message scheduler according to every kind of type of message, it is determined that being used for message team
Arrange the priority dispatched by CPU;
Step S104:Message scheduler is according to the scheduling time weight, CPU dispatching priorities and will adjust
The buffer status of the FIFO of the message queue of degree, the message queue of round-robin scheduling different priorities is to CPU.
Wherein, during the message queue of message scheduler root scheduling different priorities is to CPU, to scheduled
Message queue carries out flow-control.Specifically, it is described flow-control is carried out to scheduled message queue to include:
The caching message amount of the FIFO of the message queue of each priority is detected;When detecting the slow of FIFO
Stored messages quantity reach caching Second Threshold when, by will enter FIFO message carry out stream control shaping come
Message of the control into FIFO message queues;When the caching message amount for detecting FIFO reaches the 3rd threshold of caching
During value, the message of all entrance FIFO message queues is abandoned;When the caching message amount for detecting FIFO is less than
During caching first threshold, recover to receive the message of all entrance FIFO message queues.Wherein, first threshold
Value is less than the 3rd threshold value less than the Second Threshold, the Second Threshold.
Wherein, priority and bandwidth demand of the message scheduler according to each message queue, it is determined that for adjusting
The scheduling time weight of degree message queue includes:According to the priority and bandwidth demand of each message queue, it is every
Individual message queue distributes the timeslice for schedules message queue;According to distribute between the different message queues when
Between piece ratio, determine the scheduling time weight of message queue.
Wherein, the message scheduler is according to the scheduling time weight, CPU dispatching priorities and will adjust
The buffer status of the FIFO of the message queue of degree, the message queue of round-robin scheduling different priorities include to CPU:
Detected by the buffer status of the FIFO of the message queue to dispatching, judge whether to need to adjust
The message of degree;When the message for judging to there is no needs scheduling, then this dispatcher can be postponed to next CPU
The queue of dispatching priority;When the message for judging to there is needs scheduling, will be described according to the scheduling time weight
Message queue is scheduled to CPU.
Fig. 2 is a kind of schematic device for preventing CPU packet congestions provided in an embodiment of the present invention, such as Fig. 2
It is shown, including:Set up message queue module 201, determine scheduling time weight and priority block 202 and
Schedules message Queue module 203.It is described to set up message queue module 201, for message scheduler according to receiving
All kinds of messages priority, set up the message queue of different priorities;The determination scheduling time weight and excellent
First level module 202, for message scheduler according to the priority and bandwidth demand of each message queue, it is determined that using
In the scheduling time weight of schedules message queue, and according to the service priority of every kind of type of message, it is determined that using
In the priority that message queue is dispatched by CPU;The schedules message Queue module 203, for message scheduler
According to the slow of the FIFO of the scheduling time weight, CPU dispatching priorities and the message queue that will be dispatched
State is deposited, the message queue of round-robin scheduling different priorities is to CPU.
Wherein, the schedules message Queue module 203 includes flow controlling unit, in message scheduler root
During the message queue of scheduling different priorities is to CPU, flow-control is carried out to scheduled message queue.Institute
Stating flow controlling unit includes:Detection sub-unit, for the message queue to each priority FIFO it is slow
Stored messages quantity is detected;Flow-control subelement, for reaching when the caching message amount for detecting FIFO
During to caching Second Threshold, stream control shaping control is carried out by the message to entering FIFO and enters FIFO
The message of message queue, when the caching message amount for detecting FIFO reaches three threshold value of caching, abandons institute
There is the message into FIFO message queues, and when the caching message amount for detecting FIFO is less than caching first
During threshold value, recover to receive the message of all entrance FIFO message queues.Wherein, the first threshold is less than institute
Second Threshold is stated, the Second Threshold is less than the 3rd threshold value.
3- Fig. 7 is introduced to the process that implements of method mentioned immediately above to the present invention below in conjunction with the accompanying drawings.
As shown in figure 3, being a message scheduler.It has 3 buckets, and respectively A/B/C, three kinds of process are excellent
The message A/B/C of first level type.Wherein for the CPU priority for processing message A is 0, holding time piece 2T;
For the CPU priority for processing message B is 2, holding time piece 1T;It is for the CPU for processing message C
Priority is 1, holding time piece 3T.Therefore priority is A/C/B, and in the state of ideal, for
The message of transmitted to CPU, is not relevant for its message length, therefore scheduling time weight is 2:1:3.(described not
With the ratio of the timeslice distributed between queue, as the scheduling time weight of message queue) when message scheduler Shen
When please arrive timeslice, whether the FIFO that can first go query message A is empty, if being not sky, preferentially by when
Between piece use to A, the timeslice is not reaching to 2T, then next timeslice uses still to A.
As shown in figure 4, after A has used timeslice, when application is to next timeslice, C buckets be it is empty,
So B buckets outgoing packet.That is, after A has used timeslice, the timeslice reason that the next one is applied
Should use to C, but if the FIFO of C is sky, no message is pending, then this timeslice will
It is supplied to B to use.Here it is round-robin, which can ensure that each message can be obtained by CPU process
Chance.
As shown in figure 5, when the number of message reaches y inside the bucket, controlling message into the speed of bucket is
S.That is, needing the message queue to each barrel of the inside to carry out flow-control, it is assumed that in queue C buckets
Message number when reaching threshold value y, then it is assumed that CPU is busy, then limit the message into enqueue C buckets
Speed be S.
As shown in fig. 6, when the number of message reaches z inside the bucket, then abandoning the message of all entrance buckets.
When the message in message queue inside bucket is counted to up to z, then it is assumed that the message of the queue congestion, then
Abandon any C bucket messages that will enter enqueue.
As shown in fig. 7, when the number of message reaches x inside the bucket, then receiving the message of all entrance buckets.
When the message number in message queue inside bucket drops to x, then do not limit C bucket messages and enter enqueue.
Therefore, this can add the form of leaky bucket to realize using token bucket.Flow control is carried out to each queue
The purpose of system is smooth for allow the speed of each messages transmitted to CPU to try one's best, simultaneously for high preferential
The message of level sets lower threshold value, it can be ensured that the message of low priority is always processed.
Total described, the solution of the present invention includes:
1st, a message scheduler is set up, can goes to realize by software or hardware (such as FPGA).Message
Scheduler is responsible for setting up message queue, receives message from equipment and sends message to CPU.Message scheduler is every
A kind of message distributes timeslice, by periodically to CPU application time pieces, circulation is distributed to each message queue to be made
With.
2nd, assume that the message A/B/C for there are 3 types needs transmitted to CPU, setting processing priority is A/C/B,
Then scheduling time weight is 2:1:3.
3rd, on the premise of application to timeslice, if three kinds of messages are all pending, from high to low according to priority
Order preferentially distribute timeslice and use to A, after A use scheduling time 2T, next timeslice general
Can use to C.If timeslice arrives, no message inside the queue of priority C, then should
Timeslice will be used to message inside priority B queue.
4th, given threshold x, y, z, x<y<z.When the message number in queue reaches y, control its message and enter
The speed of enqueue is S, and when the message number in queue reaches z, queue starts to abandon all to enter to join the team
The message of row, when the message number in queue is snapped down to x, queue starts no longer to control the message into enqueue
Speed.In order to the message for ensureing low priority can fully obtain processed chance, high priority message can be by
Its threshold value is matched somebody with somebody smaller.
5th, the difference of message length in the state of ideal, is not considered.The speed of hypothesis Message processing is per second
N number of message, time leaf length are T, and the timeslice number applied each second is M, and the weight of this queue is w,
Total weight and be Q, then the type message report each second the speed of CPU to be:N*T*M*(w/Q).Then exist
Above-mentioned 4, in, speed S should be less than the speed.
During actual use, not exclusively only the message of 3 types can report CPU, it may be possible to
There are many kinds, then I can create message by screening the eigenvalue of message in message scheduler
Buffer queue.By the priority of configuration message queue, the weight of scheduling time, x/y/z threshold values, and
Speed S is calculated according to the resource situation of current system, you can realize transmitted to CPU message provided by the present invention
Dispatching method.
In sum, the present invention has following technique effect:
The present invention realizes message queue to CPU by each message queue using the algorithm of robin scheduling
The feedback of resource situation, realizes the priority of queue and avoids to CPU congestions, be that each queue wins
Using the chance of cpu resource, by the way of weighted queue, the speed to go out FIFO to different queue is carried out
Control, by setting different stream control threshold values and random drop speed, so as to control each queue FIFO receptions
The speed of message, so that it is guaranteed that the queue of low priority can obtain the resource of CPU.
Although being described in detail to the present invention above, the invention is not restricted to this, the art technology
Personnel can carry out various modifications with principle of the invention.Therefore, all modifications made according to the principle of the invention,
All should be understood to fall into protection scope of the present invention.
Claims (10)
1. a kind of method for preventing CPU packet congestions, it is characterised in that comprise the following steps:
Message scheduler sets up the message queue of different priorities according to the priority of all kinds of messages for receiving;
Priority and bandwidth demand of the message scheduler according to each message queue, it is determined that being used for schedules message queue
Scheduling time weight;
Service priority of the message scheduler according to every kind of type of message, it is determined that being dispatched by CPU for message queue
Priority;
Message scheduler is according to the scheduling time weight, CPU dispatching priorities and the message team that will be dispatched
The buffer status of the FIFO of row, the message queue of round-robin scheduling different priorities is to CPU.
2. method according to claim 1, it is characterised in that different excellent in the scheduling of message scheduler root
During the message queue of first level is to CPU, flow-control is carried out to scheduled message queue.
3. method according to claim 2, it is characterised in that described that scheduled message queue is entered
Row flow-control includes:
The caching message amount of the FIFO of the message queue of each priority is detected;
When the caching message amount for detecting FIFO reaches caching Second Threshold, by FIFO will be entered
Message carry out stream control shaping controlling the message into FIFO message queues;
When the caching message amount for detecting FIFO reaches three threshold value of caching, all entrance FIFO are abandoned
The message of message queue;
When the caching message amount of FIFO is detected less than caching first threshold, recover to receive all entrance
The message of FIFO message queues.
4. method according to claim 3, it is characterised in that the first threshold is less than described second
Threshold value, the Second Threshold are less than the 3rd threshold value.
5. method according to claim 1, it is characterised in that the message scheduler is reported according to each
The priority and bandwidth demand of literary queue, it is determined that the scheduling time weight for schedules message queue includes:
According to the priority and bandwidth demand of each message queue, it is that each message queue distributes for schedules message
The timeslice of queue;
According to the ratio of the timeslice distributed between the different message queues, the scheduling time power of message queue is determined
Weight.
6. method according to claim 5, it is characterised in that the message scheduler is according to the tune
The buffer status of the FIFO of degree time weighting, CPU dispatching priorities and the message queue that will be dispatched, follow
The message queue of ring scheduling different priorities includes to CPU:
Detected by the buffer status of the FIFO of the message queue to dispatching, judge whether to need
Message to be dispatched;
When the message for judging to there is no needs scheduling, then this dispatcher can be postponed and dispatch excellent to next CPU
The queue of first level;
When the message for judging to there is needs scheduling, the message queue is scheduled to according to the scheduling time weight
CPU。
7. a kind of device for preventing CPU packet congestions, it is characterised in that include:
Message queue module is set up, for message scheduler according to the priority of all kinds of messages for receiving, is set up not
The message queue of same priority;
Determine scheduling time weight and priority block, for message scheduler according to the preferential of each message queue
Level and bandwidth demand, it is determined that for the scheduling time weight of schedules message queue, and according to every kind of type of message
Service priority, it is determined that the priority dispatched by CPU for message queue;
Schedules message Queue module, for message scheduler according to the scheduling time weight, CPU dispatching priorities
The buffer status of the FIFO of level and the message queue that will be dispatched, the message team of round-robin scheduling different priorities
Arrange to CPU.
8. device according to claim 7, it is characterised in that the schedules message Queue module includes
Flow controlling unit, it is for during the message queue of message scheduler root scheduling different priorities is to CPU, right
Scheduled message queue carries out flow-control.
9. device according to claim 7, it is characterised in that the flow controlling unit includes:
Detection sub-unit, for examining to the caching message amount of the FIFO of the message queue of each priority
Survey;
Flow-control subelement, for when the caching message amount for detecting FIFO reach caching Second Threshold when,
Carry out flowing control shaping control into the message of FIFO message queues by the message to FIFO will be entered, work as inspection
When measuring the caching message amount of FIFO and reaching three threshold value of caching, all entrance FIFO message queues are abandoned
Message, and when the caching message amount of FIFO is detected less than caching first threshold, recover to receive all
Into the message of FIFO message queues.
10. device according to claim 9, it is characterised in that the first threshold is less than described second
Threshold value, the Second Threshold are less than the 3rd threshold value.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510629688.4A CN106559354A (en) | 2015-09-28 | 2015-09-28 | A kind of method and device for preventing CPU packet congestions |
PCT/CN2016/091879 WO2017054566A1 (en) | 2015-09-28 | 2016-07-27 | Method of preventing cpu packet congestion and device utilizing same |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510629688.4A CN106559354A (en) | 2015-09-28 | 2015-09-28 | A kind of method and device for preventing CPU packet congestions |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106559354A true CN106559354A (en) | 2017-04-05 |
Family
ID=58416733
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510629688.4A Pending CN106559354A (en) | 2015-09-28 | 2015-09-28 | A kind of method and device for preventing CPU packet congestions |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106559354A (en) |
WO (1) | WO2017054566A1 (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107547537A (en) * | 2017-08-29 | 2018-01-05 | 新华三技术有限公司 | Request message processing method, device, equipment and storage medium |
CN108173784A (en) * | 2017-12-29 | 2018-06-15 | 湖南恒茂高科股份有限公司 | A kind of aging method and device of the data pack buffer of interchanger |
CN108965005A (en) * | 2018-07-18 | 2018-12-07 | 烽火通信科技股份有限公司 | The adaptive method for limiting speed and its system of the network equipment |
CN109412973A (en) * | 2018-09-19 | 2019-03-01 | 咪咕数字传媒有限公司 | A kind of audio-frequency processing method, device and storage medium |
CN109586780A (en) * | 2018-11-30 | 2019-04-05 | 四川安迪科技实业有限公司 | The method for preventing message from blocking in satellite network |
CN110941483A (en) * | 2019-10-23 | 2020-03-31 | 创达特(苏州)科技有限责任公司 | Queue processing method, device and equipment |
WO2020155185A1 (en) * | 2019-02-03 | 2020-08-06 | 华为技术有限公司 | Message scheduling method, scheduler, network device and network system |
CN113590030A (en) * | 2021-06-30 | 2021-11-02 | 济南浪潮数据技术有限公司 | Queue scheduling method, system, equipment and medium |
CN113923171A (en) * | 2021-08-26 | 2022-01-11 | 江苏智臻能源科技有限公司 | Communication management method based on load identification detection platform |
WO2022252085A1 (en) * | 2021-05-31 | 2022-12-08 | 华为技术有限公司 | Computer system and bus flow control method |
WO2023109188A1 (en) * | 2021-12-14 | 2023-06-22 | 中兴通讯股份有限公司 | Message scheduling method, and network device and computer-readable storage medium |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113268326A (en) * | 2021-05-25 | 2021-08-17 | 西安交通大学 | Fair flow bundle scheduling method and system based on time slice rotation |
CN113824651B (en) * | 2021-11-25 | 2022-02-22 | 上海金仕达软件科技有限公司 | Market data caching method and device, storage medium and electronic equipment |
CN117014384A (en) * | 2022-04-29 | 2023-11-07 | 华为技术有限公司 | Message transmission method and message forwarding equipment |
CN116016362B (en) * | 2023-03-28 | 2023-07-07 | 北京六方云信息技术有限公司 | Flow control method, device, terminal equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002039680A2 (en) * | 2000-11-08 | 2002-05-16 | Motorola, Inc., A Corporation Of The State Of Delaware | Method for class of service weight adaptation depending on the queue residence time |
CN101459699A (en) * | 2008-12-25 | 2009-06-17 | 华为技术有限公司 | Method and apparatus for network address conversion |
CN101651615A (en) * | 2009-09-14 | 2010-02-17 | 中兴通讯股份有限公司 | Method and device for scheduling messages |
CN102231697A (en) * | 2011-06-17 | 2011-11-02 | 瑞斯康达科技发展股份有限公司 | Bandwidth dispatching method of message queues as well as message reporting method and device |
CN102420776A (en) * | 2012-01-12 | 2012-04-18 | 盛科网络(苏州)有限公司 | Method and system for dynamically regulating portal resource allocation threshold value |
CN104079501A (en) * | 2014-06-05 | 2014-10-01 | 深圳市邦彦信息技术有限公司 | Queue scheduling method based on multiple priorities |
-
2015
- 2015-09-28 CN CN201510629688.4A patent/CN106559354A/en active Pending
-
2016
- 2016-07-27 WO PCT/CN2016/091879 patent/WO2017054566A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002039680A2 (en) * | 2000-11-08 | 2002-05-16 | Motorola, Inc., A Corporation Of The State Of Delaware | Method for class of service weight adaptation depending on the queue residence time |
CN101459699A (en) * | 2008-12-25 | 2009-06-17 | 华为技术有限公司 | Method and apparatus for network address conversion |
CN101651615A (en) * | 2009-09-14 | 2010-02-17 | 中兴通讯股份有限公司 | Method and device for scheduling messages |
CN102231697A (en) * | 2011-06-17 | 2011-11-02 | 瑞斯康达科技发展股份有限公司 | Bandwidth dispatching method of message queues as well as message reporting method and device |
CN102420776A (en) * | 2012-01-12 | 2012-04-18 | 盛科网络(苏州)有限公司 | Method and system for dynamically regulating portal resource allocation threshold value |
CN104079501A (en) * | 2014-06-05 | 2014-10-01 | 深圳市邦彦信息技术有限公司 | Queue scheduling method based on multiple priorities |
Non-Patent Citations (1)
Title |
---|
三郎扎西: "浅析网络拥塞管理技术", 《西南民族大学学报 自然科学版》 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107547537B (en) * | 2017-08-29 | 2020-12-18 | 新华三技术有限公司 | Request message processing method, device, equipment and storage medium |
CN107547537A (en) * | 2017-08-29 | 2018-01-05 | 新华三技术有限公司 | Request message processing method, device, equipment and storage medium |
CN108173784A (en) * | 2017-12-29 | 2018-06-15 | 湖南恒茂高科股份有限公司 | A kind of aging method and device of the data pack buffer of interchanger |
CN108965005A (en) * | 2018-07-18 | 2018-12-07 | 烽火通信科技股份有限公司 | The adaptive method for limiting speed and its system of the network equipment |
CN108965005B (en) * | 2018-07-18 | 2021-05-14 | 烽火通信科技股份有限公司 | Self-adaptive speed limiting method and system of network equipment |
CN109412973A (en) * | 2018-09-19 | 2019-03-01 | 咪咕数字传媒有限公司 | A kind of audio-frequency processing method, device and storage medium |
CN109412973B (en) * | 2018-09-19 | 2022-09-13 | 咪咕数字传媒有限公司 | Audio processing method and device and storage medium |
CN109586780A (en) * | 2018-11-30 | 2019-04-05 | 四川安迪科技实业有限公司 | The method for preventing message from blocking in satellite network |
WO2020155185A1 (en) * | 2019-02-03 | 2020-08-06 | 华为技术有限公司 | Message scheduling method, scheduler, network device and network system |
US11968111B2 (en) | 2019-02-03 | 2024-04-23 | Huawei Technologies Co., Ltd. | Packet scheduling method, scheduler, network device, and network system |
CN110941483A (en) * | 2019-10-23 | 2020-03-31 | 创达特(苏州)科技有限责任公司 | Queue processing method, device and equipment |
WO2022252085A1 (en) * | 2021-05-31 | 2022-12-08 | 华为技术有限公司 | Computer system and bus flow control method |
CN113590030A (en) * | 2021-06-30 | 2021-11-02 | 济南浪潮数据技术有限公司 | Queue scheduling method, system, equipment and medium |
CN113590030B (en) * | 2021-06-30 | 2023-12-26 | 济南浪潮数据技术有限公司 | Queue scheduling method, system, equipment and medium |
CN113923171A (en) * | 2021-08-26 | 2022-01-11 | 江苏智臻能源科技有限公司 | Communication management method based on load identification detection platform |
CN113923171B (en) * | 2021-08-26 | 2024-02-06 | 江苏智臻能源科技有限公司 | Communication management method based on load identification detection platform |
WO2023109188A1 (en) * | 2021-12-14 | 2023-06-22 | 中兴通讯股份有限公司 | Message scheduling method, and network device and computer-readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
WO2017054566A1 (en) | 2017-04-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106559354A (en) | A kind of method and device for preventing CPU packet congestions | |
CN104854831B (en) | The output queue of equipment based on input rank postpones behavior | |
US11700204B2 (en) | Phantom queue link level load balancing system, method and device | |
EP2939380B1 (en) | Supporting quality of service differentiation using a single shared buffer | |
US8553538B2 (en) | Packet relay device and congestion control method | |
Hua et al. | Scheduling design and analysis for end-to-end heterogeneous flows in an avionics network | |
US20080225705A1 (en) | Monitoring, Controlling, And Preventing Traffic Congestion Between Processors | |
CN107431667A (en) | Packet is dispatched in the network device | |
US20130343398A1 (en) | Packet-based communication system with traffic prioritization | |
CN104718734A (en) | Congestion notification in a network | |
CN102780630B (en) | A kind of method and apparatus realizing QoS queue based on FPGA queue | |
CN100466593C (en) | Method of implementing integrated queue scheduling for supporting multi service | |
CN101616096B (en) | Method and device for scheduling queue | |
US20150131446A1 (en) | Enabling virtual queues with qos and pfc support and strict priority scheduling | |
EP3468122B1 (en) | Packet admission | |
Kozačinski et al. | Configuration of quality of service parameters in communication networks | |
Jha et al. | New Queuing Technique for Improving Computer Networks QoS | |
US9847949B1 (en) | Arbitrating data packets | |
US20180091430A1 (en) | Transmission apparatus and transmission processing method | |
CN102333035B (en) | Router and its rsvp reservation resource regulating method and device | |
Salah et al. | Analysis and simulation of interrupt overhead impact on OS throughput in high‐speed networks | |
Musa et al. | A Comparative Study of Different Queuing Scheduling Disciplines | |
Garba et al. | Efficient bandwidth management and implementation of cross-layer queuing model in a wireless campus area network | |
Lau | A target-delay-driven packet scheduling scheme for real-time interactive applications | |
Прокопович et al. | Mathematical modeling of Internet of Things traffic in AD-HOC networks with hybrid routing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20200617 Address after: 518057 Zhongxing building, A3-01, A3-02, Nanshan District hi tech Industrial Park, Shenzhen, Guangdong Applicant after: Shenzhen ZTE Technical Service Co.,Ltd. Address before: 518057 Nanshan District science and Technology Industrial Park, Guangdong high tech Industrial Park, ZTE building Applicant before: ZTE Corp. |
|
TA01 | Transfer of patent application right | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170405 |
|
RJ01 | Rejection of invention patent application after publication |