CN108206787A - A kind of congestion-preventing approach and device - Google Patents

Abstract

The invention discloses a kind of congestion-preventing approach and devices, belong to field of communication technology.The method includes：Data packet to receive sets the time to mark；The data packet is stored in buffer queue；It is whether overtime according to data packet described in the time marker for judgment of the data packet；When the data packet time-out, the data packet is abandoned from the buffer queue.The present invention passes through when receiving data packet, the time is set to mark for it, then the data packet marked there is provided the time is stored in buffer queue, when going out time time-out of the data packet in buffer queue according to time marker for judgment, data packet is abandoned from buffer queue, do so avoid the data packet in buffer queue can not go out team always due to system performance reasons caused by time delay it is excessive, while reduce data packet to system performance occupy.

Description

A congestion avoidance method and device

technical field

The invention relates to the field of communication technology, in particular to a congestion avoidance method and device.

Background technique

With the popularity of the Internet, information exchange and information sharing have become an indispensable part of people's daily life. The continuous growth of information (data packets) in the network will inevitably cause network congestion. Therefore, congestion avoidance is particularly important. The currently widely used congestion avoidance mechanisms include tail drop (English: Tail-Drop), random early detection (English: Random Early Detection, referred to as: RED), weighted random early detection (English: Weighted Random Early Detection, referred to as: WRED) and so on.

Among them, tail drop, RED and WRED all discard all or part of the data packets waiting to enter the queue when the length of the queue reaches the set threshold, or discard low-priority data packets according to the priority.

In the above congestion avoidance scheme, for the data packets that have entered the queue, they may not be able to leave the queue due to system performance reasons. , The data packets are cached in the queue for a long time, which also takes up the performance of the system.

Contents of the invention

In order to solve the problem in the prior art that the data packets buffered in the queue may have too long delay and occupy the performance of the system, the embodiments of the present invention provide a congestion avoidance method and device. Described technical scheme is as follows:

In a first aspect, an embodiment of the present invention provides a congestion avoidance method, the method comprising: setting a time stamp for a received data packet, and the time stamp is used to determine whether the time of the data packet in the buffer has expired; storing the data packet in a cache queue; judging whether the data packet times out according to the time stamp of the data packet; when the data packet times out, discarding the data packet from the cache queue.

The present invention sets a time stamp for the data packet when it is received, and then stores the data packet with the time stamp in the cache queue, and when it is judged that the time of the data packet in the cache queue is overtime according to the time stamp, the data packet is stored in the cache queue. Packets are discarded from the cache queue. This avoids the excessive delay caused by the fact that the data packets in the cache queue cannot be dequeued due to system performance reasons, and at the same time reduces the occupation of the system performance by the data packets.

With reference to the first aspect, in the first implementation manner of the first aspect, setting the time stamp for the received data packet includes: judging the type of the data packet; when the type of the data packet is the set type , set a timestamp for the data packet. In this implementation, before setting the time stamp, the type of the data packet is judged first, and only when the data packet is of the set type, the time stamp is set for it, thereby realizing differentiated processing of services with different delay requirements.

With reference to the first implementation of the first aspect, in the second implementation of the first aspect, the judging the type of the data packet includes: judging the type of the data packet according to the field in the header of the data packet type; or, judging the type of the data packet according to the size of the data packet. In this implementation manner, the type of the data packet is judged by the field of the header of the data packet or the size of the data packet, the judgment method is simple, and the judgment result is accurate.

Wherein, the type of the data packet may be a service type or a length type (data packet length). Wherein, the service type may be a voice data packet, an IPTV data packet, and the like. The length type can be less than 1500 bytes, 1500-9600 bytes, etc. The service type of the data packet can be judged according to the fields in the header of the data packet, and the length type of the data packet can be judged according to the size of the data packet.

With reference to the first implementation manner of the first aspect, in a third implementation manner of the first aspect, setting a time stamp for the data packet includes: storing an identifier of the data packet and a corresponding time stamp.

For example, a storage structure such as a linked list may be used to store the identifier of the data packet and its corresponding time stamp.

Wherein, the identifier of the data packet may be the number of the data packet.

In combination with the first aspect or any one of the first to third implementations of the first aspect, in the fourth implementation of the first aspect, the initial value of the time stamp is 0, and the The time stamp gradually increases with time; or, the initial value of the time stamp is a time threshold, and the time stamp gradually decreases with time; or, the time stamp is a time value when the time stamp is set.

In this implementation, the time stamp includes three implementations. For the way that the time stamp gradually increases or decreases with time, it is relatively simple to judge whether it is timed out; for the way that uses the current time value as the time stamp, the time stamp setting Simpler.

With reference to the fourth implementation of the first aspect, in the fifth implementation of the first aspect, the judging whether the data packet times out according to the time stamp of the data packet includes: judging the value of the time stamp Whether the time threshold is reached, when the value of the time stamp reaches the time threshold, it is judged that the data packet has timed out; or, it is judged whether the value of the time stamp is reduced to 0, when the value of the time stamp When it decreases to 0, it is judged that the data packet has timed out; or, it is judged whether the difference between the current time value and the time value when the time stamp is set in the time stamp is greater than or equal to the time threshold, when the difference When the value is greater than or equal to the time threshold, it is determined that the data packet times out.

In this implementation manner, methods for judging timeout corresponding to three implementation manners of time stamps are provided.

Furthermore, for different types of data packets, the time threshold can be set differently, so as to realize the flexible configuration of different services by the operator, improve the utilization rate of network bandwidth, and reduce unnecessary cache occupation.

In combination with the first aspect or any one of the first to third implementations of the first aspect, in the sixth implementation of the first aspect, the method further includes: after deleting the data packet One or at least two consecutive data packets, and one or at least two consecutive data packets following the data packet are of the same type as the data packet. For example, data packets 1, 2 and 3 are three consecutive data packets in the queue, and data packet 1 is at the front, and the types of these three data packets are the same. At this time, if it is judged that data packet 1 times out, the data packet 1, 2 and 3 are discarded together. Doing so reduces the timeout judgment on data packets 2 and 3 and reduces the amount of processing.

In a second aspect, an embodiment of the present invention provides a congestion avoidance device, which includes units for implementing the method described in the first aspect, such as a setting unit, a cache unit, a judging unit, and a processing unit.

In a third aspect, an embodiment of the present invention provides a congestion avoidance device, the congestion avoidance device includes a processor, a memory, and a communication interface; the processor, the memory, and the communication interface are coupled through a bus; the memory is used to store programs Instructions, the processor enables the congestion avoidance device to execute the method described in the first aspect by executing program instructions stored in the memory.

In a fourth aspect, an embodiment of the present invention further provides a computer-readable medium, which is used to store program codes for execution by a congestion avoidance device, where the program codes include instructions for executing the method described in the first aspect.

In the fifth aspect, the embodiment of the present invention also provides a communication chip, which is applied in communication system equipment, and the communication chip includes: a processor, a memory, and a communication interface; the processor, the memory, and the communication interface are coupled through a bus, The memory is used to store program instructions, and the processor executes the program instructions stored in the memory so that the communication system equipment loaded with the communication chip can execute any possible implementation manner in the first aspect above provided method.

Description of drawings

FIG. 1 is a schematic diagram of data packet storage provided by an embodiment of the present invention;

FIG. 2 is a schematic diagram of a hardware structure of a congestion avoidance device provided by an embodiment of the present invention;

FIG. 3 is a flowchart of a congestion avoidance method provided by an embodiment of the present invention;

FIG. 4 is a flowchart of another congestion avoidance method provided by an embodiment of the present invention;

FIG. 4A is a diagram of a data packet processing process provided by an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a communication device provided by an embodiment of the present invention;

Fig. 6 is a schematic structural diagram of a communication chip provided by an embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the implementation manner of the present invention will be further described in detail below in conjunction with the accompanying drawings.

In order to facilitate the description of the embodiment, the following briefly describes the congestion avoidance scenario:

Fig. 1 is the schematic diagram of data packet cache that the embodiment of the present invention provides, as shown in Fig. 1, the reading module of router or computer is used for getting data packet from cache queue, and data packet adopts first in first out (English: First Input First Output, Abbreviation: FIFO) is stored in the cache queue, and data packet 1 enters the queue first and then leaves the queue first. The process of data packets entering the queue is usually not equal to the speed at which the reading module fetches data packets from the cache queue. For example, in a non-congested system, there is little opportunity to keep packets in the cache queue, each packet is processed by the reading module in a timely manner, and only enters the cache when the reading module performance is insufficient or fails The data packets in the queue will not get the opportunity to leave the queue for a long time. When the data packet reading module returns to normal, the cached data packets are likely to have too much delay for the network, and there is no need to consume performance to process them. Therefore, it is beneficial to system performance to actively discard data packets with excessive delay. The present invention is realized based on the idea of discarding the data packets with too long delay in the queue, see the following for details.

The congestion avoidance scheme proposed by the embodiment of the present invention can be applied not only in fixed network technology, but also in Linux message communication and computer task processing. Specifically, it can be applied to devices such as routers and computers.

The implementation of the congestion avoidance device provided by the embodiment of the present invention will be described below in combination with a specific hardware structure.

FIG. 2 shows a hardware structure for implementing a congestion avoidance device 120 provided by an embodiment of the present invention. The congestion avoidance device 120 may be a router, a computer, or the like. As shown in FIG. 2 , the congestion avoidance device 120 includes: a processor 21 , a communication interface 22 , and a memory 23 .

The processor 21 includes one or more processing cores, and the processor 21 executes various functional applications and information processing by running software programs and modules.

The communication interface 22, the memory 23 and the processor 21 are coupled through a bus. The memory 23 can be used to store software programs as well as modules. The memory can store an operating system 24 and an application program module 25 for at least one function.

The application program module 25 at least includes: a setting module 250 for setting a time stamp, a cache module 251 for storing data packets, a judging module 252 for judging whether to timeout, and a processing module 253 for deleting data packets. The setting module 250 is used to set the time stamp for the received data packet, and the time stamp is used to judge whether the time of the data packet in the cache is overtime; the cache module 251 is used to store the data packet in the cache queue The judging module 252 is used to judge whether the data packet times out according to the time stamp of the data packet; the processing module 253 is used to discard the data packet from the cache queue when the data packet times out.

Optionally, the processor 21 is configured to execute various modules in the application program module 25 to implement the steps required by the congestion avoidance device as shown in FIG. 3 and FIG. 4 .

In addition, the memory 23 is a computer-readable storage medium, which can be implemented by any type of volatile or non-volatile storage devices or their combination, such as static anytime access memory (SRAM), electrically erasable and programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic or Optical Disk.

Those skilled in the art can understand that the structure of the congestion avoidance device 120 shown in FIG. layout of the components.

Referring to FIG. 3 , it shows a congestion avoidance method provided by an embodiment of the present invention, executed by the congestion avoidance device 120 shown in FIG. 2 , the method includes:

S101: Set a time stamp for the received data packet, where the time stamp is used to determine whether the time in the cache of the data packet has expired.

In the embodiment of the present invention, the time stamp is used to judge whether the time for the data packet to stay in the cache queue has expired. Specifically, the time mark may be a static time, such as the time when the time mark is set; it may also be a dynamic time, such as gradually increasing from 0, or gradually decreasing from a certain time value.

In the embodiment of the present invention, the time stamp may be stored using a storage structure such as a linked list.

S102: Store the data packet in a buffer queue.

After setting the timestamp, put the packet into the buffer queue. When the time stamp is a dynamic time, the value of the above time stamp can be changed as the storage time of the data packet in the cache queue increases.

S103: Determine whether the data packet times out according to the time stamp of the data packet.

The specific judging method is different according to the type of time stamp. For example, when the initial value of the time stamp is 0 and the time stamp gradually increases with time, it is judged whether the value of the time stamp reaches the time Threshold, when the value of the time stamp reaches the time threshold, it is judged that the data packet has timed out; when the initial value of the time stamp is a time threshold, and the time stamp gradually decreases with time, the time is judged Whether the value of mark is reduced to 0, when the value of described time mark is reduced to 0, judge described packet overtime; When described time mark is the time value when setting described time mark (time mark of setting time ), determine whether the difference between the current time value and the time value when the time stamp is set in the time stamp is greater than or equal to the time threshold, and when the difference is greater than or equal to the time threshold, determine whether the The above packet timed out.

S104: When the data packet times out, discard the data packet from the buffer queue.

In the embodiment of the present invention, after the data packet is deleted from the cache queue, subsequent data packets in the queue are moved forward to fill the position of the deleted data packet.

Referring to FIG. 4 , it shows another congestion avoidance method provided by an embodiment of the present invention, executed by the congestion avoidance device 120 shown in FIG. 2 , the method includes:

S201: Receive a data packet.

In different application scenarios, the received data packets may also be different. For example, in a fixed network, a router can receive service data packets sent by other network nodes; in Linux message communication, a process in a computer can receive message data packets sent by other processes. In computer task processing, a central processing unit (English: Central Processing Unit, CPU for short) of the computer may receive a task request data packet.

S202: Determine the type of the data packet.

In the embodiment of the present invention, there are different delay requirements for different types. Therefore, different types of data packets are processed in different ways. For those with strict delay requirements, it is necessary to set a time stamp for it, and periodically judge whether the data packet retention time exceeds, and delete it in time when the retention time exceeds. For other services with lower requirements on delay, no time mark may be set for them, and no follow-up processing needs to be performed. In services with time-delay requirements, different services have different time-delay requirements, so the time thresholds for time-out judgment may be different for different services.

Wherein, the type of the data packet may be a service type or a length type (data packet length). Wherein, the service type may be a voice data packet, an Internet Protocol Television (English: Internet Protocol Television, IPTV for short) data packet, and the like. The length type can be less than 1500 bytes, 1500-9600 bytes, etc.

Different scenarios can use different types for classification. For example, in a fixed network, data packets can be classified by service type, while in Linux message communication, data packets can be classified by length type.

In a possible implementation manner, judging the type of the data packet may include:

The type of the data packet (service type) is judged according to the field in the header of the data packet; or, the type of the data packet (length type) is judged according to the size of the data packet.

For example, the business of the data packet is judged according to fields such as the label type in the header of the data packet, the Internet Protocol (English: Internet Protocol, referred to as: IP) address or the media access control (English: Media Access Control, referred to as: MAC) address, etc. type. Among them, the IP address or MAC address in the header of the data packet corresponds to a certain user or a certain service flow, and a certain user or a certain service flow corresponds to a specific service, so according to the IP address or MAC address can be Determine the service type of the data packet.

Or, after receiving the data packet, determine its length type by detecting the length of the data packet. This method can perform congestion avoidance for data packets of a certain length, for example, it is desired to perform congestion avoidance configuration for data packets with a length of 1500-9600 bytes.

In the embodiment of the present invention, it is possible to implement the configuration of services requiring congestion avoidance according to requirements. For example, the user wishes to configure congestion avoidance for the IPTV service. At this time, when the router receives the data packet, it determines whether the data packet is an IPTV data packet.

S203: When the type of the data packet is a set type, set a time stamp for the data packet, where the time stamp is used to determine whether the time in the cache of the data packet has expired.

Wherein, the setting time stamp for the data packet includes:

The identifier of the data packet and its corresponding time stamp are stored.

For example, a storage structure such as a linked list may be used to store the identifier of the data packet and its corresponding time stamp.

Wherein, the identifier of the data packet may be the number of the data packet.

As shown in FIG. 4A , after receiving a data packet, the type of the data packet is judged first, and then different time stamps T are set for different data packets.

Wherein, the initial value of the time stamp is 0, and the time stamp gradually increases with time; or, the initial value of the time stamp is a time threshold, and the time stamp gradually decreases with time; or, the The time stamp is the time value when the time stamp is set.

Taking Figure 4A as an example, the time stamps T set for data packets 1, 2, and 3 in the figure are NA, X, and NA respectively, wherein T=X indicates that the data packet is set with a time stamp, and T=NA indicates that the data packet is not set time stamp.

In the embodiment of the present invention, the time threshold can be set as nanoseconds in seconds or milliseconds.

S204: Store the data packet into a buffer queue.

After setting the timestamp, put the packet into the buffer queue. When the time stamp is a dynamic time, the value of the above time stamp can be changed as the storage time of the data packet in the cache queue increases.

If the initial value of the time stamp is T=X, then T will gradually decrease with time, changing to X-1, X-2... until 0. For example, if X=3ms, T=0 after 3ms from the initial value.

If the initial value of the time stamp is T=0, then T will gradually increase with time, changing to 1, 2... until X in sequence. For example, if X=3ms, T=3ms after 3ms from the initial value.

S205: Determine whether the data packet times out according to the time stamp of the data packet.

In the embodiment of the present invention, it is periodically judged whether the data packet times out according to the time stamp of the data packet. Wherein, the period can be specifically configured as required, and the minimum is 1 ms. If the period is set too large, the processing volume will be too large, and if the period is set too small, it cannot meet the needs of services with high latency requirements.

The specific judging method varies according to the type of time stamp, for example, when the initial value of the time stamp is 0 and the time stamp gradually increases with time, it is judged whether the value of the time stamp reaches the time threshold, When the value of the time stamp reaches the time threshold, it is judged that the data packet has timed out; when the initial value of the time stamp is a time threshold, and the time stamp gradually decreases with time, it is judged that the time stamp is Whether the value is reduced to 0, when the value of the time stamp is reduced to 0, it is judged that the data packet is overtime; when the time stamp is the time value when the time stamp is set (the time of setting the time stamp) , judging whether the difference between the current time value and the time value when the time stamp is set in the time stamp is greater than or equal to the time threshold, and when the difference is greater than or equal to the time threshold, it is judged that the data Packet timed out.

For the third method, since the size of the time threshold corresponds to the type of the data packet, after calculating the difference, determine the time threshold according to the type of the data packet, and then compare the difference with the time threshold the size of. For example, the time to set the time stamp is T1, the current time value is T2, and the time threshold is determined to be X according to the type of the data packet. If T2-T1≥X, it is judged that the data packet times out. Among them, the time thresholds corresponding to each type are configured as needed, because each equipment manufacturer or operator has different requirements for time delay.

For the third method, when recording the number of the data packet and its corresponding time stamp, its type can also be stored, so that it is convenient to obtain the corresponding time threshold when comparing the difference with the time threshold.

S206: When the data packet times out, discard the data packet from the buffer queue.

Optionally, the method also includes:

One or at least two consecutive data packets after the data packet are deleted, and the type of the one or at least two consecutive data packets after the data packet is the same as the data packet.

For example, data packets 1, 2 and 3 are three consecutive data packets in the queue, and data packet 1 is at the front, and the types of these three data packets are the same. At this time, if it is judged that data packet 1 times out, the data packet 1, 2 and 3 are discarded together. Doing so reduces the timeout judgment on data packets 2 and 3 and reduces the amount of processing.

In the embodiment of the present invention, when there are multiple consecutive data packets of the same type entering the cache queue, time stamps can be set for multiple data packets, or time stamps can only be set for some of the data packets, for example, the first and the last.

In the embodiment of the present invention, when there are multiple consecutive data packets of the same type, when the first data packet times out, all the data packets in the multiple data packets can be deleted, or only multiple data packets can be deleted Some packets in the . For example, a total of x data packets of a certain service are continuously stored in the cache queue. When the first data packet times out, the subsequent data packets may also time out, but not all data packets enter the cache at the same time. Therefore, you can Only delete a set number of packets. The number set here needs to be set according to actual needs. The following is an example to illustrate the impact of the set number on this scheme: if the first 10 data packets among multiple data packets enter the cache queue at the same time, when these 10 data packets are about to time out, there is another data of the same type The packets enter the cache queue. If the set number is greater than 10, the last data packet that just entered the cache will be discarded in advance, but this can reduce the processing capacity of the hardware.

In the embodiment of the present invention, after the data packet is deleted from the cache queue, subsequent data packets in the queue are moved forward to fill the position of the deleted data packet.

Optionally, the method further includes: when more than a predetermined number of data packets are continuously discarded within a set time, outputting an alarm message, thereby prompting that the current packet loss is relatively large.

Different alarms can be realized according to the different settings of the predetermined number. For example, if the predetermined number is 0, an alarm will be issued as long as there is data packet discarding; when the predetermined number is greater than 0, an alarm can be used to prompt that the current packet loss is large.

In addition, the above-mentioned warning information is generated by the congestion avoidance device, and the warning information may also include corresponding equipment information (such as a router identifier), and the warning information may be transmitted to other management and control equipment, so that the network administrator can be notified in time of the equipment sending the congestion, Thus, congestion can be solved even if it is.

The above congestion avoidance method provided by the present invention can be used together with traditional congestion avoidance strategies (such as tail drop, RED, WRED, etc.). For example, the method provided by this embodiment is used to process the data packets in the queue, while the tail discard, RED or WRED mechanism can be used to process the data packets not entering the queue.

The following is the device embodiment of the embodiment of the present invention. For the details not described in detail in the device embodiment, please refer to the above corresponding method embodiment.

Fig. 5 shows a block diagram of a communication device provided by an embodiment of the present invention. The communication device can be implemented as all or a part of the congestion avoidance device through a dedicated hardware circuit, or a combination of software and hardware. The congestion avoidance device includes: a setting unit 301 , a cache unit 302 , a judging unit 303 and a processing unit 304 . Wherein, the setting unit 301 is used to set a time stamp for the received data packet, and the time stamp is used to judge whether the time of the data packet in the cache is overtime; the cache unit 302 is used to store the data packet into In the cache queue; the judging unit 303 is used to judge whether the data packet has timed out according to the time stamp of the data packet; the processing unit 304 is used to remove the data packet from the cache queue when the data packet times out discarded.

Wherein, the setting unit 301 is configured to: determine the type of the data packet; when the type of the data packet is a set type, set a time stamp for the data packet.

Wherein, the setting unit 301 is configured to: judge the type of the data packet according to the fields in the header of the data packet; or judge the type of the data packet according to the size of the data packet.

Wherein, the setting unit 301 is configured to: store the identifier of the data packet and its corresponding time stamp.

Wherein, the initial value of the time stamp is 0, and the time stamp gradually increases with time; or, the initial value of the time stamp is a time threshold, and the time stamp gradually decreases with time; or, the The time stamp is the time value when the time stamp is set.

Wherein, the judging unit 303 is configured to: judge whether the value of the time stamp reaches the time threshold, and when the value of the time stamp reaches the time threshold, judge that the data packet times out; Whether the value of the time stamp is reduced to 0, when the value of the time stamp is reduced to 0, it is judged that the data packet is timed out; Whether the difference of the time values is greater than or equal to the time threshold, and when the difference is greater than or equal to the time threshold, it is judged that the data packet times out.

Optionally, the processing unit 304 is further configured to: delete one or at least two consecutive data packets after the data packet, and the one or at least two consecutive data packets after the data packet are the same as the same type.

Relevant details may be combined with the method embodiments described with reference to FIG. 3 and FIG. 4 .

It should be noted that the above-mentioned setting unit 301 , cache unit 302 , judging unit 303 and processing unit 304 may be implemented by a processor or implemented by the processor executing program instructions in the memory.

FIG. 6 shows a structural diagram of a communication chip provided by an embodiment of the present invention, which is applied in the foregoing congestion avoidance device. The communication chip includes: a processor 410 , a memory 420 and a communication interface 430 . The processor 410 is respectively connected to the memory 420 and the communication interface 430 through a bus.

The communication interface 430 is used to communicate with other communication devices.

Processor 410 includes one or more processing cores. The processor 410 runs an operating system or application program modules.

Optionally, the memory 420 can store an operating system 422 and an application program module 424 required by at least one function. Optionally, the application module 424 includes: a receiving module 424a, a processing module 424b and a sending module 424c. Among them, the receiving module 424a is used to implement steps related to data packet reception; the processing module 424b is used to implement steps related to calculation or processing such as time stamp setting, timeout judgment, and data packet discarding; the sending module 424c is used to implement steps related to sending.

In addition, the memory 420 can be implemented by any type of volatile or non-volatile storage devices or their combination, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable In addition to programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

Those skilled in the art can understand that the structure shown in FIG. 6 does not constitute a limitation of the above-mentioned communication chip, and may include more or less components than shown in the figure, or combine certain components, or arrange different components.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (14)

1. A congestion avoidance method, characterized in that the method comprises: Set timestamps for received packets; storing the data packet in a cache queue; judging whether the data packet times out according to the time stamp of the data packet; When the data packet times out, the data packet is discarded from the buffer queue. 2. The method according to claim 1, wherein said setting a time stamp for the received data packet comprises: Determine the type of the data packet; When the type of the data packet is a set type, a time stamp is set for the data packet. 3. The method according to claim 2, wherein said judging the type of said data packet comprises: Judging the type of the data packet according to the field in the header of the data packet; or, The type of the data packet is judged according to the size of the data packet. 4. The method according to claim 2, wherein said setting a time stamp for said data packet comprises: The identifier of the data packet and its corresponding time stamp are stored. 5. The method according to any one of claims 1 to 4, wherein the initial value of the time stamp is 0, and the time stamp gradually increases with time; or, the initial value of the time stamp is a time threshold, and the time stamp gradually decreases with time; or, the time stamp is a time value when the time stamp is set. 6. The method according to claim 5, wherein said judging whether the data packet is timed out according to the time stamp of the data packet comprises: judging whether the value of the time stamp reaches the time threshold, and when the value of the time stamp reaches the time threshold, judging that the data packet times out; or, judging whether the value of the time stamp decreases to 0, and when the value of the time stamp decreases to 0, it is judged that the data packet times out; or, judging whether the difference between the current time value and the time value when the time stamp is set in the time stamp is greater than or equal to the time threshold, and when the difference is greater than or equal to the time threshold, determine whether the data packet time out. 7. The method according to any one of claims 1 to 4, wherein the method further comprises: One or at least two consecutive data packets after the data packet are deleted, and the type of the one or at least two consecutive data packets after the data packet is the same as the data packet. 8. A congestion avoidance device, characterized in that the device comprises: A setting unit, configured to set a time stamp for the received data packet, and the time stamp is used to judge whether the time of the data packet in the buffer is overtime; a cache unit, configured to store the data packet in a cache queue; a judging unit, configured to judge whether the data packet times out according to the time stamp of the data packet; A processing unit, configured to discard the data packet from the buffer queue when the data packet times out. 9. The device according to claim 8, wherein the setting unit is configured to: Judging the type of the data packet; when the type of the data packet is a set type, setting a time stamp for the data packet. 10. The device according to claim 9, wherein the setting unit is configured to: The type of the data packet is judged according to the field in the header of the data packet; or, the type of the data packet is judged according to the size of the data packet. 11. The device according to claim 9, wherein the setting unit is used for: The identifier of the data packet and its corresponding time stamp are stored. 12. The device according to any one of claims 8 to 11, wherein the initial value of the time stamp is 0, and the time stamp gradually increases with time; or, the initial value of the time stamp is a time threshold, and the time stamp gradually decreases with time; or, the time stamp is a time value when the time stamp is set. 13. The device according to claim 12, wherein the judging unit is configured to: judging whether the value of the time stamp reaches the time threshold, and when the value of the time stamp reaches the time threshold, judging that the data packet times out; or, judging whether the value of the time stamp decreases to 0, When the value of the time stamp decreases to 0, it is judged that the data packet has timed out; or, it is judged whether the difference between the current time value and the time value when the time stamp is set in the time stamp is greater than or equal to the A time threshold, when the difference is greater than or equal to the time threshold, it is judged that the data packet times out. 14. The device according to any one of claims 8 to 11, wherein the processing unit is further configured to: One or at least two consecutive data packets after the data packet are deleted, and the type of the one or at least two consecutive data packets after the data packet is the same as the data packet.