WO2016101549A1 - 一种拥塞流量的管理方法、装置及计算机存储介质 - Google Patents

一种拥塞流量的管理方法、装置及计算机存储介质 Download PDF

Info

Publication number
WO2016101549A1
WO2016101549A1 PCT/CN2015/080801 CN2015080801W WO2016101549A1 WO 2016101549 A1 WO2016101549 A1 WO 2016101549A1 CN 2015080801 W CN2015080801 W CN 2015080801W WO 2016101549 A1 WO2016101549 A1 WO 2016101549A1
Authority
WO
WIPO (PCT)
Prior art keywords
congestion
token
bucket
congestion level
level
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.)
Ceased
Application number
PCT/CN2015/080801
Other languages
English (en)
French (fr)
Inventor
马骞
张灵燕
郑利
徐云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sanechips Technology Co Ltd
Original Assignee
Sanechips Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sanechips Technology Co Ltd filed Critical Sanechips Technology Co Ltd
Publication of WO2016101549A1 publication Critical patent/WO2016101549A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Definitions

  • the present invention relates to data communication technologies, and in particular, to a method and an apparatus for managing congestion traffic, and a computer storage medium.
  • the actual situation is: when a traffic source starts to flow to the switching network or multiple traffic sources start to flow to the switching network at the same time, it will cause a burst, and because the switching network cache is not large, it cannot It is good to absorb the bursts, causing congestion and even packet loss. In addition, from the stop to the flow, the traffic will be large or large fluctuations will occur, as will congestion or even packet loss.
  • embodiments of the present invention provide a method, an apparatus, and a computer storage medium for managing congestion traffic.
  • An embodiment of the present invention provides a method for managing congestion traffic, where the method includes:
  • the switching device determines the congestion level according to the depth of the congestion of the packet transmission path
  • a standard token is sent to the congestion leaking bucket of the packet transmission path, and the depth of the token controlled leaky bucket is subtracted from the standard token value.
  • the congestion level is divided by a congestion threshold, and N congestion thresholds are divided into N+1 congestion levels, and each congestion level is set corresponding to a token addition value.
  • the method further includes: after determining the congestion level of the congested leaky bucket, the switching device waits for the first time according to the depth change of the congestion leaking bucket when the congestion level of the congestion leaking bucket falls to the next congestion level.
  • a congestion level update is performed; wherein the first time is a random waiting time constructed by a random sequence.
  • the method further includes: if the congestion level of the congestion leaking bucket rises to the original congestion level again in the first time, stopping waiting, and clearing the counter of the first time;
  • the congestion level of the congestion leaking bucket is decreased by multiple levels, then at the first time, the congestion level is lowered to the current congestion level of the congestion leaking bucket according to the second time interval;
  • the second time is the fixed time set.
  • the method further includes: when the switching device sends a standard token to the congestion leaking bucket of the packet transmission path, detecting a valid link number of the packet transmission path, and selecting a corresponding token according to the number of the effective links. The rate is delivered and the standard token is issued.
  • Another embodiment of the present invention provides a method for managing congestion traffic, where the method includes:
  • the switching device When the switching device sends a standard token to the congestion leaking packet of the packet transmission path, the number of valid links of the packet transmission path is detected, and the corresponding token delivery rate is selected according to the number of the effective links, and the standard token is delivered. .
  • the corresponding rate of the token is sent at a lower rate.
  • the rate of the tokens with a corresponding rate is selected.
  • An embodiment of the present invention provides a device for managing congestion traffic, where the device includes: a level determining module, an update depth module, and a token sending module;
  • a level determining module configured to determine a congestion level according to a depth of a congestion leaking bucket of the packet transmission path
  • Updating the depth module configured to add a value according to the token corresponding to the congestion level, and update the token to control the depth of the leaky bucket;
  • the token issuing module is configured to send a standard token to the congestion leaking bucket of the packet transmission path when the depth of the token control bucket reaches a standard token value, and control the token to leak the bucket Depth minus the standard token value.
  • the congestion level is divided by a congestion threshold, and N congestion thresholds are divided into N+1 congestion levels, and each congestion level is set corresponding to a token addition value.
  • the device further includes: a level update module, configured to: after determining a congestion level of the congestion leaking bucket, according to a depth change of the congestion leaking bucket, when the congestion level of the congestion leaking bucket falls to a next congestion level, wait for the first At a time, at the first time, a congestion level update is performed; wherein the first time is a random waiting time constructed by a random sequence.
  • a level update module configured to: after determining a congestion level of the congestion leaking bucket, according to a depth change of the congestion leaking bucket, when the congestion level of the congestion leaking bucket falls to a next congestion level, wait for the first At a time, at the first time, a congestion level update is performed; wherein the first time is a random waiting time constructed by a random sequence.
  • the level update module is further configured to: if the congestion level of the congestion leaking bucket rises to the original congestion level again in the first time, stop waiting, and clear the counter of the first time;
  • the congestion level of the congestion leaking bucket is decreased by multiple levels, then at the first time, the congestion level is lowered to the current congestion level of the congestion leaking bucket according to the second time interval;
  • the second time is the fixed time set.
  • the device further includes: an effective link number detecting module and a rate selecting module; wherein
  • the effective link number detecting module is configured to send a label to the congestion leaking bucket of the packet transmission path.
  • the quasi-token is used, the number of valid links of the packet transmission path is detected;
  • the rate selection module is configured to select a corresponding token delivery rate according to the number of the effective links, and deliver a standard token.
  • Another embodiment of the present invention provides a device for managing congestion traffic, where the device includes: an effective link number detecting module and a rate selecting module;
  • the effective link number detecting module is configured to detect the effective link number of the message transmission path when the standard token is sent to the congestion leaking bucket of the packet transmission path;
  • the rate selection module is configured to select a corresponding token delivery rate according to the number of the effective links, and deliver a standard token.
  • Embodiments of the present invention provide a computer storage medium storing a computer program for performing the foregoing method for managing congestion traffic.
  • the embodiment of the present invention provides a method and a device for managing congestion traffic, and a computer storage medium.
  • the switching device determines a congestion level according to the depth of the congestion of the packet transmission path, and adds a value according to the token corresponding to the congestion level.
  • the update token controls the depth of the leaky bucket.
  • a standard token is sent to the congestion leaking bucket of the packet transmission path, and the token is controlled to leak the bucket.
  • the depth of the standard minus the value of the standard token; in this way, in the case of a small exchange network cache, avoid large traffic bursts, smooth traffic, and reduce the pressure on the switching network.
  • FIG. 1 is a schematic flowchart of a method for managing congestion traffic according to an embodiment of the present disclosure
  • FIG. 2 is a schematic diagram of a bucket depth corresponding to a congestion threshold according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of a method for performing congestion traffic management by controlling a token sending rate according to an embodiment of the present disclosure
  • FIG. 4 is a schematic structural diagram of a device for managing congestion traffic according to an embodiment of the present disclosure
  • FIG. 5 is a schematic structural diagram of a management apparatus for controlling a token sending rate according to an embodiment of the present invention.
  • the switching device determines the congestion level according to the depth of the congestion leaking of the packet transmission path, and according to the token adding value corresponding to the congestion level, the update token controls the depth of the leaky bucket, and when the token is controlled to leak
  • the depth of the bucket reaches a standard token value
  • a standard token is sent to the congestion leaking bucket of the packet transmission path, and the depth of the token control leak bucket is subtracted from the standard token value.
  • the embodiment of the invention implements a method for managing congestion traffic. As shown in FIG. 1 , the method includes the following steps:
  • Step 101 The switching device determines a congestion level according to a depth of a congestion leaking bucket of the packet transmission path.
  • the congestion level is divided by a preset congestion threshold, and N+1 congestion levels may be divided by N congestion thresholds, and the congestion level may be represented by a control signal, and each congestion level is set in advance by a token. Value, the higher the congestion level, the smaller the corresponding token addition value.
  • 9 congestion levels are divided by 8 congestion thresholds TH0 to TH7, and 8 congestion thresholds TH0 to TH7 are in accordance with the depth of congestion. From 0 to the maximum (MAX), the 9 congestion levels can be represented by a 4-bit control signal, as shown in Table 1:
  • Step 102 The switching device adds a value according to the token corresponding to the congestion level, and the update token controls the depth of the leaky bucket.
  • the switching device adds a value of VALUE5 according to the congestion level corresponding to the congestion level 6, and adds the depth of the token control leak bucket to VALUE5.
  • Step 103 When the depth of the token control leaky bucket reaches a standard token value, the switching device sends a standard token to the congestion leaking bucket of the packet transmission path, and subtracts the depth of the token control leak bucket. Standard token value.
  • the method further includes: after the switching device determines the congestion level of the congested leaky bucket, according to the congestion
  • the depth of the leaky bucket changes, when the congestion level of the congestion leaking bucket drops to the next congestion level, waits for the first time, and at the first time arrives, performs a congestion level update; wherein the first time is constructed by a random sequence Random waiting time, the random sequence may be a configuration mask field, and some bits in the mask field are masked to generate a 16-bit pseudo-random sequence; thus, each level is randomly waited for a period of time before falling, and the staggered
  • the method further includes: stopping, in the first time, if the congestion level of the congestion leaking bucket rises to the original congestion level, stopping the waiting, and clearing the counter of the first time;
  • the congestion level of the congestion leaking bucket is decreased by multiple levels, then at the first time, the congestion level is lowered to the current congestion level of the congestion leaking bucket according to the second time interval;
  • the second time is a fixed time set in advance; thus, it can play a smoothing effect on multi-level bursts.
  • the method further includes: when the switching device sends a standard token to the congestion leaking bucket of the packet transmission path, detecting the number of valid links of the packet transmission path, and selecting a corresponding token sending rate according to the number of the effective links, The standard token is delivered; when the number of valid links is decreased, the corresponding rate of the token is delivered at a lower rate, and when the number of valid links is increased, the rate at which the corresponding rate is sent is selected.
  • the corresponding token delivery rate needs to be set in advance for different effective link numbers.
  • the embodiment of the invention also implements a method for managing congestion traffic. As shown in FIG. 3, the method includes the following steps:
  • Step 301 When the switching device sends a standard token to the congestion leaking bucket of the packet transmission path, the number of valid links of the packet transmission path is detected.
  • the effective link of the message transmission path can be detected by counting the number of valid links.
  • the number of the effective link is that when the link continuously receives the correct number of packets, the link is considered as a valid link.
  • Step 302 Select a corresponding token delivery rate according to the number of valid links, and deliver a standard token.
  • the corresponding token delivery rate is selected, and the effective chain is used.
  • the corresponding rate of token delivery is selected.
  • the method needs to set a corresponding token delivery rate for different effective link numbers in advance. The more the effective link number, the lower the set token delivery rate.
  • the embodiment of the present invention further provides a device for managing congestion traffic.
  • the device includes: a level determining module 41, an update depth module 42, and a token sending module 43.
  • the level determining module 41 is configured to determine a congestion level according to a depth of a congestion leaking bucket of the packet transmission path;
  • the update depth module 42 is configured to update the token to control the depth of the leaky bucket according to the token addition value corresponding to the congestion level;
  • the token issuing module 43 is configured to send a standard token to the congestion leaking bucket of the packet transmission path when the token controls the depth of the leaky bucket to reach a standard token value, and the token is controlled to leak the bucket.
  • the depth of the standard minus the value of the token;
  • the congestion level is divided by a preset congestion threshold, and N+1 congestion levels may be divided by N congestion thresholds.
  • the congestion level may be represented by a control signal, and each congestion level is preset to set a token addition value, and the congestion level is preset. The higher the corresponding token added value, the smaller.
  • the device further includes: a level update module 44, configured to wait for the first time when the congestion level of the congestion leaking bucket drops to the next congestion level after determining the congestion level of the congestion leaking bucket according to the congestion level of the congestion leaking bucket.
  • the congestion level update is performed; wherein the first time is a random waiting time constructed by a random sequence, and the random sequence may be a configuration mask field, and some bits in the mask field are masked to generate 16-bit pseudo-random sequence;
  • the level update module 44 is further configured to: if the congestion level of the congestion leaking bucket rises to the original congestion level again in the first time, stop waiting, and clear the counter of the first time;
  • the congestion level of the congestion leaking bucket is decreased by multiple levels, then at the first time, the congestion level is lowered to the current congestion level of the congestion leaking bucket according to the second time interval;
  • the second time is a fixed time set in advance.
  • the device further includes: an effective link number detecting module 45, a rate selecting module 46, wherein
  • the effective link number detecting module 45 is configured to detect the effective link number of the message transmission path when the standard token is sent to the congestion leaking bucket of the packet transmission path;
  • the rate selection module 46 is configured to select a corresponding token delivery rate according to the number of valid links, and deliver a standard token; when the number of valid links decreases, select a corresponding token delivery rate with a low rate. When the number of valid links increases, the corresponding rate of token delivery is selected.
  • the corresponding token delivery rate needs to be set in advance for different effective link numbers.
  • Each of the above modules can be implemented by a processor in a switching device in combination with a memory and an interface.
  • Each of the above modules may be implemented by a central processing unit (CPU, Central Processing Unit), a microprocessor (MPU, MicroProcessor Unit), or a digital signal processor (DSP, Digital Signal) located in a management device for congestion traffic. Processor), or Field-Programmable Gate Array (FPGA) implementation.
  • CPU Central Processing Unit
  • MPU Microprocessor
  • DSP Digital Signal
  • FPGA Field-Programmable Gate Array
  • the embodiment of the present invention further provides a device for managing congestion traffic.
  • the device includes: an effective link number detecting module 45, and a rate selecting module 46.
  • the effective link number detecting module 45 is configured to detect the effective link number of the message transmission path when the standard token is sent to the congestion leaking bucket of the packet transmission path;
  • the rate selection module 46 is configured to select a corresponding token delivery rate according to the number of valid links, and deliver a standard token; when the number of valid links decreases, select a corresponding token delivery rate with a low rate. When the number of valid links increases, the corresponding rate of token delivery is selected.
  • the corresponding token delivery rate needs to be set in advance for different effective link numbers.
  • Each of the above modules may be implemented by a central processing unit (CPU, Central Processing Unit), a microprocessor (MPU, MicroProcessor Unit), or a digital signal processor (DSP, Digital Signal) located in a management device for congestion traffic. Processor), or Field-Programmable Gate Array (FPGA) implementation.
  • CPU Central Processing Unit
  • MPU Microprocessor
  • DSP Digital Signal
  • FPGA Field-Programmable Gate Array
  • the embodiment of the present invention controls the delivery of the standard token according to the congestion level corresponding to the depth of the congestion leaking bucket, and can avoid a large traffic burst, smooth traffic, and mitigate the situation when the switching network cache is small.
  • the pressure of the exchange network is small.
  • the apparatus for tracking the service signaling may also be stored in a computer readable storage medium if it is implemented in the form of a software function module and sold or used as a separate product.
  • the technical solution of the embodiments of the present invention may be embodied in the form of a software product in essence or in the form of a software product stored in a storage medium, including a plurality of instructions.
  • a computer device (which may be a personal computer, server, or network device, etc.) is caused to perform all or part of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read only memory (ROM), a magnetic disk, or an optical disk.
  • program codes such as a USB flash drive, a mobile hard disk, a read only memory (ROM), a magnetic disk, or an optical disk.
  • an embodiment of the present invention further provides a computer storage medium, wherein a computer program is stored, and the computer program is used to execute a method for managing congestion traffic according to an embodiment of the present invention.

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明公开了一种拥塞流量的管理方法，交换设备根据报文传输路径的拥塞漏桶的深度，确定拥塞等级，根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度，当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值；本发明同时还公开了一种拥塞流量的管理装置及计算机存储介质。

Description

一种拥塞流量的管理方法、装置及计算机存储介质 技术领域
本发明涉及数据通信技术，尤其涉及一种拥塞流量的管理方法、装置及计算机存储介质。
背景技术
随着因特网业务的爆炸性增长，数据链路的流量越来越大，所需的交换设备的处理能力越来越高。目前的各种交换芯片通常面临着处理流量大，升级困难的问题，动则需要整个芯片系统架构进行颠覆性改动；这种架构的变动不仅需要投入大量人力进行重新开发、验证、测试，而且新项目的时间进度以及项目质量都受到非常大的影响。
在我们的使用过程中，实际情况为：当一个流量源向交换网开始发流或者多个流量源同时向交换网开始发流，会引起突发，而由于交换网的缓存不大，就不能很好的将突发吸收掉，导致拥塞甚至丢包。另外从停流到发流时，流量会很大或者会发生较大的波动，同样也会产生拥塞甚至丢包。
基于以上问题，如何在现有小缓存的情况下，避免较大的突发，平缓流量，减轻交换网的压力，是亟需解决的问题。
发明内容
为解决现有存在的技术问题，本发明实施例提供了一种拥塞流量的管理方法、装置及计算机存储介质。
本发明实施例的技术方案是这样实现的：
本发明实施例提供一种拥塞流量的管理方法，该方法包括：
交换设备根据报文传输路径的拥塞漏桶的深度，确定拥塞等级；
根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度；
当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值。
上述方案中，所述拥塞等级通过拥塞门限划分，N个拥塞门限划分出N+1个拥塞等级，每个拥塞等级对应设置一个令牌添加值，拥塞等级越高，对应的令牌添加值越小。
上述方案中，该方法还包括：交换设备在确定拥塞漏桶的拥塞等级后，根据拥塞漏桶的深度变化，当拥塞漏桶的拥塞等级下降到下一拥塞等级时，等待第一时间，在第一时间到时，进行拥塞等级更新；其中，所述第一时间为通过随机序列构造的随机等待时间。
上述方案中，该方法还包括：在第一时间内，如果拥塞漏桶的拥塞等级又升到原拥塞等级，则停止等待，将第一时间的计数器清零；
如果拥塞漏桶的拥塞等级又下降了多个等级，则在第一时间到时，按照第二时间为间隔，将拥塞等级一个等级一个等级地下降到拥塞漏桶的当前拥塞等级；所述第二时间为设置的固定时间。
上述方案中，该方法还包括：交换设备向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目，根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
本发明另一实施例提供一种拥塞流量的管理方法，该方法包括：
交换设备向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目，根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
上述方案中，当有效链路数目减少时，选择对应的速率低的令牌下发速率，当有效链路数目增多时，选择对应的速率高的令牌下发速率。
本发明实施例提供一种拥塞流量的管理装置，该装置包括：等级确定模块、更新深度模块、令牌下发模块；其中，
等级确定模块，配置为根据报文传输路径的拥塞漏桶的深度，确定拥塞等级；
更新深度模块，配置为根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度；
令牌下发模块，配置为当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值。
上述方案中，所述拥塞等级通过拥塞门限划分，N个拥塞门限划分出N+1个拥塞等级，每个拥塞等级对应设置一个令牌添加值，拥塞等级越高，对应的令牌添加值越小。
上述方案中，该装置还包括：等级更新模块，配置为在确定拥塞漏桶的拥塞等级后，根据拥塞漏桶的深度变化，当拥塞漏桶的拥塞等级下降到下一拥塞等级时，等待第一时间，在第一时间到时，进行拥塞等级更新；其中，所述第一时间为通过随机序列构造的随机等待时间。
上述方案中，所述等级更新模块，还配置为在第一时间内，如果拥塞漏桶的拥塞等级又升到原拥塞等级，则停止等待，将第一时间的计数器清零；
如果拥塞漏桶的拥塞等级又下降了多个等级，则在第一时间到时，按照第二时间为间隔，将拥塞等级一个等级一个等级地下降到拥塞漏桶的当前拥塞等级；所述第二时间为设置的固定时间。
上述方案中，该装置还包括：有效链路数目检测模块、速率选择模块；其中，
有效链路数目检测模块，配置为在向报文传输路径的拥塞漏桶下发标 准令牌时，检测报文传输路径的有效链路数目；
速率选择模块，配置为根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
本发明另一实施例提供一种拥塞流量的管理装置，该装置包括：有效链路数目检测模块、速率选择模块；其中，
有效链路数目检测模块，配置为在向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目；
速率选择模块，配置为根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
本发明实施例提供一种计算机存储介质，该计算机存储介质存储有计算机程序，该计算机程序用于执行上述拥塞流量的管理方法。
本发明实施例提供了一种拥塞流量的管理方法、装置及计算机存储介质，交换设备根据报文传输路径的拥塞漏桶的深度，确定拥塞等级，根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度，当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值；如此，能够在交换网缓存小的情况下，避免较大的流量突发，平缓流量，减轻交换网的压力。
附图说明
图1为本发明实施例提供的拥塞流量的管理方法的流程示意图；
图2为本发明实施例提供的拥塞门限对应的桶深示意图；
图3为本发明实施例提供的通过控制令牌下发速率进行拥塞流量管理的方法流程示意图；
图4为本发明实施例提供的拥塞流量的管理装置的结构示意图；
图5为本发明实施例提供的控制令牌下发速率的管理装置的结构示意 图。
具体实施方式
本发明实施例中，交换设备根据报文传输路径的拥塞漏桶的深度，确定拥塞等级，根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度，当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值。
下面通过附图及具体实施例对本发明做进一步的详细说明。
本发明实施例实现一种拥塞流量的管理方法，如图1所示，该方法包括以下几个步骤：
步骤101：交换设备根据报文传输路径的拥塞漏桶的深度，确定拥塞等级；
本步骤中，所述拥塞等级通过预先设置的拥塞门限划分，可以通过N个拥塞门限划分出N+1个拥塞等级，拥塞等级可以由控制信号表示，每个拥塞等级预先对应设置一个令牌添加值，拥塞等级越高，对应的令牌添加值越小，如图2所示，通过8个拥塞门限TH0～TH7划分出9个拥塞等级，8个拥塞门限TH0～TH7按照拥塞漏桶的深度从0到最大(MAX)依次增大，9个拥塞等级可以由4比特的控制信号表示，如表1所示：
Figure PCTCN2015080801-appb-000001
Figure PCTCN2015080801-appb-000002
步骤102：交换设备根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度；
例如：交换设备根据所述拥塞等级6对应的令牌添加值为VALUE5，将令牌控制漏桶的深度加上VALUE5。
步骤103：当令牌控制漏桶的深度达到一个标准令牌价值时，交换设备向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值。
上述方法还包括：交换设备在确定拥塞漏桶的拥塞等级后，根据拥塞 漏桶的深度变化，当拥塞漏桶的拥塞等级下降到下一拥塞等级时，等待第一时间，在第一时间到时，进行拥塞等级更新；其中，所述第一时间为通过随机序列构造的随机等待时间，所述随机序列可以是配置mask字段，将mask字段中的某些bit屏蔽，产生16bit的伪随机序列；这样，让每个等级在下降之前先随机等待一段时间，可以达到错开不同流量源的效果，并且伪随机序列的随机性越好，错开的效果越好，理想情况下就是所有流量源等待的时间都不相同，这样就能够完全错开。
该方法还包括：在第一时间内，如果拥塞漏桶的拥塞等级又升到原拥塞等级，则停止等待，将第一时间的计数器清零；
如果拥塞漏桶的拥塞等级又下降了多个等级，则在第一时间到时，按照第二时间为间隔，将拥塞等级一个等级一个等级地下降到拥塞漏桶的当前拥塞等级；所述第二时间为预先设置的固定时间；这样，能够起到对多等级突发的平滑作用。
该方法还包括：交换设备向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目，根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌；当有效链路数目减少时，选择对应的速率低的令牌下发速率，当有效链路数目增多时，选择对应的速率高的令牌下发速率；
这里，需要预先针对不同的有效链路数目设置对应的令牌下发速率，有效链路数目越多，设置的令牌下发速率越低。
本发明实施例还实现一种拥塞流量的管理方法，如图3所示，该方法包括以下几个步骤：
步骤301：交换设备向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目；
这里，可以通过统计有效链路的数量来检测报文传输路径的有效链路 数目，所述有效链路为当链路连续接收正确报文超过一定数量时，该链路即认定为有效链路。
步骤302：根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌；这里，当有效链路数目减少时，选择对应的速率低的令牌下发速率，当有效链路数目增多时，选择对应的速率高的令牌下发速率。
该方法需要预先针对不同的有效链路数目设置对应的令牌下发速率，有效链路数目越多，设置的令牌下发速率越低。
为了实现上述方法，本发明实施例还提供一种拥塞流量的管理装置，如图4所示，该装置包括：等级确定模块41、更新深度模块42、令牌下发模块43；其中，
等级确定模块41，配置为根据报文传输路径的拥塞漏桶的深度，确定拥塞等级；
更新深度模块42，配置为根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度；
令牌下发模块43，配置为当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值；
所述拥塞等级通过预先设置的拥塞门限划分，可以通过N个拥塞门限划分出N+1个拥塞等级，拥塞等级可以由控制信号表示，每个拥塞等级预先对应设置一个令牌添加值，拥塞等级越高，对应的令牌添加值越小。
该装置还包括：等级更新模块44，配置为在确定拥塞漏桶的拥塞等级后，根据拥塞漏桶的深度变化，当拥塞漏桶的拥塞等级下降到下一拥塞等级时，等待第一时间，在第一时间到时，进行拥塞等级更新；其中，所述第一时间为通过随机序列构造的随机等待时间，所述随机序列可以是配置mask字段，将mask字段中的某些bit屏蔽，产生16bit的伪随机序列；
所述等级更新模块44，还配置为在第一时间内，如果拥塞漏桶的拥塞等级又升到原拥塞等级，则停止等待，将第一时间的计数器清零；
如果拥塞漏桶的拥塞等级又下降了多个等级，则在第一时间到时，按照第二时间为间隔，将拥塞等级一个等级一个等级地下降到拥塞漏桶的当前拥塞等级；所述第二时间为预先设置的固定时间。
该装置还包括：有效链路数目检测模块45、速率选择模块46；其中，
有效链路数目检测模块45，配置为在向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目；
速率选择模块46，配置为根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌；当有效链路数目减少时，选择对应的速率低的令牌下发速率，当有效链路数目增多时，选择对应的速率高的令牌下发速率；
这里，需要预先针对不同的有效链路数目设置对应的令牌下发速率，有效链路数目越多，设置的令牌下发速率越低。
上述各模块均可以由交换设备中的处理器结合存储器和接口实现。
上述各个模块在实际应用中，均可由位于拥塞流量的管理装置中的中央处理器(CPU，Central Processing Unit)、或微处理器(MPU，MicroProcessor Unit)、或数字信号处理器(DSP，Digital Signal Processor)、或现场可编程门阵列(FPGA，Field－Programmable Gate Array)实现。
基于上述装置，本发明实施例还提供一种拥塞流量的管理装置，如图5所示，该装置包括：有效链路数目检测模块45、速率选择模块46；其中，
有效链路数目检测模块45，配置为在向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目；
速率选择模块46，配置为根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌；当有效链路数目减少时，选择对应的速率低的令牌下发速率，当有效链路数目增多时，选择对应的速率高的令牌下发速率；
这里，需要预先针对不同的有效链路数目设置对应的令牌下发速率，有效链路数目越多，设置的令牌下发速率越低。
上述各个模块在实际应用中，均可由位于拥塞流量的管理装置中的中央处理器(CPU，Central Processing Unit)、或微处理器(MPU，MicroProcessor Unit)、或数字信号处理器(DSP，Digital Signal Processor)、或现场可编程门阵列(FPGA，Field－Programmable Gate Array)实现。
综上所述，本发明实施例根据拥塞漏桶的深度对应的拥塞等级，控制标准令牌的下发，能够在交换网缓存小的情况下，避免较大的流量突发，平缓流量，减轻交换网的压力。
本发明实施例上述业务信令跟踪的装置如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实施例的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机、服务器、或者网络设备等)执行本发明各个实施例所述方法的全部或部分。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read Only Memory)、磁碟或者光盘等各种可以存储程序代码的介质。这样，本发明实施例不限制于任何特定的硬件和软件结合。
相应地，本发明实施例还提供一种计算机存储介质，其中存储有计算机程序，该计算机程序用于执行本发明实施例的拥塞流量的管理方法。
以上所述，仅为本发明的较佳实施例而已，并非用于限定本发明的保护范围，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (14)

  1. 一种拥塞流量的管理方法，该方法包括：
    交换设备根据报文传输路径的拥塞漏桶的深度，确定拥塞等级；
    根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度；
    当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值。
  2. 根据权利要求1所述的管理方法，其中，所述拥塞等级通过拥塞门限划分，N个拥塞门限划分出N+1个拥塞等级，每个拥塞等级对应设置一个令牌添加值，拥塞等级越高，对应的令牌添加值越小。
  3. 根据权利要求2所述的管理方法，其中，该方法还包括：交换设备在确定拥塞漏桶的拥塞等级后，根据拥塞漏桶的深度变化，当拥塞漏桶的拥塞等级下降到下一拥塞等级时，等待第一时间，在第一时间到时，进行拥塞等级更新；其中，所述第一时间为通过随机序列构造的随机等待时间。
  4. 根据权利要求3所述的管理方法，其中，该方法还包括：在第一时间内，如果拥塞漏桶的拥塞等级又升到原拥塞等级，则停止等待，将第一时间的计数器清零；
    如果拥塞漏桶的拥塞等级又下降了多个等级，则在第一时间到时，按照第二时间为间隔，将拥塞等级一个等级一个等级地下降到拥塞漏桶的当前拥塞等级；所述第二时间为设置的固定时间。
  5. 根据权利要求1所述的管理方法，其中，该方法还包括：交换设备向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目，根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
  6. 一种拥塞流量的管理方法，该方法包括：
    交换设备向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输 路径的有效链路数目，根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
  7. 根据权利要求6所述的管理方法，其中，当有效链路数目减少时，选择对应的速率低的令牌下发速率，当有效链路数目增多时，选择对应的速率高的令牌下发速率。
  8. 一种拥塞流量的管理装置，该装置包括：等级确定模块、更新深度模块、令牌下发模块；其中，
    等级确定模块，配置为根据报文传输路径的拥塞漏桶的深度，确定拥塞等级；
    更新深度模块，配置为根据所述拥塞等级对应的令牌添加值，更新令牌控制漏桶的深度；
    令牌下发模块，配置为当令牌控制漏桶的深度达到一个标准令牌价值时，向所述报文传输路径的拥塞漏桶下发一个标准令牌，并将令牌控制漏桶的深度减去标准令牌价值。
  9. 根据权利要求8所述的管理装置，其中，所述拥塞等级通过拥塞门限划分，N个拥塞门限划分出N+1个拥塞等级，每个拥塞等级对应设置一个令牌添加值，拥塞等级越高，对应的令牌添加值越小。
  10. 根据权利要求9所述的管理装置，其中，该装置还包括：等级更新模块，配置为在确定拥塞漏桶的拥塞等级后，根据拥塞漏桶的深度变化，当拥塞漏桶的拥塞等级下降到下一拥塞等级时，等待第一时间，在第一时间到时，进行拥塞等级更新；其中，所述第一时间为通过随机序列构造的随机等待时间。
  11. 根据权利要求10所述的管理装置，其中，所述等级更新模块，还配置为在第一时间内，如果拥塞漏桶的拥塞等级又升到原拥塞等级，则停止等待，将第一时间的计数器清零；
    如果拥塞漏桶的拥塞等级又下降了多个等级，则在第一时间到时，按照第二时间为间隔，将拥塞等级一个等级一个等级地下降到拥塞漏桶的当前拥塞等级；所述第二时间为设置的固定时间。
  12. 根据权利要求8所述的管理装置，其中，该装置还包括：有效链路数目检测模块、速率选择模块；其中，
    有效链路数目检测模块，配置为在向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目；
    速率选择模块，配置为根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
  13. 一种拥塞流量的管理装置，该装置包括：有效链路数目检测模块、速率选择模块；其中，
    有效链路数目检测模块，配置为在向报文传输路径的拥塞漏桶下发标准令牌时，检测报文传输路径的有效链路数目；
    速率选择模块，配置为根据所述有效链路数目选择对应的令牌下发速率，下发标准令牌。
  14. 一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，该计算机可执行指令配置为执行权利要求1-7任一项所述的拥塞流量的管理方法。
PCT/CN2015/080801 2014-12-24 2015-06-04 一种拥塞流量的管理方法、装置及计算机存储介质 Ceased WO2016101549A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410816492.1 2014-12-24
CN201410816492.1A CN105791155B (zh) 2014-12-24 2014-12-24 一种拥塞流量的管理方法和装置

Publications (1)

Publication Number Publication Date
WO2016101549A1 true WO2016101549A1 (zh) 2016-06-30

Family

ID=56149114

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/080801 Ceased WO2016101549A1 (zh) 2014-12-24 2015-06-04 一种拥塞流量的管理方法、装置及计算机存储介质

Country Status (2)

Country Link
CN (1) CN105791155B (zh)
WO (1) WO2016101549A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110365530A (zh) * 2019-07-11 2019-10-22 电子科技大学 一种独立于片上网络的测试令牌传递网络
CN112737970B (zh) * 2019-10-28 2024-06-14 华为技术有限公司 一种数据传输方法及相关设备
CN112825511B (zh) * 2019-11-20 2025-09-16 深圳市中兴微电子技术有限公司 一种交换网络拥塞管理方法、装置、设备和存储介质
CN112995058B (zh) * 2019-12-13 2023-11-24 深圳市中兴微电子技术有限公司 一种令牌的调整方法及装置
CN112838991B (zh) * 2021-02-09 2023-03-24 芯河半导体科技(无锡)有限公司 一种自适应调节小带宽调度器令牌价值的方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1859206A (zh) * 2006-03-24 2006-11-08 华为技术有限公司 一种剩余带宽复用的方法及网络设备
CN1870589A (zh) * 2005-05-26 2006-11-29 日立通讯技术株式会社 信息包传输装置及网络系统
US7965635B1 (en) * 2007-01-31 2011-06-21 Alcatel Lucent Per-class scheduling with rate limiting

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100426738C (zh) * 2005-03-16 2008-10-15 华为技术有限公司 一种数据流量约束控制的实现方法
CN101286947B (zh) * 2008-05-30 2010-12-01 杭州华三通信技术有限公司 数据输出控制方法和装置
CN101360049B (zh) * 2008-09-23 2011-02-09 福建星网锐捷网络有限公司 一种报文转发方法及装置
US8000235B2 (en) * 2008-10-05 2011-08-16 Contextream Ltd. Bandwidth allocation method and apparatus
CN102710514B (zh) * 2012-05-25 2015-01-21 华为技术有限公司 一种报文流量控制方法及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1870589A (zh) * 2005-05-26 2006-11-29 日立通讯技术株式会社 信息包传输装置及网络系统
CN1859206A (zh) * 2006-03-24 2006-11-08 华为技术有限公司 一种剩余带宽复用的方法及网络设备
US7965635B1 (en) * 2007-01-31 2011-06-21 Alcatel Lucent Per-class scheduling with rate limiting

Also Published As

Publication number Publication date
CN105791155A (zh) 2016-07-20
CN105791155B (zh) 2019-08-06

Similar Documents

Publication Publication Date Title
WO2016101549A1 (zh) 一种拥塞流量的管理方法、装置及计算机存储介质
US8826381B2 (en) Node device and method to prevent overflow of pending interest table in name based network system
CN110599136B (zh) 区块链交易池流量管控方法以及装置
CN106330951B (zh) 一种网络防护方法、装置和系统
CN101997766A (zh) 一种基于优先级的令牌桶限速的方法及系统
CN105045723A (zh) 一种缓存数据处理方法、装置和系统
CN104537285A (zh) 一种网络用户注册防刷方法及装置
CN109040140A (zh) 一种慢速攻击检测方法及装置
CN104079499A (zh) 基于令牌桶的报文处理方法及装置
CN104301248A (zh) 报文限速方法和装置
WO2016086542A1 (zh) 一种传输报文的方法及装置、计算机存储介质
WO2017067476A1 (zh) Mac地址处理方法及装置
CN111182551B (zh) 网络安全防护方法和系统
US20230224374A1 (en) Adaptive Signaling for Network Performance Measurement, Access, and Control
CN109756566B (zh) 基于区块链的数据存储方法、相关设备及存储介质
CN106254394B (zh) 一种攻击流量的记录方法和装置
CN1206880C (zh) 通信系统
CN111478860A (zh) 一种网络控制方法、装置、设备及机器可读存储介质
CN107395554B (zh) 流量攻击的防御处理方法及装置
US12079201B2 (en) Blockchain-based data storage method, related device and storage medium
KR20230156262A (ko) 기계 학습 기반 멀웨어 검출을 위한 시스템 및 방법
CN103905335A (zh) 流量控制方法及装置
CN107800683A (zh) 一种挖掘恶意ip的方法及装置
CN105959300B (zh) 一种DDoS攻击防护的方法及装置
CN109617893B (zh) 一种僵尸网络DDoS攻击的防护方法、装置及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15871624

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15871624

Country of ref document: EP

Kind code of ref document: A1