WO2011153860A1 - 处理网络拥塞的方法、装置和核心网络实体 - Google Patents

处理网络拥塞的方法、装置和核心网络实体 Download PDF

Info

Publication number
WO2011153860A1
WO2011153860A1 PCT/CN2011/072056 CN2011072056W WO2011153860A1 WO 2011153860 A1 WO2011153860 A1 WO 2011153860A1 CN 2011072056 W CN2011072056 W CN 2011072056W WO 2011153860 A1 WO2011153860 A1 WO 2011153860A1
Authority
WO
WIPO (PCT)
Prior art keywords
congestion
send
source terminal
data packet
data packets
Prior art date
Application number
PCT/CN2011/072056
Other languages
English (en)
French (fr)
Inventor
宋晓丽
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
Priority to US13/701,398 priority Critical patent/US8904034B2/en
Priority to EP11791852.4A priority patent/EP2582097A4/en
Publication of WO2011153860A1 publication Critical patent/WO2011153860A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/02Protocol performance
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/12Avoiding congestion; Recovering from congestion
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/26Flow control; Congestion control using explicit feedback to the source, e.g. choke packets

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a method, apparatus, and core network entity for handling network congestion. Background technique
  • the congestion state of the network is generally eliminated in the following manner:
  • the core network entity configured with the congestion detection function detects its own egress port and samples the outbound port queue; if the port is detected to be congested, it is created. Congestion message; sending the congestion message to the terminal corresponding to the sample point causing congestion; the terminal reduces the data transmission rate according to the received congestion message.
  • the terminals 1, 2, 3, and 4 send data to the terminal 5 through the core network bodies B1, B2, and B3.
  • the B3 is provided with a congestion detection and congestion information sending function, and periodically performs an outbound port queue. kind. When it is detected that the port is in a congested state, the congestion message is sent to the source terminal corresponding to the data packet of the sampled point.
  • the B3 outbound port queue sequentially caches multiple data packets, and the data packet contains the source address information of the transmitted data packet, which can be seen from FIG.
  • the congestion of the B3 is caused by the terminal 1.
  • the source terminal corresponding to the data packet extracted from the egress port queue is the terminal 2
  • the B3 determines that the terminal 2 is caused by the terminal 2
  • the port is congested, and then a congestion message is sent to the terminal 2 to cause the terminal 2 to reduce the data transmission rate.
  • Terminal 1 that actually causes B3 congestion continues to transmit data at high speed. In this case, the congestion status of B3 can not only be lifted, but also may cause The deterioration of the plug state. Summary of the invention
  • the main object of the present invention is to provide a method, apparatus and core network entity for handling network congestion, which improves the success rate of relieving network congestion.
  • a method of handling network congestion comprising:
  • the process of sending congestion information to a source terminal that sends the most data packets includes:
  • the congestion information is sent to the source terminal that sends the most packets.
  • the method for determining the degree of congestion is:
  • the congestion degree value is greater than or equal to the preset threshold, it indicates that the congestion is large; otherwise, it indicates that the congestion is small.
  • the process of transmitting congestion information to all or one of the source terminals that send the most packets includes:
  • a device for handling network congestion including:
  • the obtaining module is configured to: when detecting that the egress port is in a congested state, extract an adjacent data packet equal to the preset interval value from the egress port queue, and obtain source terminal address information of the adjacent data packet;
  • the searching module is configured to: find, according to source terminal address information, a source terminal that sends the most data packet; and send a module, configured to send congestion information to the source terminal that sends the most data packet.
  • the method is specifically configured to: determine that one or more source terminals send the most data packets; and determine the degree of congestion when the source terminals that send the most data packets are multiple. the size of;
  • the sending module When the sending module sends the congestion information, the sending module is specifically configured to: when the source terminal that sends the most data packets has multiple, send congestion information to all source terminals or one of the source terminals that send the most data packets according to the determination result; and when sending When the source terminal of the largest data packet is one, congestion information is sent to the source terminal that sends the most data packet.
  • the searching module is specifically configured to: determine a congestion degree value and a preset threshold; when the congestion degree value is greater than or equal to a preset threshold, indicating that the congestion degree is large; otherwise, indicating the congestion degree. small.
  • the sending module is specifically configured to: when the congestion degree is small, send congestion information to one of all source terminals that send the most data packets; when the congestion degree is large, send all the source terminals that send the most data packets. Send congestion information.
  • a core network entity includes means for handling network congestion, and the means for processing network congestion includes:
  • the obtaining module is configured to: when detecting that the egress port is in a congested state, extract an adjacent data packet equal to the preset interval value from the egress port queue, and obtain source terminal address information of the adjacent data packet;
  • the apparatus for processing network congestion is the apparatus for processing network congestion according to any one of claims 6 to 8.
  • the method, device and core network entity for processing network congestion provided by the present invention extracts adjacent data packets equal to the preset interval value from the outbound port queue, and obtains source terminal address information of the adjacent data packet, according to Source terminal address information, finds the source terminal that sends the most data packets, and sends congestion information to the source terminal to reduce the data transmission rate to relieve network congestion. Improve the success rate of network congestion.
  • FIG. 1 is a schematic structural diagram of a system for handling network congestion in the prior art
  • FIG. 2 is a schematic diagram of a source terminal corresponding to a data packet in an outbound port queue in the prior art
  • FIG. 3 is a flowchart of an embodiment of a method for processing network congestion according to the present invention
  • FIG. 4 is a flow chart of transmitting congestion information in an embodiment of a method for processing network congestion according to the present invention
  • FIG. 5 is a schematic structural diagram of an apparatus for processing network congestion according to the present invention
  • FIG. 6 is another schematic structural diagram of an apparatus for processing network congestion according to the present invention.
  • FIG. 7 is a schematic structural diagram of a core network entity implementation flow of the present invention. detailed description
  • Step S101 When detecting that an egress port is in a congested state, randomly extracting a phase equal to a preset interval value from an egress port queue Neighboring data packet, obtaining source terminal address information of the adjacent data packet;
  • the random extraction described above is just one of the many ways to extract a packet; in fact, pumping
  • the method of taking packets is not limited to random extraction, but there are many ways, such as sequential extraction, extraction at certain time intervals, etc.
  • Step S102 Search for a source terminal that sends the most data packet according to the source terminal address information.
  • Step S103 Send congestion information to the source terminal that sends the most data packet.
  • the source terminal of the above-mentioned most transmitted data packet is considered to be a terminal that causes the port to be congested.
  • the data packets to be sent in the egress port are sequentially arranged according to the order in which the data packets arrive at the egress port, and are buffered into the egress port queue, where each data packet carries the data packet.
  • Source terminal address information When the egress port processes the congestion state, the data packets to be sent in the egress port are sequentially arranged according to the order in which the data packets arrive at the egress port, and are buffered into the egress port queue, where each data packet carries the data packet.
  • congestion information sent to the source terminal that sends the most data packet is also created according to the detected congestion degree.
  • the congestion information includes a congestion level value and a step value to reduce the data transmission rate.
  • the source terminal receives the congestion information, and reduces the data transmission rate by a corresponding step value, thereby releasing congestion of the out port.
  • the size of the preset interval value may be set according to the throughput of the network and the number of network devices.
  • step S103 in the foregoing embodiment specifically includes:
  • Step S201 Determine that one or more source terminals that send the most data packets are
  • Step S202 When there are multiple source terminals that send the most data packets, determine the size of the congestion level
  • Step S203 Send congestion information to all source terminals or one of the source terminals that send the most data packets according to the determination result;
  • Step S204 When the source terminal that sends the most data packet is one, the congestion information is sent to the source terminal that sends the most data packet.
  • the method for determining congestion in the method for processing network congestion according to the present invention is specifically: determining a congestion degree value and a preset threshold; and when the congestion degree value is greater than or equal to a preset threshold, indicating that the congestion degree is large; When the congestion degree value is less than the preset threshold, it indicates that the congestion level is small.
  • the above preset threshold needs to be set according to the actual network running status.
  • step S204 is specifically: when the congestion degree is small, the congestion information is sent to one of the source terminals that send the most data packets; when the congestion degree is large, the congestion information is sent to the source terminal that sends the most data packets.
  • the congestion information is sent to one of the source terminals that send the most packets, so as to ensure the full utilization of the network bandwidth on the basis of releasing the congestion.
  • one source terminal is selected as the target terminal for transmitting the congestion information from the plurality of source terminals that send the most data packets, and the specific selection manner is selected according to the priority or MAC address of the data packet sent by the source terminal. For example, the congestion information is sent to the source terminal corresponding to the data packet with the lower priority to ensure the transmission rate of the high priority packet; or the importance of the source terminal corresponding to the MAC address is compared, and the congestion information is sent to the source terminal with less importance. To ensure the transmission rate of data packets in the source terminal with a high degree of importance.
  • the source terminal by randomly extracting adjacent data packets equal to the preset interval value from the outbound port queue, and acquiring source terminal address information of the adjacent data packet, searching for the most data packet according to the source terminal address information.
  • the source terminal sends congestion information to the source terminal to reduce the data transmission rate to relieve network congestion. Improve the success rate of lifting network congestion.
  • an apparatus 10 for processing network congestion includes: an acquisition module 11, a lookup module 12, and a transmission module 13.
  • the obtaining module 11 is configured to: when detecting that the egress port is in a congested state, randomly extract, from the egress port queue, an adjacent data packet equal to the preset interval value, and obtain source terminal address information of the adjacent data packet.
  • the searching module 12 is configured to search for a source terminal that sends the most data packet according to the source terminal address information, and send module 13, configured to send congestion information to the source terminal that sends the most data packet.
  • the source terminal of the above-mentioned most transmitted data packet is considered to be a terminal that causes the port to be congested.
  • the data packets to be sent in the egress port are sequentially arranged according to the order in which the data packets arrive at the egress port, and are buffered to the egress port queue, and each data is buffered.
  • the packet carries the source terminal address information for transmitting the data packet.
  • the apparatus 10 for processing network congestion described above further includes a creation module 14.
  • the creating module 14 is configured to create congestion information sent to the source terminal that sends the most data packets according to the detected congestion level when detecting that the egress port is in a congested state.
  • the congestion information includes a congestion level value and a step value for which the data transmission rate needs to be lowered.
  • the source terminal After receiving the congestion information, the source terminal reduces the data transmission rate by a corresponding step value, thereby releasing congestion of the out port.
  • the size of the preset interval value may be set according to the throughput of the network and the number of network devices.
  • the foregoing searching module 12 is further configured to determine that one or more source terminals send the most data packets, and determine the degree of congestion when the source terminals that send the most data packets are multiple.
  • the sending module 13 is configured to: when the number of source terminals that send the most data packets is multiple, determine the degree of congestion, send congestion information to one of the source terminals that send the most data packets according to the determination result; and send the most data when When the source terminal of the packet is one, congestion information is sent to the source terminal that sends the most data packet.
  • the searching module 12 is further configured to determine a congestion level value and a preset threshold value; when the congestion degree value is greater than or equal to a preset threshold, indicating that the congestion degree is large; when the congestion degree value is less than a preset threshold, Indicates that the degree of congestion is small.
  • the above preset threshold needs to be set according to the actual network running status.
  • the sending module 13 is further configured to: when the congestion degree is small, send the congestion information to one of the source terminals that send the most data packets; when the congestion degree is large, send the congestion information to the source terminal that sends the most data packets.
  • the foregoing sending module 13 sends congestion information to one of the source terminals that send the most data packets when the degree of congestion is small, in order to ensure the full utilization of the network bandwidth on the basis of releasing the congestion.
  • the above searching module 12 selects one source terminal from a plurality of source terminals that send the most data packets.
  • the specific selection method is to select according to the priority or MAC address of the data packet sent by the source terminal. For example, the congestion information is sent to the source terminal corresponding to the data packet with the lower priority to ensure the transmission rate of the high priority packet; or the importance of the source terminal corresponding to the MAC address is compared, and the congestion information is sent to the source terminal with less importance. To ensure the transmission rate of data packets in the source terminal with a high degree of importance.
  • the source terminal by randomly extracting adjacent data packets equal to the preset interval value from the outbound port queue, and acquiring source terminal address information of the adjacent data packet, searching for the most data packet according to the source terminal address information.
  • the source terminal sends congestion information to the source terminal to reduce the data transmission rate to relieve network congestion. Improve the success rate of lifting network congestion.
  • a core network entity of the present invention which includes a device for processing network congestion.
  • the device for processing network congestion is the device 10 for processing network congestion as described above, and details are not described herein again.
  • the above core network entity is a core switch.

Abstract

本发明公开了一种处理网络拥塞的方法、装置和核心网络实体,均可在检测到出端口处于拥塞状态时,从出端口队列中抽取与预设区间值相等个数的相邻数据包,并获取相邻数据包的源终端地址信息;根据源终端地址信息,查找发送最多数据包的源终端;向发送最多数据包的源终端发送拥塞信息。本发明通过从出端口队列中抽取与预设区间值相等个数的相邻数据包,并获取相邻数据包的源终端地址信息,根据源终端地址信息,查找发送最多数据包的源终端,并向源终端发送拥塞信息,使之降低数据传输速率,以解除网络拥塞。提高解除网络拥塞的成功率。

Description

处理网络拥塞的方法、 装置和核心网络实体 技术领域
本发明涉及通信技术领域, 尤其是涉及一种处理网络拥塞的方法、 装 置和核心网络实体。 背景技术
随着互联网技术的高速发展, 网络中交互的业务量越来越大, 用户对 网络性能的要求也越来越高。 同时在业务数据交互过程中, 网络的拥塞、 延时、 丟包等问题也日趋严重。
现有技术中一般通过以下方式消除网络的拥塞状态: 配置有拥塞检测 功能的核心网络实体对其自身出端口进行检测, 以及对出端口队列进行釆 样; 若检测出端口为拥塞状态, 则创建拥塞消息; 将该拥塞消息发送给导 致拥塞的釆样点对应的终端; 该终端根据接收到的拥塞消息降低自身数据 传送速率。 参见图 1 , 终端 1、 2、 3、 4通过核心网络体 Bl、 B2和 B3向终 端 5发送数据; 其中, B3设有拥塞检测和拥塞信息发送功能, 其周期性的 对出端口队列进行釆样。 当检测到出端口处于拥塞状态, 则将拥塞消息发 往被釆样点的数据包对应的源终端。
以上所述的消除网络拥塞状态的方式常常出现判断错误, 参见图 2 , B3出端口队列中顺序緩存多个数据包, 数据包中包含有发送数据包的源地 址信息, 从图 2中可以看出 B3的拥塞是由终端 1导致的, 然而 B3检测到 出端口处于拥塞状态时, 从出端口队列抽到釆样的数据包对应的源终端为 终端 2, 此时 B3确定是终端 2导致出端口拥塞, 然后向终端 2发送拥塞消 息使终端 2降低数据传输速率。而真正导致 B3拥塞的终端 1依然持续高速 的数据传送。 在这种情况下 B3的拥塞状态不但不能解除, 还可能会造成拥 塞状态的恶化。 发明内容
本发明的主要目的在于提供一种处理网络拥塞的方法、 装置和核心网 络实体, 提高解除网络拥塞的成功率。
为了达到上述目的, 本发明的技术方案是这样实现的:
一种处理网络拥塞的方法, 该方法包括:
在检测到出端口处于拥塞状态时, 从出端口队列中抽取与预设区间值 相等个数的相邻数据包, 并获取相邻数据包的源终端地址信息;
根据所获取的源终端地址信息, 查找发送最多数据包的源终端; 向发送最多数据包的所述源终端发送拥塞信息。
所述向发送最多数据包的源终端发送拥塞信息的过程包括:
判断发送最多数据包的源终端为一个或多个;
当发送最多数据包的源终端为多个时, 判断拥塞程度的大小; 根据判断结果向发送最多数据包的所有源终端或其中之一发送拥塞信 息;
当发送最多数据包的源终端为一个时, 向发送最多数据包的源终端发 送拥塞信息。
判断拥塞程度大小的方法为:
判断拥塞程度值与预设阔值的大小;
当拥塞程度值大于或等于预设阔值时, 表明拥塞程度大; 否则, 表明 拥塞程度小。
根据判断结果, 向发送最多数据包的所有源终端或其中之一发送拥塞 信息的过程包括:
当拥塞程度小时, 向发送最多数据包的源终端其中之一发送拥塞信息; 当拥塞程度大时, 向发送最多数据包的所有源终端发送拥塞信息。 一种处理网络拥塞的装置, 包括:
获取模块, 用于在检测到出端口处于拥塞状态时, 从出端口队列中抽 取与预设区间值相等个数的相邻数据包, 并获取相邻数据包的源终端地址 信息;
查找模块, 用于根据源终端地址信息, 查找发送最多数据包的源终端; 发送模块, 用于向发送最多数据包的源终端发送拥塞信息。
所述查找模块在查找发送最多数据包的源终端时, 具体用于: 判断发 送最多数据包的源终端为一个或多个; 以及当发送最多数据包的源终端为 多个时, 判断拥塞程度的大小;
所述发送模块在发送拥塞信息时, 具体用于: 在发送最多数据包的源 终端为多个时, 根据判断结果向发送最多数据包的所有源终端或其中之一 发送拥塞信息; 以及当发送最多数据包的源终端为一个时, 向该发送最多 数据包的源终端发送拥塞信息。
所述查找模块在判断拥塞程度大小时, 具体用于: 判断拥塞程度值与 预设阔值的大小; 当拥塞程度值大于或等于预设阔值时, 表明拥塞程度大; 否则, 表明拥塞程度小。
所述发送模块在发送拥塞信息时, 具体用于: 当拥塞程度小时, 向发 送最多数据包的所有源终端其中之一发送拥塞信息; 当拥塞程度大时, 向 发送最多数据包的所有源终端发送拥塞信息。
一种核心网络实体, 包括处理网络拥塞的装置, 所述处理网络拥塞的 装置包括:
获取模块, 用于在检测到出端口处于拥塞状态时, 从出端口队列中抽 取与预设区间值相等个数的相邻数据包, 并获取相邻数据包的源终端地址 信息;
查找模块, 用于根据源终端地址信息, 查找发送最多数据包的源终端; 发送模块, 用于向发送最多数据包的源终端发送拥塞信息。 所述处理网络拥塞的装置为权利要求 6至 8任一项所述的处理网络拥 塞的装置。
本发明提供的处理网络拥塞的方法、 装置和核心网络实体, 通过从出 端口队列中抽取与预设区间值相等个数的相邻数据包, 并获取相邻数据包 的源终端地址信息, 根据源终端地址信息, 查找发送最多数据包的源终端, 并向源终端发送拥塞信息, 使之降低数据传输速率, 以解除网络拥塞。 提 高解除网络拥塞的成功率。 附图说明
图 1是现有技术中处理网络拥塞的系统的结构示意图;
图 2是现有技术中对出端口队列中数据包对应的源终端釆样的示意图; 图 3 是本发明的处理网络拥塞的方法一实施例的流程图;
图 4是本发明的处理网络拥塞的方法实施例中发送拥塞信息的流程图; 图 5 是本发明的处理网络拥塞的装置实施例的结构示意图;
图 6是本发明的处理网络拥塞的装置实施例的另一结构示意图; 图 7是本发明的核心网络实体实施流的结构示意图。 具体实施方式
应当理解, 此处所描述的具体实施例仅仅用以解释本发明, 并不用于 限定本发明。
参见图 3 , 提出本发明的处理网络拥塞的方法一实施例, 其包括: 步骤 S101、 在检测到出端口处于拥塞状态时, 从出端口队列中随机抽 取与预设区间值相等个数的相邻数据包, 获取相邻数据包的源终端地址信 息;
当然, 上述的随机抽取只是抽取数据包的众多方式之一; 实际上, 抽 取数据包的方式并不限定于随机抽取, 而是还可以有很多方式, 如顺序抽 取、 按一定时间间隔抽取等;
步骤 S102、 根据源终端地址信息, 查找发送最多数据包的源终端; 步骤 S103、 向发送最多数据包的源终端发送拥塞信息。
上述发送最多的数据包的源终端被认为导致出端口拥塞的终端。
上述出端口在处理拥塞状态时, 出端口中待发送的数据包, 根据数据 包到达该出端口的先后顺序依次排列, 并緩存至出端口队列中, 每个数据 包携带有发送该数据包的源终端地址信息。
上述实例中, 在检测到出端口处于拥塞状态时, 还根据检测的拥塞程 度创建向发送最多数据包的源终端发送的拥塞信息。 该拥塞信息包括拥塞 程度值和需降低数据传送率的步长值。 源终端接收该拥塞信息, 将数据传 送速率降低相应的步长值, 从而解除出端口的拥塞。
所述预设区间值的大小可以根据网络的吞吐量和网络设备数量的多少 进行设定。
进一步地, 参见图 4, 上述实施例中步骤 S103具体包括:
步骤 S201、 判断发送最多数据包的源终端为一个或多个;
步骤 S202、 当发送最多数据包的源终端为多个时, 判断拥塞程度的大 小;
步骤 S203、 根据判断结果向发送最多数据包的所有源终端或其中之一 发送拥塞信息;
步骤 S204、 当发送最多数据包的源终端为一个时, 向发送最多数据包 的源终端发送拥塞信息。
进一步地, 本发明的处理网络拥塞的方法中判断拥塞程度的大小具体 为: 判断拥塞程度值与预设阔值的大小; 当拥塞程度值大于或等于预设阔 值时, 表明拥塞程度大; 当拥塞程度值小于预设阈值时, 表明拥塞程度小。 上述预设阔值需根据实际网络运行状况进行设置。
进一步地, 上述步骤 S204具体为: 当拥塞程度小时, 向发送最多数据 包的源终端其中之一发送拥塞信息; 当拥塞程度大时, 向发送最多数据包 的源终端发送拥塞信息。
上述当拥塞程度小时, 向发送最多数据包的源终端其中之一发送拥塞 信息, 是为了在解除拥塞的基础上, 保证网络宽带的充分利用。
其中, 从多个发送最多数据包的源终端中选择一个源终端作为发送拥 塞信息的目标终端, 其具体的选择方式是根据源终端发送的数据包的优先 级或 MAC地址进行选择。例如, 向优先级低的数据包对应的源终端发送拥 塞信息, 以保证优先级高数据包的传送速率; 或比较 MAC地址对应的源终 端的重要程度, 向重要程度小的源终端发送拥塞信息, 以保证重要程度大 的源终端中数据包的发送速率。
本实施例, 通过从出端口队列中随机抽取与预设区间值相等个数的相 邻数据包, 并获取相邻数据包的源终端地址信息, 根据源终端地址信息, 查找发送最多数据包的源终端, 并向源终端发送拥塞信息, 使之降低数据 传输速率, 以解除网络拥塞。 提高解除网络拥塞的成功率。
参见图 5, 提出一种处理网络拥塞的装置 10—实施例, 其包括: 获取 模块 11、 查找模块 12和发送模块 13。 其中, 获取模块 11 , 用于在检测到 出端口处于拥塞状态时, 从出端口队列中随机抽取与预设区间值相等个数 的相邻数据包, 并获取相邻数据包的源终端地址信息; 查找模块 12, 用于 根据源终端地址信息, 查找发送最多数据包的源终端; 发送模块 13, 用于 向发送最多数据包的源终端发送拥塞信息。
上述发送最多的数据包的源终端被认为导致出端口拥塞的终端。
上述出端口在处理拥塞状态时, 出端口中待发送的数据包, 根据数据 包到达该出端口的先后顺序依次排列, 并緩存至出端口队列中, 每个数据 包携带有发送该数据包的源终端地址信息。
参见图 6, 上述处理网络拥塞的装置 10还包括创建模块 14。 其中, 创 建模块 14, 用于在检测到出端口处于拥塞状态时, 根据检测的拥塞程度创 建向发送最多数据包的源终端发送的拥塞信息。 该拥塞信息包括拥塞程度 值和需降低数据传送率的步长值。 源终端接收该拥塞信息后, 将数据传送 速率降低相应的步长值, 从而解除出端口的拥塞。
所述预设区间值的大小可以根据网络的吞吐量和网络设备数量的多少 进行设定。
进一步地, 上述查找模块 12, 还用于判断发送最多数据包的源终端为 一个或多个; 以及当发送最多数据包的源终端为多个时, 判断拥塞程度的 大小。 上述发送模块 13 , 用于在发送最多数据包的源终端为多个, 判断拥 塞程度的大小后, 根据判断结果向发送最多数据包的源终端或其中之一发 送拥塞信息; 以及当发送最多数据包的源终端为一个时, 向该发送最多数 据包的源终端发送拥塞信息。
进一步地, 查找模块 12, 还用于判断拥塞程度值与预设阔值的大小; 当拥塞程度值大于或等于预设阈值时, 表明拥塞程度大; 当拥塞程度值小 于预设阔值时, 表明拥塞程度小。
上述预设阔值需根据实际网络运行状况进行设置。
进一步地, 发送模块 13 , 还用于当拥塞程度小时, 向发送最多数据包 的源终端其中之一发送拥塞信息; 当拥塞程度大时, 向发送最多数据包的 源终端发送拥塞信息。
上述发送模块 13 , 当拥塞程度小时, 向发送最多数据包的源终端其中 之一发送拥塞信息, 是为了在解除拥塞的基础上, 保证网络宽带的充分利 用。
上述查找模块 12, 从多个发送最多数据包的源终端中选择一个源终端 作为发送拥塞信息的目标终端, 其具体的选择方式是根据源终端发送的数 据包的优先级或 MAC地址进行选择。 例如, 向优先级低的数据包对应的源 终端发送拥塞信息, 以保证优先级高数据包的传送速率; 或比较 MAC地址 对应的源终端的重要程度, 向重要程度小的源终端发送拥塞信息, 以保证 重要程度大的源终端中数据包的发送速率。
本实施例, 通过从出端口队列中随机抽取与预设区间值相等个数的相 邻数据包, 并获取相邻数据包的源终端地址信息, 根据源终端地址信息, 查找发送最多数据包的源终端, 并向源终端发送拥塞信息, 使之降低数据 传输速率, 以解除网络拥塞。 提高解除网络拥塞的成功率。
参见图 7, 提出本发明的核心网络实体, 其包括处理网络拥塞的装置, 其中, 该处理网络拥塞的装置为以上所述的处理网络拥塞的装置 10, 在此 不再赘述。 上述核心网络实体为核心交换机。
应当理解的是, 以上仅为本发明的优选实施例, 不能因此限制本发明 的专利范围, 凡是利用本发明说明书及附图内容所作的等效结构或等效流 程变换, 或直接或间接运用在其他相关的技术领域, 均同理包括在本发明 的专利保护范围内。

Claims

权利要求书
1、 一种处理网络拥塞的方法, 其特征在于, 该方法包括:
在检测到出端口处于拥塞状态时, 从出端口队列中抽取与预设区间值 相等个数的相邻数据包, 并获取相邻数据包的源终端地址信息;
根据所获取的源终端地址信息, 查找发送最多数据包的源终端; 向发送最多数据包的所述源终端发送拥塞信息。
2、 根据权利要求 1所述的方法, 其特征在于, 所述向发送最多数据包 的源终端发送拥塞信息的过程包括:
判断发送最多数据包的源终端为一个或多个;
当发送最多数据包的源终端为多个时, 判断拥塞程度的大小; 根据判断结果向发送最多数据包的所有源终端或其中之一发送拥塞信 息;
当发送最多数据包的源终端为一个时, 向发送最多数据包的源终端发 送拥塞信息。
3、 根据权利要求 2所述的方法, 其特征在于, 判断拥塞程度大小的方 法为:
判断拥塞程度值与预设阔值的大小;
当拥塞程度值大于或等于预设阔值时, 表明拥塞程度大; 否则, 表明 拥塞程度小。
4、 根据权利要求 2或 3所述的方法, 其特征在于, 根据判断结果, 向 发送最多数据包的所有源终端或其中之一发送拥塞信息的过程包括:
当拥塞程度小时, 向发送最多数据包的源终端其中之一发送拥塞信息; 当拥塞程度大时, 向发送最多数据包的所有源终端发送拥塞信息。
5、 一种处理网络拥塞的装置, 其特征在于, 包括:
获取模块, 用于在检测到出端口处于拥塞状态时, 从出端口队列中抽 取与预设区间值相等个数的相邻数据包, 并获取相邻数据包的源终端地址 信息;
查找模块, 用于根据源终端地址信息, 查找发送最多数据包的源终端; 发送模块, 用于向发送最多数据包的源终端发送拥塞信息。
6、 根据权利要求 5所述的装置, 其特征在于,
所述查找模块在查找发送最多数据包的源终端时, 具体用于: 判断发 送最多数据包的源终端为一个或多个; 以及当发送最多数据包的源终端为 多个时, 判断拥塞程度的大小;
所述发送模块在发送拥塞信息时, 具体用于: 在发送最多数据包的源 终端为多个时, 根据判断结果向发送最多数据包的所有源终端或其中之一 发送拥塞信息; 以及当发送最多数据包的源终端为一个时, 向该发送最多 数据包的源终端发送拥塞信息。
7、 根据权利要求 6所述的装置, 其特征在于,
所述查找模块在判断拥塞程度大小时, 具体用于: 判断拥塞程度值与 预设阔值的大小; 当拥塞程度值大于或等于预设阔值时, 表明拥塞程度大; 否则, 表明拥塞程度小。
8、 根据权利要求 6或 7所述的装置, 其特征在于,
所述发送模块在发送拥塞信息时, 具体用于: 当拥塞程度小时, 向发 送最多数据包的所有源终端其中之一发送拥塞信息; 当拥塞程度大时, 向 发送最多数据包的所有源终端发送拥塞信息。
9、 一种核心网络实体, 包括处理网络拥塞的装置, 其特征在于, 所述 处理网络拥塞的装置包括:
获取模块, 用于在检测到出端口处于拥塞状态时, 从出端口队列中抽 取与预设区间值相等个数的相邻数据包, 并获取相邻数据包的源终端地址 信息; 查找模块, 用于根据源终端地址信息, 查找发送最多数据包的源终端; 发送模块, 用于向发送最多数据包的源终端发送拥塞信息。
10、 根据权利要求 9所述的核心网络实体, 其特征在于, 所述处理网 络拥塞的装置为权利要求 6至 8任一项所述的处理网络拥塞的装置。
PCT/CN2011/072056 2010-06-09 2011-03-22 处理网络拥塞的方法、装置和核心网络实体 WO2011153860A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US13/701,398 US8904034B2 (en) 2010-06-09 2011-03-22 Method and apparatus for processing network congestion and core network entity
EP11791852.4A EP2582097A4 (en) 2010-06-09 2011-03-22 Method, apparatus and core network entity for processing network congestion

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010196092.7 2010-06-09
CN201010196092.7A CN102281183B (zh) 2010-06-09 2010-06-09 处理网络拥塞的方法、装置和核心网络实体

Publications (1)

Publication Number Publication Date
WO2011153860A1 true WO2011153860A1 (zh) 2011-12-15

Family

ID=45097511

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/072056 WO2011153860A1 (zh) 2010-06-09 2011-03-22 处理网络拥塞的方法、装置和核心网络实体

Country Status (4)

Country Link
US (1) US8904034B2 (zh)
EP (1) EP2582097A4 (zh)
CN (1) CN102281183B (zh)
WO (1) WO2011153860A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104270324A (zh) * 2014-10-16 2015-01-07 中国电子科技集团公司第五十四研究所 一种适用于受限卫星网络的拥塞控制方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FI20135989L (fi) * 2013-10-03 2015-04-04 Tellabs Oy Kytkinlaite tiedonsiirtoverkon verkkoelementtiä varten
CN109417720B (zh) * 2016-07-22 2021-03-30 华为技术有限公司 拥塞控制方法、基站及终端
CN108064058B (zh) * 2016-11-07 2022-11-01 中兴通讯股份有限公司 拥塞控制方法及装置、基站
CN113194501A (zh) * 2021-04-29 2021-07-30 中南民族大学 基于ZigBee的医疗监控系统及网络拥塞控制方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070268830A1 (en) * 2006-05-19 2007-11-22 Cisco Technology, Inc. Flow based flow control in an ethernet switch backplane
CN101227495A (zh) * 2008-02-20 2008-07-23 中兴通讯股份有限公司 公用电信分组数据网系统及其拥塞控制方法
CN101296187A (zh) * 2008-06-13 2008-10-29 北京市翌晨通信技术研究所 一种通信网络拥塞控制的方法
CN101582852A (zh) * 2009-06-10 2009-11-18 中兴通讯股份有限公司 一种网络拥塞管理的方法及系统

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7180857B2 (en) * 2000-11-24 2007-02-20 Matsushita Electric Industrial Co., Ltd Apparatus and method for flow control
JP2002252640A (ja) * 2001-02-23 2002-09-06 Fujitsu Ltd ネットワーク中継装置及び方法並びにシステム
US7239636B2 (en) * 2001-07-23 2007-07-03 Broadcom Corporation Multiple virtual channels for use in network devices
JP2007528636A (ja) * 2003-10-30 2007-10-11 チーク テクノロジーズ,インク. ノンブロッキングで決定論的マルチレートマルチキャストパケットスケジューリング
US7613184B2 (en) * 2003-11-07 2009-11-03 Alcatel Lucent Method and apparatus for performing scalable selective backpressure in packet-switched networks using internal tags
US20050129008A1 (en) * 2003-12-16 2005-06-16 Intel Corporation Congestion management apparatus, systems, and methods
US20060056308A1 (en) * 2004-05-28 2006-03-16 International Business Machines Corporation Method of switching fabric for counteracting a saturation tree occurring in a network with nodes
US7830801B2 (en) * 2004-10-29 2010-11-09 Broadcom Corporation Intelligent fabric congestion detection apparatus and method
US7961621B2 (en) * 2005-10-11 2011-06-14 Cisco Technology, Inc. Methods and devices for backward congestion notification
CN101383694A (zh) * 2007-09-03 2009-03-11 电子科技大学 基于数据挖掘技术的拒绝服务攻击防御方法和系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070268830A1 (en) * 2006-05-19 2007-11-22 Cisco Technology, Inc. Flow based flow control in an ethernet switch backplane
CN101227495A (zh) * 2008-02-20 2008-07-23 中兴通讯股份有限公司 公用电信分组数据网系统及其拥塞控制方法
CN101296187A (zh) * 2008-06-13 2008-10-29 北京市翌晨通信技术研究所 一种通信网络拥塞控制的方法
CN101582852A (zh) * 2009-06-10 2009-11-18 中兴通讯股份有限公司 一种网络拥塞管理的方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2582097A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104270324A (zh) * 2014-10-16 2015-01-07 中国电子科技集团公司第五十四研究所 一种适用于受限卫星网络的拥塞控制方法

Also Published As

Publication number Publication date
CN102281183B (zh) 2015-08-26
US20130111061A1 (en) 2013-05-02
EP2582097A4 (en) 2017-09-13
US8904034B2 (en) 2014-12-02
CN102281183A (zh) 2011-12-14
EP2582097A1 (en) 2013-04-17

Similar Documents

Publication Publication Date Title
CN109617931B (zh) 一种SDN控制器的DDoS攻击防御方法及防御系统
EP2434689B1 (en) Method and apparatus for detecting message
US8732833B2 (en) Two-stage intrusion detection system for high-speed packet processing using network processor and method thereof
CN107770132A (zh) 一种对算法生成域名进行检测的方法及装置
WO2015007099A1 (zh) 数据包传输方法及装置
WO2011153860A1 (zh) 处理网络拥塞的方法、装置和核心网络实体
EP2073457A1 (en) A method and apparatus for preventing igmp message attack
US10924457B2 (en) Packet cleaning method and apparatus
CN105743815B (zh) 抖动缓冲数据处理的方法和装置
WO2019072072A1 (zh) 一种拥塞流识别方法及网络设备
RU2013106261A (ru) Способ и устройство для распределения виртуальной локальной сети
CN101699795B (zh) 一种报文拥塞处理方法及系统
JP6502902B2 (ja) 攻撃検知装置、攻撃検知システムおよび攻撃検知方法
CN103281257A (zh) 一种协议报文处理方法和设备
CN112910793A (zh) 用于七层负载均衡中连接复用的方法和负载均衡器
US20110149776A1 (en) Network interface card device and method of processing traffic using the network interface card device
CN104753813B (zh) Dma传送报文的方法
CN113242260A (zh) 攻击检测方法、装置、电子设备及存储介质
JP6376609B2 (ja) アクセス制御装置および認証制御方法
CN108024291B (zh) 一种移动网络中共享上网检测的方法及装置
JP2009111562A (ja) 通信量制御システム、通信量制御システム用サーバ、および通信量制御システム用クライアント
CN103248605A (zh) 一种基于ipv6的tcp流汇聚方法及系统
CN107251593B (zh) 数据包处理方法和相关设备
CN112612670B (zh) 一种会话信息统计方法、装置、交换设备及存储介质
CN112019589B (zh) 一种多层级负载均衡数据包处理方法

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: 11791852

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 13701398

Country of ref document: US

WWE Wipo information: entry into national phase

Ref document number: 2011791852

Country of ref document: EP