WO2022042282A1 - 一种无损流量拥塞自适应方法、系统和网络设备 - Google Patents

一种无损流量拥塞自适应方法、系统和网络设备 Download PDF

Info

Publication number
WO2022042282A1
WO2022042282A1 PCT/CN2021/111568 CN2021111568W WO2022042282A1 WO 2022042282 A1 WO2022042282 A1 WO 2022042282A1 CN 2021111568 W CN2021111568 W CN 2021111568W WO 2022042282 A1 WO2022042282 A1 WO 2022042282A1
Authority
WO
WIPO (PCT)
Prior art keywords
traffic
ecn
incast
value
sensitive
Prior art date
Application number
PCT/CN2021/111568
Other languages
English (en)
French (fr)
Inventor
刘毅
杨庆
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2022042282A1 publication Critical patent/WO2022042282A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/11—Identifying congestion
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/12—Avoiding congestion; Recovering from congestion
    • H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS

Definitions

  • the embodiments of the present application relate to the field of communications, and in particular, to a lossless traffic congestion adaptive method, system, and network device.
  • PFC Priority-based Flow Control
  • ECN Explicit Congestion Notification
  • the traditional ECN function is to manually set a static threshold value, which is easy to trigger the PFC threshold value when the network is congested, resulting in aggravated network congestion.
  • the embodiments of the present application provide a lossless traffic congestion adaptive method, system, and network device.
  • An embodiment of the present application provides a lossless traffic congestion adaptive method, including: performing statistical analysis on transmitted network traffic, and obtaining traffic analysis parameters, wherein the traffic analysis parameters include a traffic incast value, a delay-sensitive traffic ratio, a throughput Sensitive traffic ratio; dynamically adjust the threshold for displaying the congestion notification ECN according to the obtained traffic analysis parameters; according to the ECN threshold, the source automatically adjusts the traffic sending window.
  • the embodiment of the present application also provides a lossless traffic congestion adaptive system, including:
  • the traffic analysis module is configured to analyze the transmitted network traffic, obtain traffic analysis parameters and send the traffic analysis parameters to the traffic queue management module, wherein the traffic analysis parameters include a traffic incast value, a delay-sensitive traffic ratio , the throughput-sensitive traffic ratio; the traffic queue management module is set to dynamically adjust the threshold for displaying the congestion notification ECN according to the traffic analysis parameters obtained by the traffic analysis module; the traffic window module is set to be set according to the The ECN threshold value dynamically adjusted by the traffic queue management module automatically adjusts the traffic sending window.
  • the traffic analysis parameters include a traffic incast value, a delay-sensitive traffic ratio , the throughput-sensitive traffic ratio
  • the traffic queue management module is set to dynamically adjust the threshold for displaying the congestion notification ECN according to the traffic analysis parameters obtained by the traffic analysis module
  • the traffic window module is set to be set according to the The ECN threshold value dynamically adjusted by the traffic queue management module automatically adjusts the traffic sending window.
  • An embodiment of the present application further provides a lossless traffic congestion adaptive network device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores data that can be used by the at least one processor. Instructions executed by a processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the lossless traffic congestion adaptation method described above.
  • FIG. 1 is a flowchart of a lossless traffic congestion adaptive method provided by a first embodiment of the present application
  • FIG. 2 is a flowchart of a lossless traffic congestion adaptive method provided by a second embodiment of the present application
  • FIG. 3 is a flowchart of a lossless traffic congestion adaptive method provided by a third embodiment of the present application.
  • FIG. 4 is a flowchart of a lossless traffic congestion adaptive method provided by a fourth embodiment of the present application.
  • FIG. 5 is a flowchart of a lossless traffic congestion adaptive method provided by a fifth embodiment of the present application.
  • FIG. 6 is a flowchart of a lossless traffic congestion adaptive method provided by the sixth embodiment of the present application.
  • FIG. 7 is a schematic structural diagram of a lossless traffic congestion adaptive system provided by a seventh embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of a network device provided by an eighth embodiment of the present application.
  • the first embodiment of the present application relates to a lossless traffic congestion adaptive method.
  • the specific process is shown in FIG. 1 , including: Step 101 , analyze the transmitted network traffic, and obtain traffic analysis parameters, wherein the traffic analysis parameters include traffic incast value, delay-sensitive traffic ratio, throughput-sensitive traffic ratio;
  • step 101 can analyze the transmitted network traffic through the Ethernet device, for example, through the switch, perform statistical analysis on the network packets entering and leaving each port, and obtain the traffic analysis parameters.
  • the traffic analysis parameters include the traffic incast value, Delay-sensitive traffic ratio, throughput-sensitive traffic ratio, etc.
  • the traffic analysis parameters may also include other parameters in the actual use process, which will not be repeated here.
  • this embodiment does not limit the method for analyzing network traffic, and any existing traffic analysis method may be used in the actual use process, which will not be repeated here.
  • Step 102 Dynamically adjust the threshold for displaying the congestion notification ECN according to the acquired traffic analysis parameters
  • Step 103 according to the ECN threshold, the source automatically adjusts the traffic sending window.
  • step 103 includes sending an ACK response message carrying the ECE flag to the source according to the ECN threshold value; after the source receives the ACK response message, checking the ECE flag; if the ECE flag is If it is marked as a congested state, the source starts to automatically adjust the traffic sending window.
  • the traffic analysis parameters are obtained, and the ECN threshold is dynamically adjusted according to the traffic analysis parameters, including the traffic incast value, the delay-sensitive traffic ratio, and the throughput-sensitive traffic ratio.
  • the problem of triggering the PFC function is caused by the long backpressure duration of the ECN response message.
  • the second embodiment of the present application relates to a lossless traffic congestion adaptive method, which is basically the same as the lossless traffic congestion adaptive method provided by the first embodiment of the present application.
  • the difference is that, as shown in FIG. 2 , the steps 101 includes:
  • Step 201 Perform statistical analysis on the transmitted network traffic to obtain a traffic incast value.
  • the traffic incast value changes dynamically in real time, and the larger the traffic incast value is, the more serious the network congestion is at this time.
  • Step 202 Distinguish the types of network traffic according to the attributes of the data packets, and obtain the proportion of delay-sensitive traffic and the proportion of throughput-sensitive traffic.
  • the methods for obtaining the proportion of delay-sensitive traffic and the proportion of throughput-sensitive traffic are not limited, and any existing method for obtaining the proportion of traffic types may be used in actual use, which will not be repeated here.
  • the attributes of the data packet may include the length of the data packet, the type of the data packet, and the like, which will not be described in detail here.
  • the types of network traffic are distinguished according to the attributes of the data packets, and the proportion of delay-sensitive traffic and the proportion of throughput-sensitive traffic is obtained, which ensures the subsequent During the dynamic adjustment of the ECN threshold, the network's requirements for delay and throughput are fully considered.
  • the third embodiment of the present application relates to a lossless traffic congestion adaptive method, which is basically the same as the lossless traffic congestion adaptive method provided by the first embodiment of the present application. The difference is that, as shown in FIG. 3 , the steps 102 includes:
  • Step 301 Determine whether the traffic incast value is greater than a preset incast value.
  • step 302 is performed to reduce the ECN threshold; when the traffic incast value is less than the preset incast value, step 303 is performed to increase the ECN threshold.
  • Th represents the dynamic ECN threshold value
  • E represents the default ECN threshold set in the initial state
  • a represents the influence coefficient of the incast value on the ECN threshold. It should be noted that the preset incast value, the default ECN threshold in the initial state, and the influence coefficient of the incast value on the ECN threshold can be set and flexibly changed according to different needs of users, network environment, and actual application scenarios.
  • the embodiments of the present application determine the network congestion by comparing the current traffic incast value with the preset incast value, and select different methods according to different situations to perform the ECN threshold.
  • the regulation ensures the dynamic adjustment of the ECN threshold.
  • the fourth embodiment of the present application relates to a lossless traffic congestion adaptive method, which is basically the same as the lossless traffic congestion adaptive method provided by the first embodiment of the present application. The difference is that, as shown in FIG. 4 , the steps 102 includes:
  • Step 401 Determine which type of traffic ratio is greater than a preset condition for the delay-sensitive traffic ratio and the throughput-sensitive traffic ratio.
  • step 402 when the proportion of the delay-sensitive traffic is greater than the preset condition, step 402 is performed to reduce the ECN threshold; when the proportion of the throughput-sensitive traffic is greater than the preset condition, step 403 is performed to increase the ECN threshold.
  • b represents the influence coefficient of the proportion of delay-sensitive traffic on the ECN threshold
  • c represents the influence coefficient of the proportion of throughput-sensitive traffic on the ECN threshold
  • R S represents the proportion of delay-sensitive traffic
  • RH represents the proportion of throughput-sensitive traffic. Proportion.
  • the ECN threshold when increasing the influence coefficient b of the delay-sensitive traffic proportion on the ECN threshold and the influence coefficient c of the throughput-sensitive traffic proportion on the ECN threshold, the ECN threshold can be dynamically adjusted by increasing b and c according to the proportion of traffic types.
  • the above are only specific examples, and any existing adjustment methods or standards may be used in the actual use process, which will not be repeated here.
  • the embodiments of the present application on the basis of realizing the beneficial effects brought by the first embodiment, respectively consider the delay-sensitive traffic and throughput-sensitive traffic, and set the delay-sensitive traffic ratio to the ECN threshold according to the different needs of users and the actual application environment.
  • the influence coefficient b and the influence coefficient c of the throughput-sensitive traffic ratio on the ECN threshold not only ensure the requirement of dynamic adjustment of the ECN threshold, but also take into account the requirements of different traffic for delay and throughput.
  • the fifth embodiment of the present application relates to a lossless traffic congestion adaptive method, which is basically the same as the lossless traffic congestion adaptive method provided by the first embodiment of the present application. The difference is that, as shown in FIG. 5 , the steps 102 includes:
  • Step 501 whether the traffic incast value is greater than the preset incast value.
  • step 502 is performed to reduce the ECN threshold; when the traffic incast value is less than the preset incast value, step 503 is performed to increase the ECN threshold.
  • the current traffic incast value and the preset incast value are used to judge the network congestion, and the delay problem and throughput problem are also taken into consideration.
  • the incast value is too large, set a low ECN threshold to meet the low latency requirement of the traffic in the queue.
  • the incast value is too small, increase the ECN threshold to ensure high traffic throughput requirements.
  • the method of dynamically adjusting the ECN threshold is further optimized.
  • the sixth embodiment of the present application relates to a lossless traffic congestion adaptive method, which is basically the same as the lossless traffic congestion adaptive method provided by the first embodiment of the present application, with the difference that, as shown in FIG. 6 , the steps 102 includes:
  • Step 601 determine which type of traffic ratio is greater than a preset condition between the delay-sensitive traffic ratio and the throughput-sensitive traffic ratio.
  • step 602 when the proportion of the delay-sensitive traffic is greater than the preset condition, step 602 is performed to reduce the ECN threshold; when the proportion of the throughput-sensitive traffic is greater than the preset condition, step 603 is performed to increase the ECN threshold.
  • the ECN threshold is regulated according to the proportion of the traffic types, so as to reduce the impact of congestion on traffic forwarding. At the same time, it also meets the requirements of different traffic types on latency and throughput.
  • the seventh embodiment of the present application relates to a lossless traffic congestion adaptive system, as shown in FIG. 7 , including:
  • the traffic analysis module 701 is configured to analyze the transmitted network traffic, obtain traffic analysis parameters and send the traffic analysis parameters to the traffic queue management module 702, wherein the traffic analysis parameters include the traffic incast value, which is time-sensitive Traffic ratio, throughput sensitive traffic ratio;
  • the traffic queue management module 702 is configured to dynamically adjust the threshold for displaying the congestion notification ECN according to the traffic analysis parameters obtained by the traffic analysis module 701;
  • the traffic window adjustment module 703 is configured to automatically adjust the traffic sending window according to the ECN threshold value dynamically adjusted by the traffic queue management module 702 .
  • the eighth embodiment of the present application relates to a network device, as shown in FIG. 8 , including:
  • At least one processor 801 and,
  • a memory 802 in communication with the at least one processor 801;
  • the memory 802 stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor 801 to enable the at least one processor 801 to execute the first to sixth embodiments of the present application
  • the lossless traffic congestion adaptive method described in the method is not limited
  • the memory and the processor are connected by a bus, and the bus may include any number of interconnected buses and bridges, and the bus connects one or more processors and various circuits of the memory.
  • the bus may also connect together various other circuits, such as peripherals, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein.
  • the bus interface provides the interface between the bus and the transceiver.
  • a transceiver may be a single element or multiple elements, such as multiple receivers and transmitters, providing a means for communicating with various other devices over a transmission medium.
  • the data processed by the processor is transmitted on the wireless medium through the antenna, and further, the antenna also receives the data and transmits the data to the processor.
  • the processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. Instead, memory may be used to store data used by the processor in performing operations.
  • a lossless traffic congestion adaptive method, system and network device proposed in the present application obtain traffic analysis parameters by analyzing the transmitted network traffic.
  • the traffic ratio is used to dynamically adjust the ECN threshold, which avoids the problem of triggering the PFC function caused by the long backpressure duration of the ECN response packet.

Abstract

一种无损流量拥塞自适应方法、系统和网络设备。无损流量拥塞自适应方法包括：对传送的网络流量进行分析，获取流量分析参数，其中，所述流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例(101)；根据所述获取的流量分析参数，动态调整显示拥塞通知ECN的门限值(102)；根据所述ECN门限值，源端自动调节流量发送窗口(103)。

Description

一种无损流量拥塞自适应方法、系统和网络设备
相关申请的交叉引用
本申请基于申请号为202010889189.X、申请日为2020年08月28日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请实施例涉及通信领域，特别涉及一种无损流量拥塞自适应方法、系统和网络设备。
背景技术
当多级网络发生拥塞时，目前通用的解决方案是同时部署基于优先级的流量控制(Priority-based Flow Control，PFC)功能和显示拥塞通知(Explicit Congestion Notification，ECN)功能。PFC是一种基于队列拥塞的功能，当队列的拥塞长度达到其阈值时，触发PFC，并向上游设备发送反压包，直至源端设备收到反压包后，会降低对应优先级流量的发送速率。ECN是当网络设备的无损队列出现拥塞，即队列已使用的缓存超过ECN的门限值时，网络设备在转发的报文中打上ECN标签，接收端收到带有ECN拥塞标记的报文后，向源端发送拥塞通知报文，源端收到报文降低发送速率。
然而，传统的ECN功能是手工设置静态的门限值，这种方案在网络拥塞时容易触发PFC的门限值，导致网络拥塞加剧。
发明内容
本申请实施例提出一种无损流量拥塞自适应方法、系统和网络设备。
本申请实施例提供了一种无损流量拥塞自适应方法，包括：对传送的网络流量进行统计分析，获取流量分析参数，其中，所述流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例；根据所述获取的流量分析参数，动态调整显示拥塞通知ECN的门限值；根据所述ECN门限值，源端自动调节流量发送窗口。
本申请实施例还提供了一种无损流量拥塞自适应系统，包括：
流量分析模块，被设置成对传送的网络流量进行分析，获取流量分析参数并将所述流量分析参数发送给流量队列管理模块，其中，所述流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例；流量队列管理模块，被设置成根据所述流量分析模块获取的所述流量分析参数，动态调整显示拥塞通知ECN的门限值；调整流量窗口模块，被设置成根据所述流量队列管理模块动态调整的ECN门限值，自动调节流量发送窗口。
本申请实施例还提供了一种无损流量拥塞自适应网络设备，包括：至少一个处理器；以及，与所述至少一个处理器通信连接的存储器；其中，所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器 能够执行以上所述的无损流量拥塞自适应方法。
附图说明
图1是本申请的第一实施方式提供的无损流量拥塞自适应方法的流程图；
图2是本申请的第二实施方式提供的无损流量拥塞自适应方法的流程图；
图3是本申请的第三实施方式提供的无损流量拥塞自适应方法的流程图；
图4是本申请的第四实施方式提供的无损流量拥塞自适应方法的流程图；
图5是本申请的第五实施方式提供的无损流量拥塞自适应方法的流程图；
图6是本申请的第六实施方式提供的无损流量拥塞自适应方法的流程图；
图7是本申请的第七实施方式提供的无损流量拥塞自适应系统的结构示意图；
图8是本申请的第八实施方式提供的网络设备的结构示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合附图对本申请的各实施例进行详细的阐述。然而，本领域的普通技术人员可以理解，在本申请各实施例中，为了使读者更好地理解本申请而提出了许多技术细节。但是，即使没有这些技术细节和基于以下各实施例的种种变化和修改，也可以实现本申请所要求保护的技术方案。以下各个实施例的划分是为了描述方便，不应对本申请的具体实现方式构成任何限定，各个实施例在不矛盾的前提下可以相互结合相互引用。
本申请的第一实施方式涉及一种无损流量拥塞自适应方法，具体流程如图1所示，包括：步骤101，对传送的网络流量进行分析，获取流量分析参数，其中，流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例；
在本实施方式中，步骤101可以通过以太网设备对传送的网络流量进行分析，比如，通过交换机对各个端口进出的网络报文进行统计分析，获取流量分析参数，流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例等。当然，以上仅为具体的举例说明，在实际使用过程中流量分析参数还可以包括其他参数，此处不做一一赘述。
需要说明的是，本实施方式不对分析网络流量的方法进行限定，在实际使用过程中可以使用任何现有的流量分析方法，此处不做一一赘述。
步骤102，根据获取的流量分析参数，动态调整显示拥塞通知ECN的门限值；
步骤103，根据ECN门限值，源端自动调节流量发送窗口。
在本实施方式中，步骤103包括根据ECN门限值，向源端发送携带ECE标志位的ACK响应报文；在源端接收到ACK响应报文后，检查ECE标志位；若ECE标志位被标记为拥塞状态，则源端开始自动调节流量发送窗口。
本申请的实施方式，通过对传送的网络流量进行分析，获取流量分析参数，根据流量分析参数，包括流量incast值，时延敏感流量比例，吞吐敏感流量比例，对ECN门限进行动态调整，避免了由于ECN响应报文较长的反压时长造成的触发PFC功能的问题。
本申请的第二实施方式涉及一种无损流量拥塞自适应方法，该方法与本申请的第一实 施方式提供的无损流量拥塞自适应方法基本相同，区别之处在于，如图2所示，步骤101包括：
步骤201，对传送的网络流量进行统计分析，获取流量incast值。
在本实施方式中，流量incast值是实时动态变化的，流量incast值越大，表示此时网络拥塞的情况越严重。
步骤202，根据数据包的属性对网络流量的类型进行区分，获取时延敏感流量比例和吞吐敏感流量比例。
在本实施方式中，不对获取时延敏感流量比例和吞吐敏感流量比例的方法进行限定，在实际使用过程中可以使用任何现有的获取流量类型所占比例的方法，此处不做赘述。另外，数据包的属性可以包括数据包的长度，数据包的类型等等，此处不做赘述。
本申请的实施方式，在实现第一实施方式带来的有益效果基础上，根据数据包的属性对网络流量的类型进行区分，获取时延敏感流量比例和吞吐敏感流量比例，保证了在后续对ECN门限进行动态调整过程中，充分考虑到网络对时延的要求和对吞吐量的要求。
本申请的第三实施方式涉及一种无损流量拥塞自适应方法，该方法与本申请的第一实施方式提供的无损流量拥塞自适应方法基本相同，区别之处在于，如图3所示，步骤102包括：
步骤301，判断流量incast值是否大于预设的incast值。
在本实施方式中，通过比较流量incast值和预设的incast值大小，判断网络拥塞情况，选择合适的调控方法。当流量incast值大于预设的incast值时，执行步骤302，降低ECN门限值；当流量incast值小于预设的incast值时，执行步骤303，提高ECN的门限值。
步骤302，通过公式Th＝E-incast*a降低ECN门限值。
步骤303，通过公式Th＝E+incast*a提高ECN门限值。
在本实施方式中，Th表示动态ECN门限值，E表示初始状态下设置的默认ECN门限，a表示incast值对ECN门限影响系数。需要说明的是，预设的incast值，初始状态下默认的ECN门限，incast值对ECN门限的影响系数都可以根据用户的不同需求，网络环境，实际应用场景等自行设定，灵活变化。
本申请的实施方式，在实现第一实施方式带来的有益效果基础上，通过把当前流量incast值与预设的incast值比较，判断网络拥塞情况，根据不同情况选择不同的方式对ECN门限进行调控，保证了对ECN门限的动态调整。
本申请的第四实施方式涉及一种无损流量拥塞自适应方法，该方法与本申请的第一实施方式提供的无损流量拥塞自适应方法基本相同，区别之处在于，如图4所示，步骤102包括：
步骤401，判断时延敏感流量比例和吞吐敏感流量比例哪种类型流量比例大于预设条件。
在本实施方式中，当时延敏感流量比例大于预设的条件时，执行步骤402，降低ECN门限；当吞吐敏感流量比例大于预设的条件时，执行步骤403，提高ECN门限。
步骤402，采用公式Th＝E-R S*b，通过提高时延敏感流量比例对ECN门限影响系数b，降低ECN门限值。
步骤403，采用公式Th＝E+R H*c，通过提高吞吐敏流量比例对ECN门限的影响系 数c，提高ECN门限值。
在本实施方式中，b表示时延敏感流量比例对ECN门限影响系数，c表示吞吐敏感流量比例对ECN门限影响系数，R S表示时延敏感流量所占比例，R H表示吞吐敏感流量所占比例。
具体地，在提高时延敏感流量比例对ECN门限影响系数b和吞吐敏流量比例对ECN门限的影响系数c时，可以按照流量类型的比例对b，c进行增大来动态调整ECN门限。当然，以上仅为具体的举例说明，在实际使用过程中可以按照任何现有的调整方法或标准，此处不作赘述。
本申请的实施方式，在实现第一实施方式带来的有益效果基础上，分别考虑到了时延敏感流量和吞吐敏感流量，根据用户不同需求和实际应用环境设定时延敏感流量比例对ECN门限影响系数b和吞吐敏流量比例对ECN门限的影响系数c，既保证了动态调整ECN门限的要求，又考虑到了满足不同的流量对时延和吞吐量的需求。
本申请的第五实施方式涉及一种无损流量拥塞自适应方法，该方法与本申请的第一实施方式提供的无损流量拥塞自适应方法基本相同，区别之处在于，如图5所示，步骤102包括：
步骤501，流量incast值是否大于预设的incast值。
在本实施方式中，通过比较流量incast值和预设的incast值大小，判断网络拥塞情况，选择合适的调控方法。当流量incast值大于预设的incast值时，执行步骤502，降低ECN门限值；当流量incast值小于预设的incast值时，执行步骤503，提高ECN的门限值。
步骤502，通过公式Th＝E-incast*a-R S*b+R H*c降低ECN门限值。
步骤503，通过公式Th＝E+incast*a-R S*b+R H*c提高ECN门限；
本申请实施方式，在实现第一实施方式带来的有益效果基础上，通过当前流量incast值和预设的incast值的大小，判断网络拥塞情况，同时又考虑到了时延问题和吞吐问题，当incast值偏大时，设置低的ECN门限，满足队列中流量的低时延要求，当incast值偏小时，提高ECN门限，保证流量的高吞吐性要求。进一步优化了动态调整ECN门限值的方法。
本申请的第六实施方式涉及一种无损流量拥塞自适应方法，该方法与本申请的第一实施方式提供的无损流量拥塞自适应方法基本相同，区别之处在于，如图6所示，步骤102包括：
步骤601，判断时延敏感流量比例和吞吐敏感流量比例哪种类型流量比例大于预设条件。
在本实施方式中，在本实施方式中，当时延敏感流量比例大于预设的条件时，执行步骤602，降低ECN门限；当吞吐敏感流量比例大于预设的条件时，执行步骤603，提高ECN门限。
步骤602，采用公式Th＝E-incast*a-R S*b+R H*c，通过增大时延敏感流量比例对ECN门限影响系数b，减小吞吐敏感流量比例对ECN门限影响系数c，降低ECN门限值。
步骤603，采用公式Th＝E+incast*a-R S*b+R H*c，通过减小时延敏感流量比例对ECN门限影响系数b，增大吞吐敏感流量比例对ECN门限影响系数c，提高ECN门限值。
本申请实施方式，在实现第一实施方式带来的有益效果基础上，通过对网络中的流量 类型进行分析，根据流量类型所占比例对ECN门限进行调控，在减小拥塞对流量转发影响的同时，又满足了不同流量类型对时延性和吞吐性的要求。
本申请的第七实施方式涉及一种无损流量拥塞自适应系统，如图7所示，包括：
流量分析模块701，被设置成对传送的网络流量进行分析，获取流量分析参数并将所述流量分析参数发送给流量队列管理模块702，其中，所述流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例；
所述流量队列管理模块702，被设置成根据所述流量分析模块701获取的所述流量分析参数，动态调整显示拥塞通知ECN的门限值；
调整流量窗口模块703，被设置成根据所述流量队列管理模块702动态调整的ECN门限值，自动调节流量发送窗口。
本申请第八实施方式涉及一种网络设备，如图8所示，包括：
至少一个处理器801；以及，
与所述至少一个处理器801通信连接的存储器802；其中，
所述存储器802存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器801执行，以使所述至少一个处理器801能够执行本申请第一至第六实施方式所述的无损流量拥塞自适应方法。
其中，存储器和处理器采用总线方式连接，总线可以包括任意数量的互联的总线和桥，总线将一个或多个处理器和存储器的各种电路连接在一起。总线还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路连接在一起，这些都是本领域所公知的，因此，本文不再对其进行进一步描述。总线接口在总线和收发机之间提供接口。收发机可以是一个元件，也可以是多个元件，比如多个接收器和发送器，提供用于在传输介质上与各种其他装置通信的单元。经处理器处理的数据通过天线在无线介质上进行传输，进一步，天线还接收数据并将数据传送给处理器。
处理器负责管理总线和通常的处理，还可以提供各种功能，包括定时，外围接口，电压调节、电源管理以及其他控制功能。而存储器可以被用于存储处理器在执行操作时所使用的数据。
本申请提出的一种无损流量拥塞自适应方法、系统和网络设备，通过对传送的网络流量进行分析，获取流量分析参数，根据流量分析参数，包括流量incast值，时延敏感流量比例，吞吐敏感流量比例，对ECN门限进行动态调整，避免了由于ECN响应报文较长的反压时长造成的触发PFC功能的问题。
本领域的普通技术人员可以理解，上述各实施方式是实现本申请的具体实施例，而在实际应用中，可以在形式上和细节上对其作各种改变，而不偏离本申请的精神和范围。

Claims (10)

  1. 一种无损流量拥塞自适应方法，包括：
    对传送的网络流量进行统计分析，获取流量分析参数，其中，所述流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例；
    根据所述获取的流量分析参数，动态调整显示拥塞通知ECN的门限值；
    根据所述ECN门限值，源端自动调节流量发送窗口。
  2. 根据权利要求1所述的无损流量拥塞自适应方法，其中，对传送的网络流量进行分析，获取流量分析参数，包括:
    对传送的网络流量进行统计分析，获取流量incast值；
    根据数据包的属性对所述网络流量的类型进行区分，获取所述时延敏感流量比例和所述吞吐敏感流量比例。
  3. 根据权利要求1所述的无损流量拥塞自适应方法，其特征在于，所述根据所述获取的流量分析参数，动态调整显示拥塞通知ECN的门限值，包括：
    比较所述流量incast值和预设的incast值的大小；
    若所述流量incast值大于预设的incast值，则通过公式Th＝E-incast*a降低ECN门限值；
    若所述流量incast值小于预设的incast值，则通过公式Th＝E+incast*a提高ECN门限；
    其中，Th表示动态ECN门限值，E表示初始状态下设置的默认ECN门限，a表示incast值对ECN门限影响系数。
  4. 根据权利要求1所述的无损流量拥塞自适应方法，其中，所述根据所述获取的流量分析参数，动态调整显示拥塞通知ECN的门限值，包括：
    比较所述时延敏感流量比例，所述吞吐敏感流量比例与预设条件的大小；
    若所述时延敏感流量比例大于预设的条件，则采用公式Th＝E-R S*b，通过提高时延敏感流量比例对ECN门限影响系数b，降低ECN门限值；
    若所述吞吐敏感流量比例大于预设的条件，则采用公式Th＝E+R H*c，通过提高吞吐敏流量比例对ECN门限的影响系数c，提高ECN门限值；
    其中，Th表示动态ECN门限值，E表示初始状态下设置的默认ECN门限，b表示时延敏感流量比例对ECN门限影响系数，c表示吞吐敏感流量比例对ECN门限影响系数，R S表示时延敏感流量所占比例，R H表示吞吐敏感流量所占比例。
  5. 根据权利要求1所述的无损流量拥塞自适应方法，其中，所述根据所述获取的流量分析参数，动态调整显示拥塞通知ECN的门限值，包括：
    比较所述流量incast值和预设的incast值的大小；
    若所述流量incast值大于预设的incast值，则通过公式Th＝E-incast*a-R s*b+R H*c降低ECN门限值；
    若所述流量incast值小于预设的incast值，则通过公式Th＝E+incast*a-R s*b+R H*c提高ECN门限；
    其中，Th表示动态ECN门限值，E表示初始状态下设置的默认ECN门限，a表示incast值对ECN门限影响系数，b表示时延敏感流量比例对ECN门限影响系数，c表示吞吐敏感 流量比例对ECN门限影响系数，R S表示时延敏感流量所占比例，R H表示吞吐敏感流量所占比例。
  6. 根据权利要求1所述的无损流量拥塞自适应方法，其中，所述根据所述获取的流量分析参数，动态调整显示拥塞通知ECN的门限值，包括：
    若所述时延敏感流量比例大于预设的条件，则采用公式Th＝E-incast*a-R s*b+R H*c，通过增大时延敏感流量比例对ECN门限影响系数b，减小吞吐敏感流量比例对ECN门限影响系数c，降低ECN门限值；
    若所述吞吐敏感流量比例大于预设的条件，则采用公式Th＝E+incast*a-R s*b+R H*c，通过减小时延敏感流量比例对ECN门限影响系数b，增大吞吐敏感流量比例对ECN门限影响系数c提高ECN门限值；
    其中，Th表示动态ECN门限值，E表示初始状态下设置的默认ECN门限，a表示incast值对ECN门限影响系数，b表示时延敏感流量比例对ECN门限影响系数，c表示吞吐敏感流量比例对ECN门限影响系数，R S表示时延敏感流量所占比例，R H表示吞吐敏感流量所占比例。
  7. 根据权利要求3至6中任一所述的无损流量拥塞自适应方法，其中，incast值对ECN门限影响系数，时延敏感流量比例对ECN门限影响系数，吞吐敏感流量比例对ECN门限影响系数均根据用户要求自行设置。
  8. 根据权利要求1所述的无损流量拥塞自适应方法，其中，所述根据所述ECN门限值，源端自动调节流量发送窗口，包括；
    根据所述ECN门限值，向源端发送携带ECE标志位的ACK响应报文；
    源端接收所述ACK响应报文，检查所述ECE标志位；
    若所述ECE标志位被标记为拥塞状态，则源端开始自动调节流量发送窗口。
  9. 一种无损流量拥塞自适应系统，包括：
    流量分析模块，被设置成对传送的网络流量进行分析，获取流量分析参数并将所述流量分析参数发送给流量队列管理模块，其中，所述流量分析参数包括流量incast值，时延敏感流量比例，吞吐敏感流量比例；流量队列管理模块，被设置成根据所述流量分析模块获取的所述流量分析参数，动态调整显示拥塞通知ECN的门限值；
    调整流量窗口模块，被设置成根据所述流量队列管理模块动态调整的ECN门限值，自动调节流量发送窗口。
  10. 一种网络设备，包括：
    至少一个处理器；以及，
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行如权利要求1至8中任意一项所述的无损流量拥塞自适应方法。
PCT/CN2021/111568 2020-08-28 2021-08-09 一种无损流量拥塞自适应方法、系统和网络设备 WO2022042282A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010889189.XA CN114205302A (zh) 2020-08-28 2020-08-28 一种无损流量拥塞自适应方法、系统和网络设备
CN202010889189.X 2020-08-28

Publications (1)

Publication Number Publication Date
WO2022042282A1 true WO2022042282A1 (zh) 2022-03-03

Family

ID=80354529

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/111568 WO2022042282A1 (zh) 2020-08-28 2021-08-09 一种无损流量拥塞自适应方法、系统和网络设备

Country Status (2)

Country Link
CN (1) CN114205302A (zh)
WO (1) WO2022042282A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116633770A (zh) * 2023-05-29 2023-08-22 深圳市海成智联科技有限公司 一种适用于局域网设备运行的自动配置运行监管系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115174476B (zh) * 2022-06-30 2023-08-04 苏州浪潮智能科技有限公司 一种ecn控制方法、装置以及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1941737A (zh) * 2005-09-30 2007-04-04 富士通株式会社 预测节点时延的方法和装置以及时延保证的方法和装置
US20110116532A1 (en) * 2009-11-17 2011-05-19 Nokia Corporation Method and apparatus for latency-aware scheduling using interference cancellation
CN103051555A (zh) * 2013-01-05 2013-04-17 北京航空航天大学 基于网络有效带宽和ecn机制的tcp拥塞控制方法
US20190146707A1 (en) * 2006-05-17 2019-05-16 Richard Fetik Secure Application Acceleration System and Apparatus
CN110061927A (zh) * 2019-04-26 2019-07-26 东南大学 一种多队列数据中心环境中面向微突发流的拥塞感知与标记方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1941737A (zh) * 2005-09-30 2007-04-04 富士通株式会社 预测节点时延的方法和装置以及时延保证的方法和装置
US20190146707A1 (en) * 2006-05-17 2019-05-16 Richard Fetik Secure Application Acceleration System and Apparatus
US20110116532A1 (en) * 2009-11-17 2011-05-19 Nokia Corporation Method and apparatus for latency-aware scheduling using interference cancellation
CN103051555A (zh) * 2013-01-05 2013-04-17 北京航空航天大学 基于网络有效带宽和ecn机制的tcp拥塞控制方法
CN110061927A (zh) * 2019-04-26 2019-07-26 东南大学 一种多队列数据中心环境中面向微突发流的拥塞感知与标记方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116633770A (zh) * 2023-05-29 2023-08-22 深圳市海成智联科技有限公司 一种适用于局域网设备运行的自动配置运行监管系统
CN116633770B (zh) * 2023-05-29 2024-02-13 深圳市海成智联科技有限公司 一种适用于局域网设备运行的自动配置运行监管系统

Also Published As

Publication number Publication date
CN114205302A (zh) 2022-03-18

Similar Documents

Publication Publication Date Title
CN108965151B (zh) 一种基于排队时延的显式拥塞控制方法
WO2022042282A1 (zh) 一种无损流量拥塞自适应方法、系统和网络设备
CN109120544B (zh) 一种数据中心网络中基于主机端流量调度的传输控制方法
US20190014053A1 (en) Network Flow Control Method And Network Device
JP5340186B2 (ja) パケット中継装置及びパケットを中継する方法
WO2021244240A1 (zh) 网络拥塞的控制方法、装置、设备、系统及存储介质
KR102350504B1 (ko) 통신 시스템에서 하향링크 전송률 제어를 위한 장치 및 방법
WO2020253488A1 (zh) 拥塞控制方法及装置、通信网络、计算机存储介质
US9444741B2 (en) Facilitating network flows
US8576850B2 (en) Band control apparatus, band control method, and storage medium
Luo et al. Standardization of low-latency TCP with explicit congestion notification: A survey
US20220407808A1 (en) Service Level Adjustment Method, Apparatus, Device, and System, and Storage Medium
KR101344014B1 (ko) 제로 지연 큐잉 방법 및 그 시스템
US20180176136A1 (en) TCP Bufferbloat Resolution
WO2021083160A1 (zh) 数据传输的方法和装置
CN112995048B (zh) 数据中心网络的阻塞控制与调度融合方法及终端设备
US20080291833A1 (en) Method for buffer control for network device
US11695629B2 (en) Method and apparatus for configuring a network parameter
Lu et al. Dynamic ECN marking threshold algorithm for TCP congestion control in data center networks
CN110391992B (zh) 基于交换机主动队列管理的拥塞控制方法和装置
US20070019550A1 (en) Shaper control method, data communication system, network interface apparatus, and network delay apparatus
US20020176443A1 (en) Method and apparatus for bandwidth management of TCP traffic employing post-acknowledgement control
US20100074111A1 (en) Method of data traffic shaping, apparatus and wireless device
Tang et al. TCP-WBQ: a backlog-queue-based congestion control mechanism for heterogeneous wireless networks
CN108243117B (zh) 一种流量监控方法、装置及电子设备

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 04/07/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21860121

Country of ref document: EP

Kind code of ref document: A1