WO2011009304A1 - Method and device for measuring available bandwidth using mixed messages - Google Patents

Method and device for measuring available bandwidth using mixed messages Download PDF

Info

Publication number
WO2011009304A1
WO2011009304A1 PCT/CN2010/071120 CN2010071120W WO2011009304A1 WO 2011009304 A1 WO2011009304 A1 WO 2011009304A1 CN 2010071120 W CN2010071120 W CN 2010071120W WO 2011009304 A1 WO2011009304 A1 WO 2011009304A1
Authority
WO
WIPO (PCT)
Prior art keywords
packet
icmp
mixed
measurement end
message
Prior art date
Application number
PCT/CN2010/071120
Other languages
French (fr)
Chinese (zh)
Inventor
李�瑞
李晋
谭然
乾春燕
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 WO2011009304A1 publication Critical patent/WO2011009304A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0876Network utilisation, e.g. volume of load or congestion level
    • H04L43/0882Utilisation of link capacity
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0823Errors, e.g. transmission errors
    • H04L43/0829Packet loss

Definitions

  • the present invention relates to the field of computer network electronic communications, and more particularly to a method and apparatus for measuring available bandwidth using hybrid messages. Background technique
  • network bandwidth measurement can be divided into different aspects such as link bandwidth, path bandwidth and available bandwidth according to different measures. Available bandwidth is a key parameter in network routing, traffic engineering, and QoS control in engineering, and is more practical than bottleneck bandwidth.
  • the link bandwidth refers to the maximum bandwidth of a link, that is, the maximum data transmission rate of the link, which is determined by the physical properties of the link.
  • the available bandwidth of the link refers to the bandwidth that is currently unused by a link.
  • the path bandwidth is the current path.
  • the minimum bandwidth of all links on the path, the path bandwidth is also called the end-to-end path bandwidth;
  • the available bandwidth of the path is the minimum of the available bandwidth of all links on the current path, and the available bandwidth of the path is also called the end-to-end path. Available bandwidth.
  • the link with the smallest available bandwidth of the link is called the urgent link of the path, and the link with the smallest link bandwidth.
  • a bottleneck link called the path.
  • Figure 1 shows the end-to-end bandwidth diagram.
  • the path between the sender and the receiver consists of three links 2 and 3 , where c 2 and c 3 are the link bandwidths of links A, 2 , and 3 , respectively.
  • the shaded portion indicates the bandwidth that the current link has used, and then 4, 4, and 4 are the unused bandwidth of link A, 2 , and 3, that is, the available bandwidth of the link.
  • ⁇ c 2 ⁇ c 3 that is, the link bandwidth of the link i is the smallest, and the path bandwidth of the sender to the receiver is determined, and the link A is the bottleneck link between the sender and the receiver;
  • 4 ⁇ 4 ⁇ 4 that is, the link available bandwidth of link 3 is the smallest, which determines the available bandwidth of the path from the sender to the receiver, and link 3 is the urgent link of the path.
  • the existing bandwidth measurement methods are mostly time-based packet queue algorithms.
  • the main measurement models include PGM (Packet Gap Model) and Detection Rate Model (PRM, Packet Rate Model).
  • the algorithm based on the PGM model mainly analyzes the packet interval variation of the probe packet queue, and determines the bandwidth estimation value according to the change trend, which is also called a timestamp-based bandwidth measurement method; the PRM model-based algorithm is mainly at the transmitting end.
  • the probe packet queue is sent at a certain rate, and the transmission rate of the probe packet queue is dynamically adjusted according to the rate at which the receiver receives the probe packet queue, thereby determining a matching critical rate to estimate the bandwidth to be tested.
  • the above two measurement methods generally require the cooperation of the passive measurement end. That is to say, the passive measurement end needs to perform corresponding calculation according to some information of the probe packet queue, and returns the calculation information to the active measurement end. Work with the link to test the link bandwidth. In this case, the passive measuring end is required to take auxiliary measures such as installing a corresponding measuring program, which reduces the practicability and flexibility of the measuring method.
  • the present invention provides a method for measuring available bandwidth by using a hybrid packet, the method comprising: the active measurement end transmitting an Internet Control Message Protocol (ICMP) message to the passive measurement end, and collecting the normal ICMP message of the network on the passive measurement end Packet loss rate;
  • ICMP Internet Control Message Protocol
  • the active measurement end sends the mixed packet to the passive measurement end, and collects the average packet loss rate of the ICMP packet of the hybrid packet at the passive measurement end.
  • the hybrid packet includes the ICMP packet and the user data.
  • Packet protocol (UDP) packet when the rate difference is greater than the preset packet loss rate threshold, the average rate at which the passive measurement end receives the mixed packet is counted as the available bandwidth from the active measurement end to the passive measurement end.
  • the method further includes: the active measurement end sends the ICMP message to the passive measurement end at the preset initial transmission rate, and the passive measurement end receives the ICMP message. Returning the ICMP message to the active measurement end;
  • the normal ICMP packet loss rate of the network is counted, and the following is: the active measurement end collects the normal ICMP packet of the network according to the ICMP packet sent by the active measurement end and the ICMP packet sent by the passive measurement end. Packet loss rate.
  • the method further includes: determining, by the active measurement end, a transmission rate of the hybrid message;
  • determining a sending rate of the mixed packet is specifically:
  • the first transmission rate of the hybrid message is the preset initial transmission rate, or is the primary The link bandwidth of the bottleneck link between the measurement end and the passive measurement end.
  • the mixed message sent by the active measuring end is n sets of mixed messages, where n is an integer greater than or equal to 1.
  • the method further includes: the active measurement end setting the destination port of the UDP packet as an incorrect destination port;
  • the method further includes: the passive measurement end discards the UDP packet according to the wrong destination port, and the hybrid message is The ICMP message in is returned to the active measurement terminal.
  • the active measurement end calculates the average packet loss rate of the ICMP ⁇ , , specifically:
  • the active measurement end collects the packet loss rate of the ICMP packet of each group of the mixed packets in the n groups according to the n sets of the mixed packets and the ICMP packets sent by the passive measurement end, and compares the packet loss rate of the ICMP packets in the n groups. The average packet loss rate of the ICMP packets is averaged, and the average packet loss rate of the ICMP packets is obtained.
  • the method further includes: the active measurement end, when the active measurement end determines that the difference between the average ICMP packet loss rate and the normal ICMP packet loss rate of the network is less than or equal to a preset packet loss rate threshold, the method further includes: And determining, according to the first sending rate of the mixed packet, the sending rate is increased by linearly increasing or exponentially increasing.
  • the active measurement end collects the average rate at which the passive measurement end receives the mixed message, and specifically: the active measurement end is based on the n sets of mixed packets sent by the active measurement end, and the ICMP message returned by the passive measurement end. And the rate at which the passive measurement end receives the mixed message of each group in the n groups, and averages the rate at which the passive measurement terminal receives the n groups of mixed packets, to obtain the passive measurement end receiving the hybrid The average rate of messages.
  • the present invention also provides an apparatus for measuring available bandwidth by using a mixed message, the apparatus comprising: an active measuring end and a passive measuring end, wherein The active measurement end is configured to send an ICMP message to the passive measurement end, and collect a packet loss rate of the normal ICMP packet on the passive measurement end; and send the mixed message to the passive measurement end. And counting, when the difference between the flat packet loss rate of the mixed packet of the passive measurement end is greater than a preset packet loss rate threshold, counting the average rate at which the passive measurement end receives the mixed packet, as The available bandwidth from the active measurement end to the passive measurement end;
  • the passive measurement end is configured to receive the ICMP packet and the hybrid packet.
  • the passive measurement end is further configured to: when receiving the ICMP message sent by the active measurement end, return the ICMP message to the active measurement end; and receive the hybrid report sent by the active measurement end The UDP packet in the mixed packet is discarded, and the ICMP packet in the mixed packet is returned to the active measurement end.
  • the active measurement terminal sends an ICMP ⁇ message to the passive measurement end, and the normal ICMP/ ⁇ packet loss rate of the network on the passive measurement end is counted; the active measurement end sends a mixed message to the passive measurement end, The average packet loss rate of the ICMP packets on the passive measurement end is counted.
  • the active measurement end determines that the difference between the average packet loss rate of the ICMP packet and the packet loss rate of the normal ICMP packet is greater than the preset packet loss rate threshold, the statistical passive measurement is performed.
  • the average rate at which the terminal receives the mixed message; the average rate is the measured path available bandwidth.
  • the destination port of the UDP packet sent by the active measurement end is set to be incorrect.
  • the passive measurement end can directly discard the UDP packet and return only the ICMP packet to the active measurement end. . It can be seen that the bandwidth measurement in the present invention does not require the cooperation of the passive measurement end, which reduces the complexity of the measurement work, and the measurement method of the present invention is also more practical and flexible.
  • Figure 1 is a schematic diagram of end-to-end bandwidth
  • FIG. 2 is a schematic flowchart of a method for measuring available bandwidth by using a mixed packet according to the present invention
  • FIG. 3 is a schematic flowchart of a method for measuring available bandwidth by using a mixed packet according to the present invention
  • Figure 4 is a schematic view showing the structure of the mixed type of the present invention
  • FIG. 5 is a schematic structural diagram of an apparatus for measuring available bandwidth by using a mixed message according to the present invention. detailed description
  • the bandwidth measurement in the present invention is mainly for measuring the available bandwidth of the end-to-end path, where one end is the active measurement end, and the other end is the passive measurement end; the active measurement end sends the mixed message to the passive measurement end,
  • the path of the end can be measured with bandwidth.
  • the mixed message refers to the User Datagram Protocol (UDP) and the Internet Control Message Protocol (ICMP).
  • UDP User Datagram Protocol
  • ICMP Internet Control Message Protocol
  • Step 201 The active measurement end sends an ICMP message to the passive measurement end, and collects the normal ICMP packet loss rate of the passive measurement end. ;
  • Step 202 The active measurement end sends the mixed packet to the passive measurement end, and collects the average packet loss rate of the ICMP packet of the mixed packet at the passive measurement end.
  • the mixed packet includes the ICMP packet and
  • the UDP packet is used.
  • the packet loss rate is greater than the preset packet loss threshold, the average rate at which the passive measurement end receives the hybrid packet is counted as the available bandwidth from the active measurement end to the passive measurement end.
  • the method for measuring bandwidth of the present invention is described below in conjunction with a specific embodiment. As shown in FIG. 3, the method includes:
  • Step 301 The active measurement end measures a normal ICMP ⁇ packet loss rate of the network.
  • the active measurement end sends the ICMP message to the passive measurement end at the preset initial transmission rate. After receiving the ICMP message, the passive measurement end returns the ICMP message to the active measurement end.
  • the active measurement end is based on the ICMP message sent by itself.
  • the ICMP packet returned by the passive measurement terminal is used to collect the normal ICMP packet loss rate from the active measurement end to the passive measurement end.
  • the specific measurement is as follows: The default initial send rate. Send an ICMP message to the passive measurement end, where. For a lower send rate, when. When the rate of ICMP packets is sent, you need to ensure that the network does not become congested. On this premise, the active measurement terminal can select the value according to its needs.
  • Step 302 The active measurement end determines a sending rate of the current mixed message.
  • the first transmission rate of the mixed message It can be the link bandwidth of the bottleneck link; otherwise, it can be the preset initial transmission rate.
  • the transmission rate of the mixed packet needs to be increased. Specifically, the linear improvement and the exponential multiplication can be adopted as needed.
  • the transmission rate is linearly increased, the transmission rate increases at a slower rate, resulting in a longer test time, but the test result is more accurate.
  • the index rate is increased, the transmission rate increases faster and the test converges quickly.
  • the test results are relatively linear and increase the error of the transmission rate.
  • Step 303 The active measurement end sends the mixed message to the passive measurement end at the currently determined transmission rate; the passive measurement end returns the ICMP ⁇ message to the active measurement end.
  • the number of UDP packets is large, and preferably, an ICMP packet is sent after a continuous transmission of ⁇ : (for a positive integer) UDP packets, as shown in FIG. Schematic diagram of the mixed message.
  • the active sender uses the ICMP file as the start of the mixed message of the group, and then sends an ICMP 2 message after sending UDP packets continuously; and then continuously sends ⁇ : UDP packets. Send an ICMP 3 packet, and so on.
  • +1 ICMP ⁇ ⁇ text is called a set of mixed 4 ⁇ a transmission interval. It can be seen that the interval between two consecutive ICMP messages is long, and the interval at which the passive measurement end returns ICMP packets is also long.
  • the mixed message has a total of s ( s is a positive integer) ICMP message, and the ICMP message number is continuous, where : ⁇ is the first ICMP message, and ICMP S is the last ICMP message.
  • the number of UDP packets is (s - l)x
  • the passive measurement end when the active measurement end sends a UDP packet, the destination port of the UDP packet is set to the wrong destination port number, so that the passive measurement end receives the mixed packet, because of the UDP packet.
  • the destination port number is incorrect. Therefore, the passive measurement end directly discards the UDP packet, but only receives the ICMP packet and returns the ICMP packet to the active measurement end. Since the ICMP text file has the function of measuring the end-to-end data transmission function (including bandwidth measurement), and the UDP data is actively discarded by the passive measurement terminal, the passive measurement terminal does not need to cooperate with the bandwidth measurement. This greatly reduces the complexity of bandwidth measurement.
  • the proportion of the UDP packets in the mixed packets is far greater than the proportion of the ICMP packets in the mixed packets. Therefore, during the uplink (the active measurement terminal sends the mixed packets to the passive measurement terminal), the data volume of the mixed packets is If the UDP packet has been discarded, only the ICMP packet is returned. In the process of the downlink (the passive measurement end returns the ICMP packet to the active measurement end), only the ICMP packet is returned. The amount of data is greatly reduced, and the interval between ICMP packets is large. Therefore, ICMP packets are greatly reduced by network conditions, which helps improve the accuracy of bandwidth measurement.
  • Step 304 The active measurement end calculates the average packet loss rate of the ICMP.
  • the active measurement end collects the ICMP packet loss rate of each group of mixed packets in the n groups according to the n sets of mixed packets and the ICMP packets sent by the passive measurement end, and the ICMP packets of the n sets of mixed packets.
  • the packet loss rate is averaged to obtain the average packet loss rate of ICMP packets.
  • the active measurement end needs to count the average ICMP packet loss rate at the current transmission rate. Measure the available bandwidth of the path based on the packet loss rate of ICMP packets in a single group of mixed packets.
  • Good, « can take an integer greater than 1, and calculate the packet loss rate of ICMP packets of each group of mixed packets, that is, the packet loss rate ccn of a single group of ICMP packets; then take the packet loss rate of the n groups of ICMP packets. Average, get
  • the calculation of the packet loss rate ocn of a single group of ICMP packets in the following step is described in detail.
  • the active measurement terminal receives the last ICMP packet in the mixed packet returned by the passive measurement terminal, it confirms whether the received ICMP packet is lost. If the ICMP packet is lost, the single packet in the step is The calculation of the packet loss rate of the ICMP packet can be calculated by using the packet loss rate in step 301.
  • the limit bandwidth (maximum bandwidth) S of the path can be determined.
  • the existing technology can be used here, and will not be described again.
  • the sending time of the ICMP packet the first ICMP packet and the last one can be calculated.
  • the interval for sending ICMP packets is based on the time when ICMP packets are returned to the active measurement end.
  • the first ICMP message and the last one can be calculated.
  • the length of the mixed packet in bits.
  • the ICMP packet is considered to be a lost ICMP.
  • the packet loss rate threshold can be set according to the physical characteristics of the current path.
  • the average packet loss rate of the ICMP packets is "the rate of packet loss of the normal ICMP packets.
  • the rate of sending mixed packets is higher than the rate of sending ICMP packets when the network is normal. Therefore, "is greater than,” ⁇ ⁇ - ⁇ ⁇ , it is determined that the transmission rate of the mixed message before the current transmission rate is too low, and the most accurate path available bandwidth cannot be measured based on the current transmission rate.
  • the active measurement end stops at the current
  • the sending rate of the mixed packet is sent, and the sending rate of the mixed packet needs to be increased.
  • the process returns to step 302.
  • the active measuring end re-determines the sending rate of the mixed packet, based on the first sending rate of the mixed packet, and adopts linear improvement.
  • step 303 is performed to send the hybrid message to the passive measurement end again.
  • the number of the mixed message can be increased. , such as sending more sets of mixed messages than the last time, and / or increasing UDP messages and ICM in each group of mixed messages. The number of P messages.
  • step 306 indicating that the current transmission rate of the mixed message does not need to be increased. For the measurement of the available bandwidth of the path, it is already a reasonable transmission rate, and step 306 is continued.
  • Step 306 The active measurement end counts the average rate of the mixed measurement message received by the passive measurement end according to the ICMP message returned by the passive measurement end, and serves as the available bandwidth from the active measurement end to the passive measurement end.
  • the active measurement end collects the rate of each group of mixed packets in the n groups according to the n sets of mixed packets sent by the passive measurement end and the ICMP packets returned by the passive measurement end, and receives the n sets of mixed reports from the passive measurement end.
  • the rate of the text is averaged to obtain the average rate at which the passive measurement end receives the mixed message.
  • the rate at which the passive measurement terminal receives a single group of mixed packets is based on the premise that the ICMP packets returned by the passive measurement terminal are not affected by the network environment. However, in actual applications, the passive measurement terminal returns. ICMP packets are basically affected by the network environment. Therefore, the active measurement end needs to filter the returned ICMP packets, and select the ICMP packets that are least affected by the network environment. The rate of the text.
  • the specific ICMP packet sample filtering of the present invention adopts the following method:
  • the active measurement end selects 7; (the time when the ICMP packet numbered X returns to the active measurement end) and 7; the closest ICMP message, that is, the ICMP message that returns to the active measurement end closest to the ideal time, As a basis, the rate at which the passive measurement end receives the set of mixed messages is calculated.
  • sample filtering of ICMP 4 can also use other sample filtering algorithms in the prior art.
  • the present invention further provides a measurement device, as shown in FIG. 5, including: an active measurement terminal 10 and a passive measurement terminal 20, wherein
  • the active measurement terminal 10 is configured to send an ICMP message to the passive measurement terminal 20, and collect statistics on the normal ICMP packet loss rate of the passive measurement terminal 20; and also send a mixed message to the passive measurement terminal 20, and collect statistics.
  • the passive measurement terminal 20 is configured to receive ICMP packets and mixed packets.
  • the ICMP message sent by the active measurement terminal 10 is received, the ICMP message is returned to the active measurement terminal 10; when the hybrid message sent by the active measurement terminal 10 is received, the UDP message in the mixed message is received. The packet is discarded, and the ICMP packet in the mixed packet is returned to the active measuring terminal 10.

Abstract

A method and device for measuring available bandwidth using mixed messages are disclosed. The method includes that: an initiative measuring end sends Internet Control Message Protocol (ICMP) messages to a passiveness measuring end, and counts a packet loss rate of normal network ICMP messages to the passiveness measuring end (201); the initiative measuring end sends mixed messages to the passiveness measuring end, and counts an average packet loss rate of ICMP messages in the mixed messages to the passiveness measuring end, in which, the mixed messages comprise ICMP messages and User Datagram Protocol (UDP) messages (202); when the initiative measuring end estimates the difference between the average packet loss rate of ICMP messages and the packet loss rate of normal network ICMP messages is greater than a predefined threshold of packet loss rate, the average speed of mixed messages received by the passiveness measuring end is counted, and regarded as available bandwidth between the initiative measuring end and the passiveness measuring end (203). The solution reduces the complex degree of bandwidth measuring.

Description

一种利用混合报文测量可用带宽的方法和装置 技术领域  Method and device for measuring available bandwidth by using mixed message
本发明涉及计算机网络电子通信领域, 特别是指一种利用混合报文测 量可用带宽的方法和装置。 背景技术  The present invention relates to the field of computer network electronic communications, and more particularly to a method and apparatus for measuring available bandwidth using hybrid messages. Background technique
随着计算机技术的发展, 因特网 (Internet )逐渐成为人们获取信息、 相互交流的重要平台之一。 因特网在全球范围的急剧膨胀、 以及因特网中 实时业务和多媒体业务等新业务的快速发展, 导致了因特网网络性能的下 降。 目前, 影响网络性能的主要原因有链路带宽、 链路瓶颈、 应用与协议 的设计和主机系统等因素, 其中, 链路带宽是最重要的网络资源, 也是传 输路径性能的主要决定因素之一, 并且与网络瓶颈、 整体性能分析、 容量 规划等密切相关, 因此, 精确地测量带宽就成为因特网研究中的一个重要 课题。  With the development of computer technology, the Internet has gradually become one of the important platforms for people to obtain information and communicate with each other. The rapid expansion of the Internet on a global scale and the rapid development of new services such as real-time services and multimedia services in the Internet have led to a decline in the performance of Internet networks. At present, the main factors affecting network performance are link bandwidth, link bottleneck, application and protocol design, and host system. Among them, link bandwidth is the most important network resource and one of the main determinants of transmission path performance. And it is closely related to network bottlenecks, overall performance analysis, capacity planning, etc. Therefore, accurately measuring bandwidth has become an important topic in Internet research.
在网络带宽测量领域中, 按不同的测度, 网络带宽测量可分为链路带 宽、 路径带宽和可用带宽等不同方面的测量。 可用带宽在工程中是网络路 由、 流量工程、 QoS 控制中的一个关键参数, 相比于瓶颈带宽更具有实际 意义。  In the field of network bandwidth measurement, network bandwidth measurement can be divided into different aspects such as link bandwidth, path bandwidth and available bandwidth according to different measures. Available bandwidth is a key parameter in network routing, traffic engineering, and QoS control in engineering, and is more practical than bottleneck bandwidth.
链路带宽指一条链路的最大带宽、 即该条链路的最大数据传输速率 , 由链路的物理性质决定; 链路可用带宽指一条链路当前未被使用的带宽; 路径带宽为当前路径上的所有链路带宽的最小值, 路径带宽也被称为端到 端路径带宽; 路径可用带宽为当前路径上的所有链路可用带宽的最小值, 路径可用带宽也被称为端到端路径可用带宽。 在当前路径上的所有链路中, 链路可用带宽最小的链路被称为该路径的紧迫链路, 链路带宽最小的链路 被称为该路径的瓶颈链路。 The link bandwidth refers to the maximum bandwidth of a link, that is, the maximum data transmission rate of the link, which is determined by the physical properties of the link. The available bandwidth of the link refers to the bandwidth that is currently unused by a link. The path bandwidth is the current path. The minimum bandwidth of all links on the path, the path bandwidth is also called the end-to-end path bandwidth; the available bandwidth of the path is the minimum of the available bandwidth of all links on the current path, and the available bandwidth of the path is also called the end-to-end path. Available bandwidth. Among all the links on the current path, the link with the smallest available bandwidth of the link is called the urgent link of the path, and the link with the smallest link bandwidth. A bottleneck link called the path.
图 1 所示为端到端带宽示意图, 发送端和接收端之间的路径由三条链 路 23组成, 其中, 、 c2、 c3分别为链路 A、 23的链路带宽; 阴影部分表示当前链路已经使用的带宽, 则 4、 4、 4分别为链路 A、 2、 3未被使用的带宽、 即链路可用带宽。 从图中可以看出 < c2 < c3、 即链 路 i的链路带宽最小, 决定了发送端到接收端的路径带宽, 则链路 A是发 送端和接收端之间的瓶颈链路; 另外, 4 < 4 <4、 即链路 3的链路可用 带宽最小, 决定了发送端到接收端的路径可用带宽, 则链路 3为该路径的 紧迫链路。 Figure 1 shows the end-to-end bandwidth diagram. The path between the sender and the receiver consists of three links 2 and 3 , where c 2 and c 3 are the link bandwidths of links A, 2 , and 3 , respectively. The shaded portion indicates the bandwidth that the current link has used, and then 4, 4, and 4 are the unused bandwidth of link A, 2 , and 3, that is, the available bandwidth of the link. It can be seen from the figure that < c 2 < c 3 , that is, the link bandwidth of the link i is the smallest, and the path bandwidth of the sender to the receiver is determined, and the link A is the bottleneck link between the sender and the receiver; In addition, 4 < 4 <4, that is, the link available bandwidth of link 3 is the smallest, which determines the available bandwidth of the path from the sender to the receiver, and link 3 is the urgent link of the path.
现有的带宽测量方法多是基于时间戳的包队列算法, 主要测量模型有 探测间隔模型( PGM, Packet Gap Model )和探测速率模型( PRM, Packet Rate Model )。其中,基于 PGM模型的算法主要是分析探测包队列的包间隔变化, 根据变化趋势来确定带宽的估计值, 也被称为基于时间戳的带宽测量方法; 基于 PRM模型的算法主要是在发送端以一定的速率发送探测包队列,根据 接收端接收探测包队列的速率, 动态调整探测包队列的发送速率, 从而确 定一个匹配的临界速率来估算待测带宽。  The existing bandwidth measurement methods are mostly time-based packet queue algorithms. The main measurement models include PGM (Packet Gap Model) and Detection Rate Model (PRM, Packet Rate Model). The algorithm based on the PGM model mainly analyzes the packet interval variation of the probe packet queue, and determines the bandwidth estimation value according to the change trend, which is also called a timestamp-based bandwidth measurement method; the PRM model-based algorithm is mainly at the transmitting end. The probe packet queue is sent at a certain rate, and the transmission rate of the probe packet queue is dynamically adjusted according to the rate at which the receiver receives the probe packet queue, thereby determining a matching critical rate to estimate the bandwidth to be tested.
但是, 现有的基于 PGM和 PRM的带宽测量方法主要存在以下缺点: However, the existing PGM and PRM based bandwidth measurement methods mainly have the following disadvantages:
1. 上述两种测量方法一般需要被动测量端的配合工作, 也就是说, 被 动测量端需要根据探测包队列的一些信息进行相应的计算, 并将这些计算 信息返回给主动测量端, 两测量端需要配合工作才能测试出链路带宽。 这 种情况下, 需要被动测量端采取安装相应的测量程序等辅助措施, 这样降 低了测量方法的实用性和灵活性。 1. The above two measurement methods generally require the cooperation of the passive measurement end. That is to say, the passive measurement end needs to perform corresponding calculation according to some information of the probe packet queue, and returns the calculation information to the active measurement end. Work with the link to test the link bandwidth. In this case, the passive measuring end is required to take auxiliary measures such as installing a corresponding measuring program, which reduces the practicability and flexibility of the measuring method.
2. 上述两种测量方法计算流程复杂, 特别是基于 PGM的带宽测量方 法, 对于定时精度等要求比较严格, 计算的复杂度较高。 发明内容 2. The calculation process of the above two measurement methods is complicated, especially the bandwidth measurement method based on PGM. The requirements for timing accuracy are strict, and the computational complexity is high. Summary of the invention
有鉴于此, 本发明的主要目的在于提供一种利用混合报文测量可用带 宽的方法和装置, 能够降低带宽测量的复杂度。  In view of this, it is a primary object of the present invention to provide a method and apparatus for measuring available bandwidth using mixed messages, which can reduce the complexity of bandwidth measurement.
为达到上述目的, 本发明的技术方案是这样实现的:  In order to achieve the above object, the technical solution of the present invention is achieved as follows:
本发明提供了一种利用混合报文测量可用带宽的方法, 该方法包括: 主动测量端发送网际控制报文协议(ICMP )报文给被动测量端, 并统 计到被动测量端的网络正常 ICMP报文丟包率;  The present invention provides a method for measuring available bandwidth by using a hybrid packet, the method comprising: the active measurement end transmitting an Internet Control Message Protocol (ICMP) message to the passive measurement end, and collecting the normal ICMP message of the network on the passive measurement end Packet loss rate;
主动测量端发送混合报文给所述被动测量端, 并统计到被动测量端的 所述混合报文的 ICMP报文平均丟包率;其中,所述混合报文包括所述 ICMP 报文和用户数据报文协议 ( UDP )报文; 率之差大于预设的丟包率阈值时, 统计所述被动测量端接收混合报文的平 均速率, 作为主动测量端到被动测量端的可用带宽。  The active measurement end sends the mixed packet to the passive measurement end, and collects the average packet loss rate of the ICMP packet of the hybrid packet at the passive measurement end. The hybrid packet includes the ICMP packet and the user data. Packet protocol (UDP) packet; when the rate difference is greater than the preset packet loss rate threshold, the average rate at which the passive measurement end receives the mixed packet is counted as the available bandwidth from the active measurement end to the passive measurement end.
主动测量端发送 ICMP 文给被动测量端时, 该方法进一步包括: 所 述主动测量端以预设的初始发送速率发送 ICMP报文给被动测量端, 所述 被动测量端接收到所述 ICMP报文后,将所述 ICMP报文返回给所述主动测 量端;  When the active measurement end sends the ICMP file to the passive measurement end, the method further includes: the active measurement end sends the ICMP message to the passive measurement end at the preset initial transmission rate, and the passive measurement end receives the ICMP message. Returning the ICMP message to the active measurement end;
相应的, 统计所述网络正常 ICMP 4艮文丟包率, 具体为: 所述主动测 量端依据自身发送的 ICMP报文、和被动测量端返回的 ICMP报文,统计所 述网络正常 ICMP报文丟包率。  Correspondingly, the normal ICMP packet loss rate of the network is counted, and the following is: the active measurement end collects the normal ICMP packet of the network according to the ICMP packet sent by the active measurement end and the ICMP packet sent by the passive measurement end. Packet loss rate.
所述主动测量端发送混合报文给所述被动测量端之前, 该方法还包括: 所述主动测量端确定所述混合报文的发送速率;  Before the active measurement end sends the hybrid message to the passive measurement end, the method further includes: determining, by the active measurement end, a transmission rate of the hybrid message;
相应的, 确定所述混合报文的发送速率, 具体为:  Correspondingly, determining a sending rate of the mixed packet is specifically:
当所述主动测量端第一次向所述被动测量端发送所述混合报文时, 所 述混合报文的第一次发送速率为所述预设的初始发送速率、 或者为所述主 动测量端到被动测量端之间瓶颈链路的链路带宽。 When the active measurement end sends the hybrid message to the passive measurement end for the first time, the first transmission rate of the hybrid message is the preset initial transmission rate, or is the primary The link bandwidth of the bottleneck link between the measurement end and the passive measurement end.
所述主动测量端发送的混合报文为 n组混合报文,其中 n为大于等于 1 的整数。  The mixed message sent by the active measuring end is n sets of mixed messages, where n is an integer greater than or equal to 1.
所述主动测量端发送混合报文给所述被动测量端时, 该方法还包括: 所述主动测量端将所述 UDP报文的目的端口设置为错误的目的端口;  When the active measurement end sends the hybrid packet to the passive measurement end, the method further includes: the active measurement end setting the destination port of the UDP packet as an incorrect destination port;
相应的, 所述被动测量端接收到所述混合报文后, 该方法还包括: 所 述被动测量端依据所述错误的目的端口将所述 UDP报文丟弃, 并将所述混 合报文中的 ICMP报文返回给所述主动测量端。  Correspondingly, after the passive measurement end receives the mixed message, the method further includes: the passive measurement end discards the UDP packet according to the wrong destination port, and the hybrid message is The ICMP message in is returned to the active measurement terminal.
所述主动测量端统计所述 ICMP ^艮文平均丟包率, 具体为:  The active measurement end calculates the average packet loss rate of the ICMP ^ , ,, specifically:
所述主动测量端依据自身发送的 n组混合报文、 和被动测量端返回的 ICMP报文, 统计所述 n组中每组混合报文的 ICMP报文丟包率, 并对所述 n组混合报文的 ICMP报文丟包率进行平均,得到所述 ICMP报文平均丟包 率。  The active measurement end collects the packet loss rate of the ICMP packet of each group of the mixed packets in the n groups according to the n sets of the mixed packets and the ICMP packets sent by the passive measurement end, and compares the packet loss rate of the ICMP packets in the n groups. The average packet loss rate of the ICMP packets is averaged, and the average packet loss rate of the ICMP packets is obtained.
所述主动测量端判定所述 ICMP ^艮文平均丟包率与所述网络正常 ICMP 报文丟包率之差小于等于预设的丟包率阈值时, 该方法还包括: 所述主动 测量端重新确定所述混合报文的发送速率, 具体为: 以所述混合报文的第 一次发送速率为基础, 采用线性提高、 或者指数倍提高的方式增加所述发 送速率。  The method further includes: the active measurement end, when the active measurement end determines that the difference between the average ICMP packet loss rate and the normal ICMP packet loss rate of the network is less than or equal to a preset packet loss rate threshold, the method further includes: And determining, according to the first sending rate of the mixed packet, the sending rate is increased by linearly increasing or exponentially increasing.
所述主动测量端统计所述被动测量端接收所述混合报文的平均速率, 具体为: 所述主动测量端依据自身发送的 n组混合 ^艮文、 和被动测量端返 回的 ICMP报文, 统计所述被动测量端接收所述 n组中每组混合报文的速 率, 并对所述被动测量端接收所述 n组混合报文的速率进行平均, 得到所 述被动测量端接收所述混合报文的平均速率。  The active measurement end collects the average rate at which the passive measurement end receives the mixed message, and specifically: the active measurement end is based on the n sets of mixed packets sent by the active measurement end, and the ICMP message returned by the passive measurement end. And the rate at which the passive measurement end receives the mixed message of each group in the n groups, and averages the rate at which the passive measurement terminal receives the n groups of mixed packets, to obtain the passive measurement end receiving the hybrid The average rate of messages.
本发明还提供了一种利用混合报文测量可用带宽的装置, 该装置包括: 主动测量端和被动测量端, 其中, 所述主动测量端, 用于发送 ICMP报文给所述被动测量端, 并统计到 所述被动测量端的网络正常 ICMP报文丟包率; 还用于发送混合报文给所 述被动测量端, 并统计到所述被动测量端的所述混合报文的 ICMP报文平 丟包率之差大于预设的丟包率阈值时, 统计所述被动测量端接收所述混合 报文的平均速率, 作为主动测量端到被动测量端的可用带宽; The present invention also provides an apparatus for measuring available bandwidth by using a mixed message, the apparatus comprising: an active measuring end and a passive measuring end, wherein The active measurement end is configured to send an ICMP message to the passive measurement end, and collect a packet loss rate of the normal ICMP packet on the passive measurement end; and send the mixed message to the passive measurement end. And counting, when the difference between the flat packet loss rate of the mixed packet of the passive measurement end is greater than a preset packet loss rate threshold, counting the average rate at which the passive measurement end receives the mixed packet, as The available bandwidth from the active measurement end to the passive measurement end;
所述被动测量端, 用于接收所述 ICMP报文和所述混合报文。  The passive measurement end is configured to receive the ICMP packet and the hybrid packet.
进一步地, 所述被动测量端还用于, 在接收到所述主动测量端发送的 ICMP报文时, 返回所述 ICMP报文给主动测量端; 在接收到所述主动测量 端发送的混合报文时, 将所述混合报文中的 UDP报文丟弃, 并将混合报文 中的 ICMP报文返回给主动测量端。  Further, the passive measurement end is further configured to: when receiving the ICMP message sent by the active measurement end, return the ICMP message to the active measurement end; and receive the hybrid report sent by the active measurement end The UDP packet in the mixed packet is discarded, and the ICMP packet in the mixed packet is returned to the active measurement end.
本发明的带宽测量方案, 由主动测量端发送 ICMP ^艮文给被动测量端 , 并统计到被动测量端的网络正常 ICMP ^艮文丟包率; 由主动测量端发送混 合报文给被动测量端, 并统计到被动测量端的 ICMP报文平均丟包率; 当 主动测量端判定 ICMP报文平均丟包率与网络正常 ICMP报文丟包率之差大 于预设的丟包率阈值时, 统计被动测量端接收混合报文的平均速率; 该平 均速率即为测量的路径可用带宽。 另外, 该方案中, 主动测量端发送的混 合报文中 UDP报文的目的端口被设置为错误的, 如此, 被动测量端可以直 接将 UDP报文丟弃, 只向主动测量端返回 ICMP报文。 由此可见, 本发明 中的带宽测量不需要被动测量端的配合工作, 降低了测量工作的复杂度, 同时本发明的测量方法的实用性和灵活性也比较高。 附图说明  In the bandwidth measurement scheme of the present invention, the active measurement terminal sends an ICMP^艮 message to the passive measurement end, and the normal ICMP/艮 packet loss rate of the network on the passive measurement end is counted; the active measurement end sends a mixed message to the passive measurement end, The average packet loss rate of the ICMP packets on the passive measurement end is counted. When the active measurement end determines that the difference between the average packet loss rate of the ICMP packet and the packet loss rate of the normal ICMP packet is greater than the preset packet loss rate threshold, the statistical passive measurement is performed. The average rate at which the terminal receives the mixed message; the average rate is the measured path available bandwidth. In addition, in this solution, the destination port of the UDP packet sent by the active measurement end is set to be incorrect. Therefore, the passive measurement end can directly discard the UDP packet and return only the ICMP packet to the active measurement end. . It can be seen that the bandwidth measurement in the present invention does not require the cooperation of the passive measurement end, which reduces the complexity of the measurement work, and the measurement method of the present invention is also more practical and flexible. DRAWINGS
图 1为端到端带宽示意图;  Figure 1 is a schematic diagram of end-to-end bandwidth;
图 2为本发明利用混合报文测量可用带宽的方法流程示意图; 图 3为本发明利用混合报文测量可用带宽一实施例的方法流程示意图; 图 4为本发明的混合 ·^艮文构成示意图; 2 is a schematic flowchart of a method for measuring available bandwidth by using a mixed packet according to the present invention; FIG. 3 is a schematic flowchart of a method for measuring available bandwidth by using a mixed packet according to the present invention; Figure 4 is a schematic view showing the structure of the mixed type of the present invention;
图 5为本发明利用混合报文测量可用带宽的装置结构示意图。 具体实施方式  FIG. 5 is a schematic structural diagram of an apparatus for measuring available bandwidth by using a mixed message according to the present invention. detailed description
下面结合附图和具体实施例对本发明的技术方案进一步详细阐述。 本发明中的带宽测量主要是对端到端的路径可用带宽的测量, 其中一 端为主动测量端, 则另外一端为被动测量端; 主动测量端通过向被动测量 端发送混合报文的方式, 对两端的路径可用带宽进行测量。  The technical solutions of the present invention are further elaborated below in conjunction with the accompanying drawings and specific embodiments. The bandwidth measurement in the present invention is mainly for measuring the available bandwidth of the end-to-end path, where one end is the active measurement end, and the other end is the passive measurement end; the active measurement end sends the mixed message to the passive measurement end, The path of the end can be measured with bandwidth.
本发明中混合报文指用户数据报文协议( UDP, User Datagram Protocol ) 才艮文和网际控制才艮文十办议 ( ICMP, Internet Control Message Protocol )才艮文。  In the present invention, the mixed message refers to the User Datagram Protocol (UDP) and the Internet Control Message Protocol (ICMP).
本发明利用混合 ·^艮文测量可用带宽的流程如图 2所示, 包括: 步驟 201 , 主动测量端发送 ICMP报文给被动测量端, 并统计到被动测 量端的网络正常 ICMP报文丟包率;  The process of measuring the available bandwidth by using the hybrid method is as shown in FIG. 2, which includes the following steps: Step 201: The active measurement end sends an ICMP message to the passive measurement end, and collects the normal ICMP packet loss rate of the passive measurement end. ;
步驟 202, 主动测量端发送混合报文给被动测量端, 并统计到被动测量 端的混合报文的 ICMP报文平均丟包率;其中,混合报文包括 ICMP报文和 Step 202: The active measurement end sends the mixed packet to the passive measurement end, and collects the average packet loss rate of the ICMP packet of the mixed packet at the passive measurement end. The mixed packet includes the ICMP packet and
UDP才艮文; 文丟包率之差大于预设的丟包率阈值时, 统计被动测量端接收混合报文的 平均速率, 作为主动测量端到被动测量端的可用带宽。 The UDP packet is used. When the packet loss rate is greater than the preset packet loss threshold, the average rate at which the passive measurement end receives the hybrid packet is counted as the available bandwidth from the active measurement end to the passive measurement end.
下面结合具体的实施例来说明本发明测量带宽的方法, 如图 3 所示, 包括:  The method for measuring bandwidth of the present invention is described below in conjunction with a specific embodiment. As shown in FIG. 3, the method includes:
步驟 301 , 主动测量端测量网络正常 ICMP ^艮文丟包率。  Step 301: The active measurement end measures a normal ICMP^ packet loss rate of the network.
主动测量端以预设的初始发送速率发送 ICMP报文给被动测量端, 被 动测量端接收到 ICMP报文后,将 ICMP报文返回给主动测量端,主动测量 端依据自身发送的 ICMP报文和被动测量端返回的 ICMP报文,统计主动测 量端到被动测量端的网络正常 ICMP报文丟包率, 具体为: 主动测量端以 预设的初始发送速率 。向被动测量端发送 ICMP报文, 其中 。为一个较低 的发送速率, 当以 。为 ICMP报文的发送速率时, 需要确保不会造成网络 的拥塞, 在该前提下, 主动测量端可以根据自身需要进行 ^值的选择。 假设主动测量端向被动测量端一次发送了 ζ ( ζ为正整数)个 ICMP报 文, 且被动测量端返回了 ζ' ( ζ'为正整数)个 ICMP报文, 则当前路径上 网络正常 ICMP报文丟包率为 = 。 The active measurement end sends the ICMP message to the passive measurement end at the preset initial transmission rate. After receiving the ICMP message, the passive measurement end returns the ICMP message to the active measurement end. The active measurement end is based on the ICMP message sent by itself. The ICMP packet returned by the passive measurement terminal is used to collect the normal ICMP packet loss rate from the active measurement end to the passive measurement end. The specific measurement is as follows: The default initial send rate. Send an ICMP message to the passive measurement end, where. For a lower send rate, when. When the rate of ICMP packets is sent, you need to ensure that the network does not become congested. On this premise, the active measurement terminal can select the value according to its needs. Suppose the active measurement end sends ζ (ζ is a positive integer) ICMP message to the passive measurement end once, and the passive measurement end returns IC' ( ζ ' is a positive integer) ICMP message, then the network is normal ICMP on the current path. Packet loss rate = .
z  z
步驟 302, 主动测量端确定当前混合报文的发送速率。  Step 302: The active measurement end determines a sending rate of the current mixed message.
当主动测量端第一次向被动测量端发送混合报文, 且在已知当前路径、 即主动测量端到被动测量端的瓶颈链路的链路带宽时, 该混合报文的第一 次发送速率可以为该瓶颈链路的链路带宽; 否则, 可以为预设的初始发送 速率 。  When the active measurement end sends a hybrid message to the passive measurement end for the first time, and the link rate of the bottleneck link of the current path, that is, the active measurement end to the passive measurement end, is known, the first transmission rate of the mixed message It can be the link bandwidth of the bottleneck link; otherwise, it can be the preset initial transmission rate.
如果混合报文的第一次发送速率不能满足路径可用带宽的测量, 则需 要提高混合报文的发送速率, 具体的, 根据需要可以采用线性提高和指数 倍提高两种手段。 其中, 线性提高发送速率时, 发送速率的增长速度较慢, 导致测试时间相对较长, 但是测试结果更为准确; 指数倍提高发送速率时, 发送速率的增长速度较快, 测试收敛快, 但是测试结果相对线性提高发送 速率的误差较大。  If the first transmission rate of the hybrid packet cannot meet the measurement of the available bandwidth of the path, the transmission rate of the mixed packet needs to be increased. Specifically, the linear improvement and the exponential multiplication can be adopted as needed. When the transmission rate is linearly increased, the transmission rate increases at a slower rate, resulting in a longer test time, but the test result is more accurate. When the index rate is increased, the transmission rate increases faster and the test converges quickly. The test results are relatively linear and increase the error of the transmission rate.
步驟 303 ,主动测量端以当前确定的发送速率向被动测量端发送混合报 文; 被动测量端向主动测量端返回 ICMP ^艮文。  Step 303: The active measurement end sends the mixed message to the passive measurement end at the currently determined transmission rate; the passive measurement end returns the ICMP^ message to the active measurement end.
本发明的混合报文中, UDP报文的数量较多, 较佳地, 可以设置连续 发送^: ( 为正整数)个 UDP报文后发送一个 ICMP报文, 如图 4所示为 一组混合报文的结构示意图, 主动发送端以 ICMP 文作为该组混合报文 的起始, 然后连续发送 ^:个 UDP报文后发送一个 ICMP2报文; 再连续发送 ^:个 UDP报文后发送一个 ICMP3报文, 以此类推, 以连续发送^:个 UDP报 文后发送一个 ICMPS报文作为该组混合报文的结束, 可以将 ^:个 UDP报文 +1个 ICMP ^艮文称为一组混合 4艮文的一个发送间隔。 由此可见, 两个连续 ICMP报文的发送间隔很长, 则被动测量端返回 ICMP报文的间隔也很长。 In the mixed message of the present invention, the number of UDP packets is large, and preferably, an ICMP packet is sent after a continuous transmission of ^: (for a positive integer) UDP packets, as shown in FIG. Schematic diagram of the mixed message. The active sender uses the ICMP file as the start of the mixed message of the group, and then sends an ICMP 2 message after sending UDP packets continuously; and then continuously sends ^: UDP packets. Send an ICMP 3 packet, and so on. After sending ^: UDP packets continuously, send an ICMP S packet as the end of the mixed packet. You can send ^: UDP packets. +1 ICMP ^ 艮 text is called a set of mixed 4 的 a transmission interval. It can be seen that the interval between two consecutive ICMP messages is long, and the interval at which the passive measurement end returns ICMP packets is also long.
该混合报文中共有 s ( s为正整数)个 ICMP报文, 且 ICMP报文的编 号是连续的, 其中, :^^^ 为第一个 ICMP报文、 ICMPS为最后一个 ICMP 报文; UDP报文的数量为(s - l)x The mixed message has a total of s ( s is a positive integer) ICMP message, and the ICMP message number is continuous, where : ^^^ is the first ICMP message, and ICMP S is the last ICMP message. The number of UDP packets is (s - l)x
需要指出的是, 本发明中主动测量端在发送 UDP报文时, 将 UDP报 文的目的端口设置为错误的目的端口号, 如此被动测量端在接收到混合报 文时, 由于 UDP报文的目的端口号错误, 因此被动测量端直接将 UDP报 文丟弃, 而只是接收 ICMP报文, 并向主动测量端返回 ICMP报文。 由于 ICMP报文本身就具有测量端到端的数据传输功能(包括带宽测量), 并且 UDP ^艮文由被动测量端主动丟弃, 因此, 被动测量端不需要对带宽测量作 任何的配合工作, 这样就大大地降低了带宽测量的复杂度。  It should be noted that, in the present invention, when the active measurement end sends a UDP packet, the destination port of the UDP packet is set to the wrong destination port number, so that the passive measurement end receives the mixed packet, because of the UDP packet. The destination port number is incorrect. Therefore, the passive measurement end directly discards the UDP packet, but only receives the ICMP packet and returns the ICMP packet to the active measurement end. Since the ICMP text file has the function of measuring the end-to-end data transmission function (including bandwidth measurement), and the UDP data is actively discarded by the passive measurement terminal, the passive measurement terminal does not need to cooperate with the bandwidth measurement. This greatly reduces the complexity of bandwidth measurement.
上述 UDP报文占混合报文的比例远远大于 ICMP报文占混合报文的比 例, 如此, 在上行(主动测量端向被动测量端发送混合报文)过程中, 由 于混合报文的数据量较大, 致使 ICMP报文的发送受网络状况的影响较大; 在下行(被动测量端向主动测量端返回 ICMP报文 )过程中, 由于 UDP报 文已经被丟弃, 返回的只有 ICMP报文, 数据量大大减少, 且 ICMP报文之 间的间隔很大, 因此, ICMP报文受到网络状况的影响大大减小, 这样有助 于提高带宽测量的精度。  The proportion of the UDP packets in the mixed packets is far greater than the proportion of the ICMP packets in the mixed packets. Therefore, during the uplink (the active measurement terminal sends the mixed packets to the passive measurement terminal), the data volume of the mixed packets is If the UDP packet has been discarded, only the ICMP packet is returned. In the process of the downlink (the passive measurement end returns the ICMP packet to the active measurement end), only the ICMP packet is returned. The amount of data is greatly reduced, and the interval between ICMP packets is large. Therefore, ICMP packets are greatly reduced by network conditions, which helps improve the accuracy of bandwidth measurement.
步驟 304, 主动测量端统计 ICMP ^艮文平均丟包率。  Step 304: The active measurement end calculates the average packet loss rate of the ICMP.
主动测量端依据自身发送的 n组混合报文和被动测量端返回的 ICMP 报文, 统计 n组中每组混合报文的 ICMP报文丟包率, 并对这 n组混合报 文的 ICMP报文丟包率进行平均, 得到 ICMP报文平均丟包率。  The active measurement end collects the ICMP packet loss rate of each group of mixed packets in the n groups according to the n sets of mixed packets and the ICMP packets sent by the passive measurement end, and the ICMP packets of the n sets of mixed packets. The packet loss rate is averaged to obtain the average packet loss rate of ICMP packets.
具体的, 主动测量端需要统计在当前发送速率下的 ICMP ^艮文平均丟 包率。 仅以单组混合报文的 ICMP报文丟包率为依据来测量路径可用带宽 时, 会产生较大的误差, 因此, 在进行一次路径可用带宽的测量时, 上述 步驟 303中需要主动测量端一次发送《 ( « = 1,2,3... )组混合报文, 较佳地, «可以取大于 1的整数, 并分别计算每组混合报文的 ICMP报文丟包率、 即 单组 ICMP报文丟包率 ccn; 然后取这 n组 ICMP报文丟包率的平均值,得到 Specifically, the active measurement end needs to count the average ICMP packet loss rate at the current transmission rate. Measure the available bandwidth of the path based on the packet loss rate of ICMP packets in a single group of mixed packets. When a measurement of the available bandwidth of the path is performed, in the above step 303, the active measurement terminal is required to transmit the mixed message of (« = 1, 2, 3...) at a time. Good, « can take an integer greater than 1, and calculate the packet loss rate of ICMP packets of each group of mixed packets, that is, the packet loss rate ccn of a single group of ICMP packets; then take the packet loss rate of the n groups of ICMP packets. Average, get
ICMP报文平均丟包率" = ai + a2 + - + 0 n ,如此可以提高 K P报文丟包率 n The average packet loss rate of ICMP packets is = ai + a2 + - + 0 n , which can increase the packet loss rate of KP packets.
的计算精度。 The accuracy of the calculation.
下面对该步驟中单组 ICMP报文丟包率 ocn的计算进行详细的说明。 当主动测量端接收到被动测量端返回的该组混合报文中的最后一个 ICMP报文时, 确认其接收到的 ICMP报文是否有丟失; 如果 ICMP报文有 丟失, 则该步驟中单组 ICMP报文丟包率 的计算可以采用步驟 301 中丟 包率的计算方法。  The calculation of the packet loss rate ocn of a single group of ICMP packets in the following step is described in detail. When the active measurement terminal receives the last ICMP packet in the mixed packet returned by the passive measurement terminal, it confirms whether the received ICMP packet is lost. If the ICMP packet is lost, the single packet in the step is The calculation of the packet loss rate of the ICMP packet can be calculated by using the packet loss rate in step 301.
如果 ICMP报文没有丟失, 则采用下面的计算方法:  If the ICMP packet is not lost, the following calculation method is used:
根据当前的路径情况可以确定该路径的极限带宽 (最大带宽) S , 此 处可以采用现有技术, 不再赘述; 根据 ICMP报文的发送时间, 可以计算 出第一个 ICMP报文和最后一个 ICMP报文的发送间隔为 根据 ICMP报 文返回到主动测量端的时间, 可以计算出第一个 ICMP ^艮文和最后一个  According to the current path situation, the limit bandwidth (maximum bandwidth) S of the path can be determined. The existing technology can be used here, and will not be described again. According to the sending time of the ICMP packet, the first ICMP packet and the last one can be calculated. The interval for sending ICMP packets is based on the time when ICMP packets are returned to the active measurement end. The first ICMP message and the last one can be calculated.
ICMP报文返回到主动测量端的时间间隔为 + 其中, Δ = , 为每 The interval at which ICMP packets are returned to the active measurement end is + where Δ = , for each
Β  Β
组混合报文的长度, 单位为 bit; 当某个 ICMP报文返回到主动测量端的时 间与第一个 ICMP报文返回到主动测量端的时间间隔大于 + 时, 认为该 ICMP报文为丟失的 ICMP报文,如此可以得出该组混合报文中丟失的 ICMP 报文的个数 则"„ = ^。 The length of the mixed packet, in bits. When the time interval between the return of an ICMP packet to the active measurement end and the time when the first ICMP packet is returned to the active measurement end is greater than +, the ICMP packet is considered to be a lost ICMP. The message, so that the number of ICMP messages lost in the mixed message of the group is "„ = ^.
s  s
差是否大于预设的丟包率阈值, 如果否, 返回步驟 302; 如果是, 执行步驟 306。 丟包率阈值 可以根据当前路径的物理特性进行设置。 Whether the difference is greater than a preset packet loss rate threshold, if no, return to step 302; if yes, go to step 306. The packet loss rate threshold can be set according to the physical characteristics of the current path.
ICMP报文平均丟包率为"、 网络正常 ICMP报文丟包率为 , 一般情 况下 , 由于混合报文的发送速率比网络正常时 ICMP报文的发送速率要高 , 因此, "要大于 , ^^ α- ≤γ , 则判定混合报文在当前发送速率下的 前的发送速率过低, 以当前发送速率为依据无法测量出最精确的路径可用 带宽, 此时, 主动测量端停止以当前的发送速率发送混合报文, 需要提高 混合报文的发送速率,返回执行步驟 302, 主动测量端重新确定混合报文的 发送速率, 以混合报文的第一次发送速率为基础, 采用线性提高、 或者指 数倍提高的方式增加发送速率, 然后执行步驟 303 , 再次发送混合报文给被 动测量端; 需要指出的是, 主动测量端再次发送混合报文时, 可以选择增 加混合报文的发送数量, 如相比上一次多发送几组混合报文、 和 /或增加每 组混合报文中的 UDP报文和 ICMP报文的数量。  The average packet loss rate of the ICMP packets is "the rate of packet loss of the normal ICMP packets. Generally, the rate of sending mixed packets is higher than the rate of sending ICMP packets when the network is normal. Therefore, "is greater than," ^^ α- ≤ γ , it is determined that the transmission rate of the mixed message before the current transmission rate is too low, and the most accurate path available bandwidth cannot be measured based on the current transmission rate. At this time, the active measurement end stops at the current The sending rate of the mixed packet is sent, and the sending rate of the mixed packet needs to be increased. The process returns to step 302. The active measuring end re-determines the sending rate of the mixed packet, based on the first sending rate of the mixed packet, and adopts linear improvement. Or, the exponential multiplication is increased to increase the transmission rate, and then step 303 is performed to send the hybrid message to the passive measurement end again. It should be noted that when the active measurement end sends the mixed message again, the number of the mixed message can be increased. , such as sending more sets of mixed messages than the last time, and / or increasing UDP messages and ICM in each group of mixed messages. The number of P messages.
^^ -β > γ, 说明混合报文当前的发送速率不需要再提高, 对于路 径可用带宽的测量来说, 已经是一个比较合理的发送速率, 继续执行步驟 306。  ^^ -β > γ, indicating that the current transmission rate of the mixed message does not need to be increased. For the measurement of the available bandwidth of the path, it is already a reasonable transmission rate, and step 306 is continued.
步驟 306, 主动测量端依据被动测量端返回的 ICMP报文, 统计被动测 量端接收混合报文的平均速率, 作为主动测量端到被动测量端的可用带宽。  Step 306: The active measurement end counts the average rate of the mixed measurement message received by the passive measurement end according to the ICMP message returned by the passive measurement end, and serves as the available bandwidth from the active measurement end to the passive measurement end.
主动测量端依据自身发送的 n组混合报文和被动测量端返回的 ICMP 报文, 统计被动测量端接收该 n组中每组混合报文的速率, 并对被动测量 端接收该 n组混合报文的速率进行平均, 得到被动测量端接收混合报文的 平均速率。  The active measurement end collects the rate of each group of mixed packets in the n groups according to the n sets of mixed packets sent by the passive measurement end and the ICMP packets returned by the passive measurement end, and receives the n sets of mixed reports from the passive measurement end. The rate of the text is averaged to obtain the average rate at which the passive measurement end receives the mixed message.
具体为, 主动测量端需要测量在当前发送速率下, 被动测量端接收《 ( « = 1,2,3... )组混合报文的平均速率: 主动测量端先计算出被动测量端接 收单组混合报文的速率 然后再统计这《组速率的平均值 即被动测量 端接收混合报文的平均速率 = Fl + F2 + - + F"。 则该平均速率即为测量得 Specifically, the active measurement terminal needs to measure the average rate at which the passive measurement terminal receives the mixed message of the ( ( « 1 , 2, 3... ) group at the current transmission rate: the active measurement terminal first calculates the passive measurement terminal reception ticket. The rate at which the group mixes the packets is then re-stated. The average of the group rates is the passive measurement. The average rate at which the terminal receives the mixed message = Fl + F2 + - + F ". Then the average rate is measured.
Π  Π
到的当前路径、 即主动测量端到被动测量端的可用带宽。 The current path to the active bandwidth from the active measurement end to the passive measurement end.
主动测量端统计被动测量端接收单组混合报文的速率是基于被动测量 端返回的 ICMP报文不受网络环境的影响这一前提条件的, 但是, 在实际 的应用中, 被动测量端返回的 ICMP报文基本上都会受到网络环境的影响, 因此, 主动测量端需要对返回的 ICMP报文进行样本过滤, 选取受网络环 境影响最小的 ICMP报文作为依据来统计被动测量端接收单组混合报文的 速率。  The rate at which the passive measurement terminal receives a single group of mixed packets is based on the premise that the ICMP packets returned by the passive measurement terminal are not affected by the network environment. However, in actual applications, the passive measurement terminal returns. ICMP packets are basically affected by the network environment. Therefore, the active measurement end needs to filter the returned ICMP packets, and select the ICMP packets that are least affected by the network environment. The rate of the text.
具体的本发明的 ICMP报文样本过滤采用如下的方法:  The specific ICMP packet sample filtering of the present invention adopts the following method:
1、 假设主动测量端接收到被动测量端返回的 ICMP报文的最小编号为 ί、 最大编号为 j , 且最小编号 ICMP ^艮文和最大编号 ICMP ^艮文返回到主 动测量端的时间分别为 7和 Tj ,则 ICMP报文返回到主动测量端的平均时间  1. It is assumed that the minimum number of ICMP messages returned by the active measurement terminal to the passive measurement terminal is ί, the maximum number is j, and the minimum number of ICMP^艮 and the maximum number ICMP^艮 returns to the active measurement end respectively. And Tj, the average time that ICMP packets are returned to the active measurement end
T—T T-T
间隔为 Δ =」 ~~ -; The interval is Δ =" ~~ -;
2、 设编号为 X ( i<x≤ j )的 ICMP报文返回到主动测量端的理想时间 ^/Τχ' = Τί + (χ-ι)χΛΤ; 2. Set the ideal time for the ICMP message numbered X (i<x≤ j) to return to the active measurement end ^/Τ χ ' = Τ ί + (χ-ι)χΛΤ;
3、 主动测量端选取 7; (编号为 X的 ICMP报文返回到主动测量端的时 间) 与 7;'最接近的 ICMP报文、 即返回到主动测量端的时间最接近理想时 间的 ICMP报文, 作为依据来计算被动测量端接收该组混合报文的速率。 3. The active measurement end selects 7; (the time when the ICMP packet numbered X returns to the active measurement end) and 7; the closest ICMP message, that is, the ICMP message that returns to the active measurement end closest to the ideal time, As a basis, the rate at which the passive measurement end receives the set of mixed messages is calculated.
当然, ICMP 4艮文的样本过滤还可以采用现有技术中其他的样本过滤算 法。  Of course, the sample filtering of ICMP 4 can also use other sample filtering algorithms in the prior art.
关于被动测量端接收一组混合报文的速率的计算: 假设该组混合报文 中 ICMP报文的长度为 //c«p、 UDP报文的长度为 单位为 bit, 则被动 测量端接收该组混合报文的速率为: Vn = ^-^xlUDP + (x-i + \)xlICMP ,其中 τ -τ 为一组混合报文的一个发送间隔中 UDP报文的个数。 The calculation of the rate at which the passive measurement terminal receives a set of mixed packets: If the length of the ICMP packet in the mixed packet is //c«p and the length of the UDP packet is in bits, the passive measurement terminal receives the The rate of group mixed messages is: Vn = ^-^xl UDP + (xi + \)xl ICMP , where τ -τ The number of UDP packets in the interval for sending a group of mixed packets.
为了实现上述带宽测量方法, 本发明还提供了一种测量装置, 如图 5 所示, 包括: 主动测量端 10和被动测量端 20, 其中,  In order to implement the foregoing bandwidth measurement method, the present invention further provides a measurement device, as shown in FIG. 5, including: an active measurement terminal 10 and a passive measurement terminal 20, wherein
主动测量端 10, 用于发送 ICMP ^艮文给被动测量端 20 , 并统计到被动 测量端 20的网络正常 ICMP报文丟包率; 还用于发送混合报文给被动测量 端 20, 并统计到被动测量端 20的混合报文的 ICMP报文平均丟包率; 还用 的丟包率阈值时, 统计被动测量端 20接收混合报文的平均速率, 作为主动 测量端 10到被动测量端 20的可用带宽;  The active measurement terminal 10 is configured to send an ICMP message to the passive measurement terminal 20, and collect statistics on the normal ICMP packet loss rate of the passive measurement terminal 20; and also send a mixed message to the passive measurement terminal 20, and collect statistics. The average packet loss rate of the ICMP packet of the mixed packet to the passive measurement terminal 20; when the packet loss rate threshold is used, the average rate at which the passive measurement terminal 20 receives the mixed packet is used as the active measurement terminal 10 to the passive measurement terminal 20 Available bandwidth;
被动测量端 20, 用于接收 ICMP报文和混合报文。 还用于在接收到主 动测量端 10发送的 ICMP报文时, 返回该 ICMP报文给主动测量端 10; 在 接收到主动测量端 10发送的混合报文时, 将该混合报文中的 UDP报文丟 弃, 并将混合报文中的 ICMP报文返回给主动测量端 10。  The passive measurement terminal 20 is configured to receive ICMP packets and mixed packets. When the ICMP message sent by the active measurement terminal 10 is received, the ICMP message is returned to the active measurement terminal 10; when the hybrid message sent by the active measurement terminal 10 is received, the UDP message in the mixed message is received. The packet is discarded, and the ICMP packet in the mixed packet is returned to the active measuring terminal 10.
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。  The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Claims

权利要求书 Claim
1、一种利用混合报文测量可用带宽的方法,其特征在于,该方法包括: 主动测量端发送网际控制报文协议(ICMP )报文给被动测量端, 并统 计到被动测量端的网络正常 ICMP报文丟包率;  A method for measuring available bandwidth by using a mixed packet, the method comprising: the active measuring end sends an Internet Control Message Protocol (ICMP) message to the passive measuring end, and the network of the passive measuring end is normal ICMP. Packet loss rate;
主动测量端发送混合报文给所述被动测量端, 并统计到被动测量端的 所述混合报文的 ICMP报文平均丟包率;其中,所述混合报文包括所述 ICMP 报文和用户数据报文协议 ( UDP )报文; 率之差大于预设的丟包率阈值时, 统计所述被动测量端接收混合报文的平 均速率, 作为主动测量端到被动测量端的可用带宽。  The active measurement end sends the mixed packet to the passive measurement end, and collects the average packet loss rate of the ICMP packet of the hybrid packet at the passive measurement end. The hybrid packet includes the ICMP packet and the user data. Packet protocol (UDP) packet; when the rate difference is greater than the preset packet loss rate threshold, the average rate at which the passive measurement end receives the mixed packet is counted as the available bandwidth from the active measurement end to the passive measurement end.
2、 根据权利要求 1所述利用混合报文测量可用带宽的方法, 其特征在 于, 主动测量端发送 ICMP 文给被动测量端时, 该方法进一步包括: 所 述主动测量端以预设的初始发送速率发送 ICMP报文给被动测量端, 所述 被动测量端接收到所述 ICMP报文后,将所述 ICMP报文返回给所述主动测 量端;  The method for measuring the available bandwidth by using the hybrid packet according to claim 1, wherein when the active measuring end sends the ICMP file to the passive measuring end, the method further comprises: the active measuring end transmitting with a preset initial The rate is sent to the passive measurement end, and the passive measurement end returns the ICMP message to the active measurement end after receiving the ICMP message;
相应的, 统计所述网络正常 ICMP 4艮文丟包率, 具体为: 所述主动测 量端依据自身发送的 ICMP报文、和被动测量端返回的 ICMP报文,统计所 述网络正常 ICMP报文丟包率。  Correspondingly, the normal ICMP packet loss rate of the network is counted, and the following is: the active measurement end collects the normal ICMP packet of the network according to the ICMP packet sent by the active measurement end and the ICMP packet sent by the passive measurement end. Packet loss rate.
3、 根据权利要求 2所述利用混合报文测量可用带宽的方法, 其特征在 于, 所述主动测量端发送混合报文给所述被动测量端之前, 该方法还包括: 所述主动测量端确定所述混合报文的发送速率;  The method for measuring the available bandwidth by using the mixed message according to claim 2, wherein before the active measuring end sends the mixed message to the passive measuring end, the method further includes: determining, by the active measuring end, The sending rate of the mixed message;
相应的, 确定所述混合报文的发送速率, 具体为:  Correspondingly, determining a sending rate of the mixed packet is specifically:
当所述主动测量端第一次向所述被动测量端发送所述混合报文时, 所 述混合报文的第一次发送速率为所述预设的初始发送速率、 或者为所述主 动测量端到被动测量端之间瓶颈链路的链路带宽。 When the active measurement end sends the hybrid message to the passive measurement end for the first time, the first transmission rate of the hybrid message is the preset initial transmission rate, or is the active measurement. The link bandwidth of the bottleneck link between the end and the passive measurement end.
4、 根据权利要求 1所述利用混合报文测量可用带宽的方法, 其特征在 于, 所述主动测量端发送的混合报文为 n组混合报文, 其中 n为大于等于 1 的整数。 The method for measuring the available bandwidth by using the mixed message according to claim 1, wherein the mixed message sent by the active measuring end is n sets of mixed messages, where n is an integer greater than or equal to 1.
5、 根据权利要求 4所述利用混合报文测量可用带宽的方法, 其特征在 于, 所述主动测量端发送混合报文给所述被动测量端时, 该方法还包括: 所述主动测量端将所述 UDP报文的目的端口设置为错误的目的端口;  The method for measuring the available bandwidth by using the mixed message according to claim 4, wherein when the active measuring end sends the mixed message to the passive measuring end, the method further includes: the active measuring end The destination port of the UDP packet is set to the wrong destination port.
相应的, 所述被动测量端接收到所述混合报文后, 该方法还包括: 所 述被动测量端依据所述错误的目的端口将所述 UDP报文丟弃, 并将所述混 合报文中的 ICMP报文返回给所述主动测量端。  Correspondingly, after the passive measurement end receives the mixed message, the method further includes: the passive measurement end discards the UDP packet according to the wrong destination port, and the hybrid message is The ICMP message in is returned to the active measurement terminal.
6、 根据权利要求 5所述利用混合报文测量可用带宽的方法, 其特征在 于, 所述主动测量端统计所述 ICMP ^艮文平均丟包率, 具体为:  The method for measuring the available bandwidth by using the mixed packet according to claim 5, wherein the active measuring end collects the average packet loss rate of the ICMP^艮, specifically:
所述主动测量端依据自身发送的 n组混合报文、 和被动测量端返回的 ICMP报文, 统计所述 n组中每组混合报文的 ICMP报文丟包率, 并对所述 n组混合报文的 ICMP报文丟包率进行平均,得到所述 ICMP报文平均丟包 率。  The active measurement end collects the packet loss rate of the ICMP packet of each group of the mixed packets in the n groups according to the n sets of the mixed packets and the ICMP packets sent by the passive measurement end, and compares the packet loss rate of the ICMP packets in the n groups. The average packet loss rate of the ICMP packets is averaged, and the average packet loss rate of the ICMP packets is obtained.
7、根据权利要求 1至 6任一项所述利用混合报文测量可用带宽的方法, 其特征在于, 所述主动测量端判定所述 ICMP报文平均丟包率与所述网络 正常 ICMP报文丟包率之差小于等于预设的丟包率阈值时, 该方法还包括: 所述主动测量端重新确定所述混合报文的发送速率, 具体为: 以所述混合 报文的第一次发送速率为基础, 采用线性提高、 或者指数倍提高的方式增 加所述发送速率。  The method for measuring the available bandwidth by using the hybrid packet according to any one of claims 1 to 6, wherein the active measurement end determines the average packet loss rate of the ICMP packet and the normal ICMP packet of the network. When the difference between the packet loss rate is less than or equal to the preset packet loss rate threshold, the method further includes: the active measurement end re-determining the sending rate of the mixed packet, specifically: the first time of the mixed packet Based on the transmission rate, the transmission rate is increased by linearly increasing, or exponentially increasing.
8、 根据权利要求 7所述利用混合报文测量可用带宽的方法, 其特征在 于, 所述主动测量端统计所述被动测量端接收所述混合报文的平均速率, 具体为: 所述主动测量端依据自身发送的 n组混合 ^艮文、 和被动测量端返 回的 ICMP报文, 统计所述被动测量端接收所述 n组中每组混合报文的速 率, 并对所述被动测量端接收所述 n组混合报文的速率进行平均, 得到所 述被动测量端接收所述混合报文的平均速率。 The method for measuring the available bandwidth by using the mixed packet according to claim 7, wherein the active measuring end collects an average rate at which the passive measuring end receives the mixed packet, specifically: the active measuring The terminal collects the speed of each group of mixed packets in the n groups according to the n sets of the mixed packets sent by the self and the ICMP packets returned by the passive measuring end. Rate, and average the rate at which the passive measurement end receives the n sets of mixed packets, and obtain an average rate at which the passive measurement end receives the mixed message.
9、一种利用混合报文测量可用带宽的装置,其特征在于,该装置包括: 主动测量端和被动测量端, 其中,  A device for measuring an available bandwidth by using a mixed message, the device comprising: an active measuring end and a passive measuring end, wherein
所述主动测量端, 用于发送 ICMP报文给所述被动测量端, 并统计到 所述被动测量端的网络正常 ICMP报文丟包率; 还用于发送混合报文给所 述被动测量端, 并统计到所述被动测量端的所述混合报文的 ICMP报文平 丟包率之差大于预设的丟包率阈值时, 统计所述被动测量端接收所述混合 报文的平均速率, 作为主动测量端到被动测量端的可用带宽;  The active measurement end is configured to send an ICMP message to the passive measurement end, and collect a packet loss rate of the normal ICMP packet on the passive measurement end; and send the mixed message to the passive measurement end. And counting, when the difference between the flat packet loss rate of the mixed packet of the passive measurement end is greater than a preset packet loss rate threshold, counting the average rate at which the passive measurement end receives the mixed packet, as The available bandwidth from the active measurement end to the passive measurement end;
所述被动测量端, 用于接收所述 ICMP报文和所述混合报文。  The passive measurement end is configured to receive the ICMP packet and the hybrid packet.
10、 根据权利要求 9所述利用混合报文测量可用带宽的装置, 其特征 在于, 所述被动测量端还用于, 在接收到所述主动测量端发送的 ICMP报 文时, 返回所述 ICMP报文给主动测量端; 在接收到所述主动测量端发送 的混合报文时, 将所述混合报文中的 UDP 报文丟弃, 并将混合报文中的 ICMP 艮文返回给主动测量端。  The apparatus for measuring the available bandwidth by using the mixed message according to claim 9, wherein the passive measuring end is further configured to: when receiving the ICMP message sent by the active measuring end, returning the ICMP The packet is sent to the active measurement end; when receiving the mixed message sent by the active measurement end, the UDP packet in the mixed message is discarded, and the ICMP message in the mixed message is returned to the active measurement. end.
PCT/CN2010/071120 2009-07-22 2010-03-18 Method and device for measuring available bandwidth using mixed messages WO2011009304A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200910089747A CN101964727B (en) 2009-07-22 2009-07-22 Method and device for measuring available bandwidth by using mixed messages
CN200910089747.8 2009-07-22

Publications (1)

Publication Number Publication Date
WO2011009304A1 true WO2011009304A1 (en) 2011-01-27

Family

ID=43498737

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/071120 WO2011009304A1 (en) 2009-07-22 2010-03-18 Method and device for measuring available bandwidth using mixed messages

Country Status (2)

Country Link
CN (1) CN101964727B (en)
WO (1) WO2011009304A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106603344A (en) * 2017-01-20 2017-04-26 烽火通信科技股份有限公司 Household gateway bandwidth monitoring system and method

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102075443B (en) * 2011-02-28 2012-11-21 电子科技大学 Active detection-based host IP flow estimation method
CN102594629A (en) * 2012-03-29 2012-07-18 迈普通信技术股份有限公司 Method and system for measuring subscriber line bandwidth
CN103905263B (en) * 2012-12-26 2017-06-20 中国移动通信集团公司 A kind of service bandwidth method of testing and device
TWI470974B (en) * 2013-01-10 2015-01-21 Univ Nat Taiwan Multimedia data rate allocation method and voice over ip data rate allocation method
CN103312632B (en) * 2013-07-01 2016-05-25 北京邮电大学 A kind of method of measuring available bandwidth
CN109995609A (en) * 2017-12-29 2019-07-09 杭州海康威视系统技术有限公司 Bandwidth determining method, device and storage medium
CN111416688B (en) * 2019-01-04 2023-06-16 阿里巴巴集团控股有限公司 Method, device, communication equipment and system for controlling message retransmission
CN112702238B (en) * 2020-12-24 2022-05-27 北京合众方达科技有限公司 Message retransmission method for severe network environment
CN115484209B (en) * 2022-09-23 2024-04-02 绿盟科技集团股份有限公司 Network traffic playback method and device, medium and electronic equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1400774A (en) * 2002-08-22 2003-03-05 陈鸣 Interconnection network end-to-end performance monitoring method and its system
CN101035037A (en) * 2007-04-17 2007-09-12 华为技术有限公司 Method, system and related device for detecting the network communication quality
CN101394355A (en) * 2008-10-29 2009-03-25 北京大学 Routing link detection method and device for mobile Ad hoc network

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1359697A1 (en) * 2002-04-30 2003-11-05 Psytechnics Ltd Method and apparatus for transmission error characterisation

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1400774A (en) * 2002-08-22 2003-03-05 陈鸣 Interconnection network end-to-end performance monitoring method and its system
CN101035037A (en) * 2007-04-17 2007-09-12 华为技术有限公司 Method, system and related device for detecting the network communication quality
CN101394355A (en) * 2008-10-29 2009-03-25 北京大学 Routing link detection method and device for mobile Ad hoc network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
HA XINTIAN ET AL., AVAILABLE BANDWIDTH MEASUREMENT WITH ICMP IN IPV6 ETWORKS MICROCOMPUTER INFORMATION, no. 1-3, 31 January 2008 (2008-01-31), pages 122 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106603344A (en) * 2017-01-20 2017-04-26 烽火通信科技股份有限公司 Household gateway bandwidth monitoring system and method

Also Published As

Publication number Publication date
CN101964727B (en) 2012-09-05
CN101964727A (en) 2011-02-02

Similar Documents

Publication Publication Date Title
WO2011009304A1 (en) Method and device for measuring available bandwidth using mixed messages
Zhang et al. Novel quick start (QS) method for optimization of TCP
Widmer et al. Extending equation-based congestion control to multicast applications
US8116225B2 (en) Method and apparatus for estimating channel bandwidth
US7675856B2 (en) Bandwidth estimation in broadband access networks
JP4348124B2 (en) Method and communication device for estimating QoS
Pokhrel et al. Improving multipath TCP performance over WiFi and cellular networks: An analytical approach
RU2677373C1 (en) Fractal telecommunication traffic transmission quality increasing method
EP3295612B1 (en) Uplink performance management
WO2012167685A1 (en) Method and device for monitoring packet-loss rate based on fpga
WO2011006372A1 (en) Method and system for end to end available bandwidth measurement
WO2013172996A1 (en) Link microbenchmarking with idle link correction
Shin et al. Loss recovery scheme for TCP using MAC MIB over wireless access networks
JP2012005087A (en) Data transmission system and method
Schlembach et al. Design and implementation of scalable admission control
Lim et al. Robust end-to-end loss differentiation scheme for transport control protocol over wired/wireless networks
Shorten et al. On queue provisioning, network efficiency and the transmission control protocol
Yuan et al. Design and implementation of scalable edge-based admission control
KR100870182B1 (en) Traffic estimation system and method for flow per routing pair in the Layer2/3
Seferoglu et al. Congestion state-based dynamic FEC algorithm for media friendly transport layer
Dalton et al. A study on high speed TCP protocols
Chen et al. Normalized queueing delay: Congestion control jointly utilizing delay and marking
Ozturk et al. Workload-dependent queuing model of an AQM-controlled wireless router with TCP traffic and its application to PER-based link adaptation
WO2023280004A1 (en) Network configuration method, device and system
Damjanovic et al. Extending the tcp steady-state throughput equation for parallel tcp flows

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: 10801864

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: 10801864

Country of ref document: EP

Kind code of ref document: A1