CN1956411A - Method for preventing congestion of data flow - Google Patents
Method for preventing congestion of data flow Download PDFInfo
- Publication number
- CN1956411A CN1956411A CNA2005101167309A CN200510116730A CN1956411A CN 1956411 A CN1956411 A CN 1956411A CN A2005101167309 A CNA2005101167309 A CN A2005101167309A CN 200510116730 A CN200510116730 A CN 200510116730A CN 1956411 A CN1956411 A CN 1956411A
- Authority
- CN
- China
- Prior art keywords
- data flow
- data
- cpu
- timeslice
- task
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
This invention provides a method for preventing congestion of data flow including: 1, setting a dispatch task independently preserving ordinary data message information needed to be reported and not taking part in the management to messages of high priorities, 2, a task opens one or several switches reporting data packets according to the needs each time arriving at a time slice based on time slices to control the rhythm of reporting ordinary data messages to CPU, which can guarantee that CPU is not congested by endless data packets nor important information is submerged.
Description
Technical field
The present invention relates to a kind ofly prevent in the network equipment that data flow is excessive and cause congested method.Wherein relate to the network equipment of various use asic chips, comprise router and switch etc.
Background technology
Network condition becomes increasingly complex in the reality, wherein has a lot of messages to be sent to CPU, is easy to cause congested.Its consequence one is to take a large amount of cpu resources, the 2nd, and may cause important message dropping or handle untimely.
The most frequently used method is a sampling different priorities queue scheduling at present.Important packet leaves in guarantees in the high-priority queue in time to handle.Yet still have two deficiencies in the reality like this: the firstth, because under the normal condition, the important bag of protecting (as protocol massages) is not a lot, and general data bag quantity will exceed several magnitude, and therefore this method can not solve the problem that cpu resource is taken in a large number; The secondth, a lot of important informations are included in (as the clean culture ARP) in the general data bag, also have some agreement to drive (as multicast protocol) by stream, these number of data packets flows are big, and they all are at same low priority, may cause some information to be submerged.In the reality, the packet of these low priorities has caused congested just.And how preventing that these congestion of data flow from producing, prior art does not provide a feasible method.
Summary of the invention
The technical problem to be solved in the present invention is, a kind of new effective ways that prevent congestion of data flow are proposed at the deficiency of present Priority Queuing, this method can make CPU allow selectively that according to the rhythm of oneself some data message enters, thereby guarantee congestedly not produce, and guarantee that important message can not be submerged.
A kind of effective ways that prevent congestion of data flow of the present invention, it may further comprise the steps:
1, a scheduler task is set separately;
This task keeps needs the general data message information that send, according to system's actual conditions, for the message (as the variety of protocol message) of high priority participative management not;
2, scheduling is based on timeslice;
Every timeslice, task are opened the switch that send packet on some or several as required, with the rhythm of control general data messages transmitted to CPU.
In the said method,, each class data message send switch for being provided with respectively on one;
In the said method, the length of timeslice can be according to the ability or the system requirements adjustment of CPU deal with data message; From the CPU angle, this transfer guarantees that can not produce unconfined packet pours in.
In the said method, to the timeslice weighting, to guarantee that some class data message is not lost; In the actual conditions, the data flow message that is used as driving action often only needs to send seldom just several, but the speed of data flow own is very high, and this mode can be guaranteed not have too much duplicate packages and enter.
Compared with prior art, the inventive method can guarantee that CPU is not congested by unconfined packet, can not flood important information simultaneously.This method does not relate to the processing of high priority message, therefore can not have any impact to the high priority message.Strengthened the stability of product.
Description of drawings
Fig. 1 is that mission flow diagram is transferred in packet receiving.
Embodiment
Embodiment below in conjunction with packet receiving scheduler task shown in Figure 1 specifies this method.
With reference to Fig. 1, this method mainly contains following steps:
1, switch is opened a packet receiving scheduler task;
2, when switch needs a certain packet, at first notify scheduler task;
3, scheduler task is registered such packet, and timeslice length is set as requested;
4, when timeslice then, scheduler task is opened the switch of accepting of this class bag, in this time period, this type of packet can both be sent to CPU;
5, after timeslice expired, scheduler task was closed this and is accepted switch, and this class packet will be dropped this moment, arrive up to next timeslice.
Claims (4)
1, a kind of method that prevents congestion of data flow is characterized in that it may further comprise the steps:
1) scheduler task is set separately;
This task keeps needs the data message information sent, according to system's actual conditions, for the message participative management of high priority or participative management not;
2) scheduling is based on timeslice;
Every timeslice, task are opened the switch that send packet on some or several as required, with the rhythm of control data messages transmitted to CPU.
2, the method that prevents congestion of data flow according to claim 1 is characterized in that, send switch for each class data message is provided with respectively on one.
3, the method that prevents congestion of data flow according to claim 1 is characterized in that, the length of described timeslice is according to the ability or the system requirements adjustment of CPU deal with data message.
4, the method that prevents congestion of data flow according to claim 1 is characterized in that, to described timeslice weighting, to guarantee that some class data message is not lost.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101167309A CN100499585C (en) | 2005-10-28 | 2005-10-28 | Method for preventing congestion of data flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101167309A CN100499585C (en) | 2005-10-28 | 2005-10-28 | Method for preventing congestion of data flow |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1956411A true CN1956411A (en) | 2007-05-02 |
| CN100499585C CN100499585C (en) | 2009-06-10 |
Family
ID=38063517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005101167309A Expired - Fee Related CN100499585C (en) | 2005-10-28 | 2005-10-28 | Method for preventing congestion of data flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100499585C (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102025694A (en) * | 2009-09-11 | 2011-04-20 | 中兴通讯股份有限公司 | DSP (Digital Signal Processor) array based device and method for sending Ethernet data |
-
2005
- 2005-10-28 CN CNB2005101167309A patent/CN100499585C/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102025694A (en) * | 2009-09-11 | 2011-04-20 | 中兴通讯股份有限公司 | DSP (Digital Signal Processor) array based device and method for sending Ethernet data |
| CN102025694B (en) * | 2009-09-11 | 2014-07-02 | 中兴通讯股份有限公司 | DSP (Digital Signal Processor) array based device and method for sending Ethernet data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100499585C (en) | 2009-06-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109479032B (en) | Congestion avoidance in network devices | |
| Hamdaoui et al. | A dynamic priority assignment technique for streams with (m, k)-firm deadlines | |
| US7573818B2 (en) | Method and equipment for controlling the congestion management and scheduling of transmission-link capacity in packet-switched telecommunications | |
| CN104579962B (en) | A kind of method and device of qos policy that distinguishing different messages | |
| CN106789721B (en) | Intelligent QOS method and system based on token bucket | |
| Mogul et al. | Eliminating receive livelock in an interrupt-driven kernel | |
| US8553543B2 (en) | Traffic shaping method and device | |
| US9106574B2 (en) | Supporting quality of service differentiation using a single shared buffer | |
| US20040081090A1 (en) | Congestion controller for ethernet switch | |
| US7646781B2 (en) | Methods, systems, and computer program products for selectively discarding packets | |
| US20030120795A1 (en) | Method for capacity enhancement of packet switched networks | |
| EP3073680A1 (en) | Methods, queueing system, network element and network system for queueing and processing of packets | |
| WO2014201934A1 (en) | Message scheduling method and device | |
| CN101834790A (en) | Multicore processor based flow control method and multicore processor | |
| CN103581056A (en) | Method for achieving congestion management in network communication through layered queue scheduling | |
| CN105007235B (en) | A kind of jamming control method in wireless multimedia sensor network | |
| WO2012109880A1 (en) | Queue scheduling method and device | |
| CN101056274A (en) | A time division traffic management method and device | |
| Li et al. | Survey on traffic management in data center network: from link layer to application layer | |
| Li et al. | Packet rank-aware active queue management for programmable flow scheduling | |
| CN106921586B (en) | Data stream shaping method, data scheduling method and device | |
| CN100499585C (en) | Method for preventing congestion of data flow | |
| Baker et al. | On queuing, marking, and dropping | |
| CN101039282A (en) | Method for managing flux of packet inflowing CPU system | |
| Jha et al. | New Queuing Technique for Improving Computer Networks QoS |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190124 Address after: Delaware Patentee after: Open Invention Network Co., Ltd. Address before: 518057 Department of law, Zhongxing building, South hi tech Industrial Park, Nanshan District hi tech Industrial Park, Guangdong, Shenzhen Patentee before: ZTE Corporation |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090610 Termination date: 20201028 |