WO2010130102A1 - 网络性能测量的方法、装置和系统 - Google Patents

网络性能测量的方法、装置和系统 Download PDF

Info

Publication number
WO2010130102A1
WO2010130102A1 PCT/CN2009/071811 CN2009071811W WO2010130102A1 WO 2010130102 A1 WO2010130102 A1 WO 2010130102A1 CN 2009071811 W CN2009071811 W CN 2009071811W WO 2010130102 A1 WO2010130102 A1 WO 2010130102A1
Authority
WO
WIPO (PCT)
Prior art keywords
performance measurement
packet
performance
parameter
measurement parameter
Prior art date
Application number
PCT/CN2009/071811
Other languages
English (en)
French (fr)
Inventor
宋建民
马鹏
陈大鹏
姚政
潘雷
薛晶
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN2009801552232A priority Critical patent/CN102301651A/zh
Priority to BRPI0924567A priority patent/BRPI0924567A2/pt
Priority to EP09844514A priority patent/EP2432160A1/en
Priority to PCT/CN2009/071811 priority patent/WO2010130102A1/zh
Publication of WO2010130102A1 publication Critical patent/WO2010130102A1/zh
Priority to US13/296,782 priority patent/US20120057497A1/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
    • H04L43/0852Delays
    • H04L43/0858One way delays
    • 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
    • H04L43/0835One way packet loss

Definitions

  • the present invention relates to the field of communications, and in particular, to a method, an apparatus, and a system for measuring network performance.
  • BACKGROUND OF THE INVENTION The development of services continues to pose more challenges to the network. For example, in a video multicast service, if a data stream is lost during transmission, and the delay is too long, the video image may be mosaic and not smooth. Such failure phenomena directly affect the user experience. Therefore, how to measure the performance of packet loss and delay of the network and locate the faults that occur is an urgent problem to be solved.
  • an LMM Lessing Measurement Message
  • a CCM Continuousity Check Message
  • a 1 DM One-way Delay Measurement
  • a DM One-way Delay Measurement
  • DMM/DMR Delay Measurement Message/Delay Measurement Reply
  • NE Network Element 1 and NE2 are both ends of a maintenance domain.
  • a 1 DM packet can be sent from NE1 to NE2.
  • the 1 DM packet carries the NE1 timestamp t1.
  • NE2 obtains the timestamp t1; and according to the local (ie, NE2) timestamp t2, the delay between NE1 and NE2 is (t2 - t1). .
  • the above method is a performance measurement method for a layer 2 network, and a stream-based network performance measurement method cannot be implemented.
  • the embodiments of the present invention provide the following technical solutions, which can solve the problem that the performance of the flow-based network cannot be detected in the prior art.
  • a method for measuring network performance information includes:
  • the receiving end receives at least one performance measurement message, where the performance measurement message carries the performance measurement parameter of the sending end of the performance measurement message, and the performance measurement parameter of the sending end corresponds to the first data stream;
  • the performance measurement packet is sent by the sender based on the first data stream.
  • a method of network performance measurement including:
  • the receiving end receives at least one performance measurement packet, and the performance measurement packet carries the performance measurement parameter of the sending end of the performance measurement packet;
  • the network performance between the sender and the receiver is measured according to the performance measurement parameters of the sender and the performance measurement parameters of the receiver.
  • a device for measuring network performance including:
  • a receiving module configured to receive at least one performance measurement message, where the performance measurement message carries a performance measurement parameter of the transmitting end, and the performance measurement parameter of the sending end corresponds to the first data stream, where the performance measurement message is that the sending end is based on the The first data stream is sent;
  • a management module configured to obtain a performance measurement parameter of the transmitting end, determine a performance measurement parameter of the device, and the performance measurement parameter of the device corresponds to the first data stream;
  • a measurement module configured to measure network performance between the transmitting end and the device according to the performance measurement parameter of the transmitting end and the performance measurement parameter of the device, where the network performance corresponds to the first data stream.
  • a system for measuring network performance comprising:
  • a first device configured to send a performance measurement packet based on the first data stream, where the performance measurement packet carries a performance measurement parameter of the device, the performance measurement parameter of the first device is corresponding to the first data flow;
  • the second device is configured to receive the performance measurement message, and obtain the performance measurement parameter of the first device carried in the performance measurement message, and determine a performance measurement parameter of the second device; a performance measurement parameter of the second device corresponding to the first data stream;
  • the network performance between the first device and the second device is measured according to the performance measurement parameter of the first device and the performance measurement parameter of the second device.
  • FIG. 1 is a schematic diagram showing an example of performing delay detection in the prior art
  • FIG. 2 is a schematic view of a specific embodiment of the present invention.
  • Figure 3 is a schematic view of still another embodiment of the present invention.
  • FIG. 5 is a flowchart of still another method for performance measurement according to an embodiment of the present invention.
  • FIG. 6 is a flowchart of still another method for performance measurement according to an embodiment of the present invention.
  • Figure 7 is a schematic view of still another embodiment of the present invention.
  • FIG. 8 is a schematic diagram of a device for performance measurement according to an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of still another apparatus for performance measurement according to an embodiment of the present invention.
  • FIG. 10 is a schematic diagram of still another apparatus for performance measurement according to an embodiment of the present invention.
  • FIG. 11 is a schematic diagram of still another apparatus for performance measurement according to an embodiment of the present invention.
  • FIG. 12 is a schematic diagram of a system for performance measurement according to an embodiment of the present invention.
  • Figure 13 is a schematic diagram of the format of an IPv4 packet
  • FIG. 14 is a schematic diagram of a fixed format of a packet loss measurement packet according to an embodiment of the present invention
  • FIG. 15 is a schematic diagram of a TLV format of a packet loss measurement packet according to an embodiment of the present invention
  • FIG. 17 is a schematic diagram of a TLV format of a delay measurement message according to an embodiment of the present invention
  • FIG. 18 is a flowchart of still another method for performance measurement according to an embodiment of the present invention.
  • NE1 and NE2 are the two ends of a maintenance domain.
  • the maintenance domain carries three data streams, namely, video stream 1, video stream 2, and voice stream 3.
  • the video stream 1 is forwarded by the NE1 to the NE2 through the forwarding path A
  • the video stream 2 is forwarded by the NE1 to the NE2 through the forwarding path B
  • the voice stream 3 is forwarded by the NE1 to the NE2 through the forwarding path C.
  • the NE1 can send the performance measurement packet to the NE2 according to a certain period.
  • the period in which the performance measurement packet is sent may be set.
  • the NE1 counts the data packet of the sent video stream 1.
  • the NE1 generates the first performance measurement packet, the first performance.
  • the measurement message carries the count value a of the data packet of the video stream 1 sent by the NE1. Then, NE1 sends the first performance measurement message based on video stream 1.
  • NE1 continues to count the data packet of the transmitted video stream 1, and after the period ends, generates a second performance measurement packet, and sends the second performance measurement packet based on the video stream 1,
  • the second performance measurement packet also carries the count value b of the data packet of the video stream 1 sent by the NE1.
  • NE1 can also send a performance measurement packet to NE2.
  • the number of packets sent by the packet can be set, and NE1 counts the data packet of the transmitted video stream 1.
  • the first performance measurement packet carries the count value a of the data packet of the video stream 1 sent by the NE1.
  • NE1 sends the first performance measurement message based on video stream 1.
  • NE1 continues to count the data packets of the transmitted video stream 1, and the number of data packets of the transmitted video stream 1 is equal to the previous one.
  • the NE1 When the number of packets sent by the packet is set, the NE1 generates a second performance measurement packet, and sends the second performance measurement packet based on the video stream 1, and the second performance measurement packet carries the data of the video stream 1 sent by the NE1.
  • the count value b of the message When the number of packets sent by the packet is set, the NE1 generates a second performance measurement packet, and sends the second performance measurement packet based on the video stream 1, and the second performance measurement packet carries the data of the video stream 1 sent by the NE1.
  • the count value b of the message The count value b of the message.
  • NE1 can send performance measurement packets to NE2 according to actual needs. For example, NE1 counts the data packet of the sent video stream 1 and generates a first performance measurement packet, and the first performance measurement packet carries the data packet of the video stream 1 sent by the NE1. Count value a. Then, NE1 sends the first performance measurement message based on video stream 1. Then, the NE1 continues to count the data packets of the sent video stream 1. When the other signals are triggered again, the NE1 generates a second performance measurement packet, and sends the second performance measurement packet based on the video stream 1. The second performance measurement message carries the count value b of the data packet of the video stream 1 sent by the NE1.
  • the NE1 may use the data packet of the video stream 1 to obtain the foregoing performance measurement packet.
  • the NE1 can copy the data packet of the video stream 1 and then replace the information originally carried in the payload of the data packet with the count value a or b of the data packet of the video stream 1 sent by the NE1.
  • the first performance measurement message or the second performance measurement message are the same as the forwarding path and the forwarding rule of the video stream 1, so that the NE1 can be sent based on the video stream 1 A performance measurement message or a second performance measurement message.
  • a specific value may be carried in a certain field of the performance measurement message, so that NE2 can recognize that the message is a performance measurement message.
  • the NE2 also counts the data packet of the video stream 1 received by the local device.
  • the NE1 receives the first performance measurement packet sent by the NE1 based on the video stream 1, the NE2 determines the data of the video stream 1 received locally.
  • the packet count value a' is obtained, and according to the received first performance measurement packet, the data packet count value a of the video stream 1 sent by the NE1 carried in the packet is obtained.
  • the NE2 receives the second performance measurement packet sent by the NE1 based on the video stream 1, the NE2 determines the data packet count value b' of the locally received video stream 1, and obtains the report according to the received second performance measurement packet.
  • NE2 can obtain video stream 1 by NE1 according to the above performance measurement parameters a, b, a', and b'.
  • the number of lost packets forwarded to NE2 is [ ( ba ) - ( b'-a' ) ].
  • (b) is the number of packets sent by NE1
  • (b'-a') is the interval between the two received performance measurement packets and the message received by NE2.
  • the difference between the two is the number of lost packets of video stream 1 forwarded by NE1 to NE2.
  • NE1 and NE2 can use the local counter to count the data packets of the transmitted/received video stream 1.
  • the above scheme can also be used to measure the number of lost packets of video stream 2 and voice stream 3 in video stream 1 forwarded by NE1 to NE2.
  • the NE1 sends a performance measurement message based on the video stream 1.
  • the NE1 can send a performance measurement packet to the NE2 according to a certain period, or send a performance measurement packet to the NE2, or send a performance measurement packet to the NE2 according to actual needs.
  • the performance measurement message transmission period may be set, or the number of packets sent by the fixed packet may be set.
  • NE1 When one cycle ends, or the number of packets sent by NE1 is equal to the preset number of packet transmission packets, NE1 generates and The first performance measurement message is sent based on the video stream 1, and the first performance measurement message carries the local time t when the first performance measurement is sent by the NE1. For example, the first performance measurement message is generated and sent according to the video stream 1 under the trigger of the other signal. The first performance measurement message carries the local time t when the first performance measurement message is sent by the NE1.
  • the NE2 When receiving the first performance measurement packet sent by the NE1 based on the video stream 1, the NE2 determines the local time t′ when the first performance measurement packet is received, and obtains the packet according to the received first performance measurement packet. Time parameter carried in
  • NE2 can obtain the delay (t'-t) of video stream 1 forwarded from NE1 to NE2.
  • the local times of NE1 and NE2 are synchronized.
  • the local time of the NE1 and the NE2 may also have a certain time difference A t.
  • the two situations described above are to measure the packet loss and delay of the data flow by using the performance measurement message respectively.
  • the packet loss and delay of the data stream can be measured simultaneously.
  • the performance measurement message can carry the data packet count value of the video stream 1 sent by the NE1 and the local time parameter of the NE1.
  • the packet carries the data packet count value of the video stream 1 sent by the NE1 and the local time parameter of the NE1, if only the packet loss or the delay is measured, only the data of the video stream 1 sent by the NE1 needs to be read.
  • the message count value or the local time parameter of NE1 is fine.
  • the process is similar to the above two cases, and will not be described here.
  • the above performance measurement packets can also be used to measure network performance such as packet loss rate, delay jitter, and throughput.
  • the data transmission amount of video stream 1 can be limited to X. If no packet loss is detected, the data transmission amount of video stream 1 is appropriately increased to Y, and then detected. Whether there is a packet loss. This adjustment is repeated until the data transmission amount of the video stream 1 is increased to Z, and the packet loss is detected. At this time, Z is the measured network throughput.
  • the data stream may be an IP (Internet Protocol), an ETH (Ethernet, Ethernet), or an MPLS (Multi-Protocol Label Switch). Exchange), or a network such as VPLS (Virtual Private LAN Service).
  • IP Internet Protocol
  • ETH Ethernet, Ethernet
  • MPLS Multi-Protocol Label Switch
  • Exchange a network such as VPLS (Virtual Private LAN Service).
  • the data stream may be a unicast stream or a multicast stream as described above.
  • the data stream may be as described above.
  • Multimedia streams such as video stream 1, video stream 2, and voice stream 3 may, of course, be ordinary data streams.
  • the data stream that needs to be measured may be selected according to the information of the data stream, and then the performance measurement message is based on the data stream. send.
  • the data stream may be based on a binary group of data streams (ie: source IP address, destination IP address), or a triplet (ie: source IP address, source port, protocol number), or a quintuple (ie: source, destination)
  • the IP address, source, destination port, protocol number, or seven-tuple ie: interface index, source, destination IP address, source, destination port number, protocol number, and service type) select the stream to be measured.
  • other information such as stream identification, is also possible.
  • the selection can be done stream by stream and the performance measurement message is sent based on the data stream.
  • the performance measurement message may further carry a sequence number of the performance measurement message, and the two consecutive performance measurement messages in the sending sequence may be carried. Ordered. For example, in the foregoing embodiment, two performance measurement messages that are adjacent in the sending sequence may be set, and the serial number carried may be incremented by one. If the sequence number of the packet carried in the first performance measurement packet is
  • the sequence number of the packet carried in the second performance measurement packet should be 2. If the second performance measurement message is received by the ⁇ 2, and the sequence number carried in the second measurement message is 3, then ⁇ 2 knows that between the first performance measurement message and the second performance measurement message, A performance measurement message was not received.
  • ⁇ 1 sends a performance measurement message based on a specific data flow
  • ⁇ 2 receives a performance measurement message corresponding to the data flow, and can detect packet loss, delay, and the like of each specific flow, and implement flow-based performance. Detection.
  • ⁇ 1 and ⁇ 2 are respectively two ends of a maintenance domain, and in actual application, it is not limited to the two ends of the maintenance domain, and the intermediate network device on the packet forwarding path can also perform performance measurement. .
  • FIG. 3 shows still another specific implementation, which is capable of performing performance detection on each physical link of each hop and each hop device.
  • ⁇ 3 is an intermediate network device on the forwarding path.
  • the first performance measurement packet and the second performance measurement packet are sent in the first performance measurement packet, and the first performance measurement packet carries the data packet of the video stream 1 sent by the NE1.
  • the performance measurement packet carries the count value b1 of the data packet of the video stream 1 sent by the NE1.
  • the port 2 and the port 3 of the NE3 and the port 4 and the port 5 of the NE2 respectively determine the data of the video stream 1 received locally.
  • the packet count value you can get the number of packets lost in video stream 1, as shown in the following table: Table 1
  • port 4 detects a packet loss.
  • port 5 also detects the same amount of packet loss.
  • port 2 and port 3 do not detect packet loss. You can determine that the link between port 3 and port 4 is faulty, resulting in packet loss.
  • N E3 internal forwarding has a performance failure, resulting in packet loss.
  • the number of detected packet loss can be used to locate multiple packet loss faults that may occur, and details are not described herein.
  • the physical link of each hop and the delay performance of each hop device can also be detected.
  • the port 1 sends a first performance measurement message based on the video stream 1, where the first performance measurement message carries the local time parameter t1 when the first performance measurement message is sent.
  • the port 2 and the port 3 of the NE3 and the port 4 and the port 5 of the NE2 respectively determine the local time parameter when the first performance measurement packet is received locally. You can get the delay of video stream 1, as shown in the following table:
  • a delay fault that may occur between NE1 and NE2.
  • H does not detect the delay of port 4, and port 5 also detects the delay anomaly, and the delay anomaly is consistent, that is, the same value increases or decreases.
  • the delay of port 2 and port 3 is normal. It can be judged that the link between port 3 and port 4 is faulty, causing the delay to change abnormally.
  • ports 3, 4, and 5 detect a delay anomaly, and the delay anomalies are consistent, that is, an increase or decrease in the same value occurs.
  • the delay of port 2 is normal. It can be judged that the performance of NE3 internal forwarding is faulty, causing abnormal delay.
  • the detected delay can be used to locate multiple delay faults that may occur, and details are not described herein again.
  • the intermediate network device that receives the performance measurement packet does not write the performance measurement packet to the performance measurement packet, and does not affect the replication and forwarding of the performance measurement packet.
  • the performance measurement packet may be discarded.
  • NE1 sends a performance measurement message based on a specific data flow
  • NE2 and NE3 receive performance measurement packets corresponding to the data flow, which can not only implement flow-based performance detection, but also can perform each hop.
  • the physical link and each hop device perform performance detection to locate the faulty place in the network.
  • FIG. 4 is a flowchart of a method for performance measurement according to an embodiment of the present invention, including:
  • the first performance measurement packet carries a first performance measurement parameter, where the first performance measurement parameter corresponds to the first data flow;
  • the first performance measurement parameter is a performance measurement parameter of the sending end, and at least includes a packet count value of the first data stream sent by the sending end.
  • the first performance measurement packet is sent by the sending end of the first performance measurement message (hereinafter referred to as a transmitting end) based on the first data stream.
  • the foregoing first performance measurement packet is sent by the sending end, or sent by a fixed packet, or sent according to actual needs.
  • the first performance measurement packet further carries a sequence number of the performance measurement message.
  • the second performance measurement parameter is corresponding to the first data stream, and is a performance measurement parameter of the receiving end of the first performance measurement message (hereinafter referred to as a receiving end), including at least receiving The data packet count value of the first data stream received by the terminal.
  • the third performance measurement parameter is a performance measurement parameter of the sending end, and at least includes a data packet count value of the first data stream sent by the sending end.
  • the second performance measurement message is sent by the sending end based on the first data stream.
  • the foregoing second performance measurement packet is sent by the sending end periodically, or sent by a fixed packet, or sent according to actual needs.
  • the second performance measurement packet further carries a sequence number of the performance measurement packet.
  • the fourth performance measurement parameter is a performance measurement parameter of the receiving end, and at least includes a data packet count value of the first data stream received by the receiving end.
  • the first performance measurement parameter and the third performance measurement parameter are data packet count values, second performance measurement parameters, and fourth performance measurement of the first data stream sent by the sending end
  • the parameter is the data packet count value of the first data stream received by the receiving end
  • the number of lost packets forwarded by the transmitting end to the receiving end of the first data stream is: [(third performance measurement parameter - first performance measurement parameter) - (Fourth performance measurement parameter - second performance measurement parameter)].
  • the first performance measurement parameter and the third performance measurement parameter are And at least the data packet count value of the first data stream sent by the sending end
  • the second performance measurement parameter and the fourth performance measurement parameter include at least the data packet count value of the first data stream received by the receiving end
  • the first data The number of lost packets forwarded by the transmitting end to the receiving end is: [(The packet count value included in the third performance measurement parameter - the packet count value included in the first performance measurement parameter) - (Included in the fourth performance measurement parameter) The message count value - the message count value included in the second performance measurement parameter).
  • the intermediate network node that receives the first performance measurement packet and the second performance measurement packet does not perform the write operation on the performance measurement packet, and does not affect the performance measurement. Copy and forward the text.
  • the performance measurement packet may be discarded.
  • FIG. 5 is a flowchart of still another method for performance measurement according to an embodiment of the present invention, including:
  • the first performance measurement parameter is a performance measurement parameter of the sending end, and at least includes a local time parameter of the sending end.
  • the first performance measurement message is sent by the sending end based on the first data stream.
  • the foregoing first performance measurement packet is sent by the sending end, or sent by a fixed packet, or sent according to actual needs.
  • the first performance measurement packet further carries a sequence number of the performance measurement packet.
  • the second performance measurement parameter is corresponding to the first data stream, and is a performance measurement parameter of the receiving end, and includes at least a local time parameter of the receiving end.
  • the first performance measurement parameter is a local time parameter of the sending end
  • the second performance measurement parameter is a local time parameter of the receiving end
  • the delay is (second performance measurement parameter - first performance measurement parameter).
  • the first performance measurement parameter includes at least a local time parameter of the transmitting end
  • the second performance measurement parameter includes at least the local time parameter of the receiving end
  • the first data stream is forwarded by the sending end to The delay of the receiving end is (the local time parameter of the receiving end included in the second performance measurement parameter - the local time parameter of the transmitting end included in the first performance measurement parameter).
  • the intermediate network node that receives the first performance measurement packet does not perform the write operation on the performance measurement packet, and does not affect the copying and forwarding of the performance measurement packet.
  • the performance measurement packet may be discarded if the performance measurement packet is not required to be forwarded after receiving the performance measurement packet.
  • FIG. 6 is a flowchart of still another method for performance measurement according to an embodiment of the present invention, including:
  • the performance measurement message is sent by the sending end based on the first data stream.
  • the foregoing performance measurement packet is sent by the sending end, or sent by the fixed packet, or sent according to actual needs.
  • the foregoing performance parameter of the sending end when measuring the packet loss of the first data stream, includes at least a data packet count value of the first data stream sent by the sending end;
  • the first performance measurement parameter when the time delay of the first data stream is measured, includes at least a local time parameter of the transmitting end.
  • the receiving end performance measurement parameter corresponds to the first data stream.
  • the performance parameter of the receiving end includes at least the receiving end.
  • the intermediate network node that receives the performance measurement packet does not perform the write operation on the performance measurement packet, and does not affect the copying and forwarding of the performance measurement packet.
  • the performance measurement packet may be discarded if the performance measurement packet is not required to be forwarded after receiving the performance measurement packet.
  • the sending end sends the performance measurement packet based on the specific data flow
  • the receiving end receives the performance measurement packet corresponding to the data flow, and can detect the packet loss, the delay, and the like of each specific flow, and implement the flow based on the flow. Performance testing.
  • the first multicast stream is accessed by port 1 of NE1.
  • the first multicast stream is copied and forwarded to NE2 and NE3.
  • the first multicast is transmitted.
  • the stream is copied and forwarded to NE4 and NE5 (of course, it can be understood that for the sake of explanation, the complete forwarding path of the first multicast stream is not shown in FIG. 7).
  • the NE1 may send the packet periodically, or send the packet, or send the performance measurement packet according to the actual requirement.
  • the performance measurement packet carries the sender. In the embodiment, it is the performance measurement parameter of NE1).
  • the performance measurement parameter of the sending end may be a data packet count value of the first multicast stream sent by the sending end, or a local time parameter of the sending end, or a packet counting value and a local time parameter sent by the sending end, etc. In the examples, no specific limitation is made.
  • the transmitting end sends a performance measurement message based on the first multicast stream.
  • the performance measurement packet is equivalent to the first multicast stream.
  • the performance measurement packet is also copied and forwarded.
  • the receiving end (in the embodiment, NE2, NE3, NE4, NE5) can forward the first multicast stream to the local end by NE1. of Loss of packet, delay and other properties are tested.
  • the specific implementation is similar to the description in the embodiment shown in FIG. 2, and will not be described again here.
  • performance testing can also be performed on each physical link of each hop and each device.
  • the specific implementation is similar to that described in the embodiment shown in FIG. 3, and will not be described again here.
  • the NE1 sends a performance measurement packet based on the specific multicast stream, and the receiving end receives the performance measurement packet corresponding to the multicast stream, and can detect the packet loss and delay of each multicast stream.
  • Embodiments of the present invention also provide a device for performance measurement, as shown in Figure 8(a), the device includes:
  • the receiving module 800 is configured to receive at least one performance measurement packet, where the performance measurement packet carries a performance measurement parameter of the transmitting end, where the performance measurement parameter of the transmitting end is corresponding to the first data stream, where the performance measurement packet is based on the sending end The first data stream is sent;
  • the management module 810 is configured to obtain a performance measurement parameter of the transmitting end, and determine a performance measurement parameter of the receiving end, where the performance measurement parameter of the device corresponds to the first data stream.
  • the measuring module 820 is configured to measure network performance between the transmitting end and the receiving end according to the foregoing sending end performance measurement parameter and the receiving end performance measurement parameter, where the network performance corresponds to the first data stream.
  • the receiving end is the device for measuring the above performance.
  • the receiving module 800 may include:
  • the packet loss measurement packet receiving unit 801 is configured to receive at least one performance measurement packet, where the performance measurement packet carries a performance parameter of the sending end, where the performance measurement parameter includes at least a data packet count of the first data stream sent by the sending end.
  • the delay measurement packet receiving unit 802 is configured to receive at least one performance measurement packet, where the performance measurement packet carries a performance measurement parameter of the transmitting end, where the performance measurement parameter includes at least a local time parameter when the sending end sends the performance measurement packet .
  • the receiving module 800 includes the packet loss measurement message receiving unit 801 and the delay measurement message receiving unit 802 is shown in FIG.
  • the foregoing management module 810 may include:
  • the packet loss management unit 811 is configured to obtain a data packet count value of the first data stream sent by the sending end, which is carried in the performance measurement packet received by the packet loss measurement packet receiving unit 801, and determine the first received by the receiving end.
  • the delay management unit 812 is configured to obtain the local time parameter when the sending end of the performance measurement message sent by the performance measurement message received by the delay measurement message receiving unit 802, and the receiving end receives the performance measurement report.
  • the local time parameter of the text is configured to obtain the local time parameter when the sending end of the performance measurement message sent by the performance measurement message received by the delay measurement message receiving unit 802, and the receiving end receives the performance measurement report. The local time parameter of the text.
  • management module 810 includes both the packet loss management unit 811 and the delay management unit 812 is shown in FIG.
  • the foregoing measurement module 820 may include:
  • the packet loss measurement unit 821 is configured to: according to the data packet count value of the first data stream sent by the sender sent by the packet loss management unit 811, and the determined data packet count value of the first data stream received by the receiver, Measuring packet loss performance of the first data stream forwarded by the sending end to the receiving end;
  • the delay measurement unit 822 is configured to: according to the local time parameter when the sending end of the performance measurement message is sent by the sending end, and the determined local time parameter when the receiving end receives the performance measurement message, the measurement The delay performance of a data stream forwarded by the sender to the receiver.
  • the measurement module 820 includes both the packet loss measurement unit 821 and the delay measurement unit 822 is shown in FIG.
  • the performance measuring apparatus may further include:
  • the sending module 830 is configured to forward the performance measurement message received by the receiving module 800; or copy the performance measurement message received by the forwarding receiving module 800.
  • the sending end sends a performance measurement message based on a specific data flow, and the performance of the receiving end is measured.
  • the device receives the performance measurement packet corresponding to the data stream, and can detect packet loss, delay, and the like of each specific stream, and implement flow-based performance detection.
  • the embodiment of the present invention further provides a system for performance measurement.
  • the system includes: a first device, configured to send a performance measurement message based on the first data stream, where the performance measurement message carries a performance measurement parameter of the device, where the performance measurement parameter of the first device corresponds to the first data stream; the second device is configured to receive the performance measurement message, and obtain performance measurement of the first device carried in the performance measurement message a parameter, determining a performance measurement parameter of the second device; the performance measurement parameter of the second device is corresponding to the first data stream;
  • the network performance between the first device and the second device is measured according to the performance measurement parameter of the first device and the performance measurement parameter of the second device.
  • the second network device after receiving the foregoing performance measurement packet, forwards or copies the foregoing performance measurement.
  • the first device sends a performance measurement message based on the specific data flow
  • the second device receives the performance measurement corresponding to the data flow, and can detect the packet loss, delay, and the like of each specific flow. Implement stream-based performance detection.
  • the data stream may be IP, or MPLS, or a network such as VPLS. Therefore, in the embodiment of the present invention, the performance measurement packet may also be constructed according to an actual network protocol.
  • the following takes the performance measurement packet as an example of an IP packet, and describes the specific format of several performance measurement packets.
  • FIG. 13 shows the basic format of an IPv4 message (hereinafter referred to as an IP message).
  • the performance measurement parameter in the embodiment of the present invention may be carried in the data part of the IP packet.
  • the format of the performance measurement packet can be a fixed format or a TLV (Type, Length, Value, Type, Length, Value) format.
  • the fixed format can be in the format shown in Figure 14, where:
  • Version used to indicate different versions of the measurement methods used for performance measurement messages.
  • the operation type is used to indicate that the type of the performance measurement packet is a packet loss measurement packet.
  • the transmission cycle/packet number is used to indicate the transmission period of the preset performance measurement message, or to set the number of packets to be sent in advance, or to send the performance measurement message mode according to the actual needs, or according to actual needs.
  • the mode sends the performance measurement message mode as an option.
  • the TLV offset is used to indicate the position of the TLV in the message.
  • the TLV can be used to extend the function and is optional.
  • the serial number is used to indicate the order of packet loss measurement packets.
  • TLV end flag used to indicate the end of TLV, encountered this flag, indicating the end of all TLV, is optional.
  • the TLV format may be in the format shown in Figure 15, where:
  • TYPE used to indicate that the type of the packet is a packet loss measurement packet, which can be extended.
  • LENGTH used to identify the length of the data in the message, can be calculated in bytes.
  • Version used to indicate different versions of the measurement methods used for performance measurement messages.
  • the transmission period/packet number is used to indicate the transmission period of the preset performance measurement message, or to set the number of packets to be sent in advance, or to send the performance measurement message mode as required.
  • the serial number is used to indicate the order of packet loss measurement packets.
  • the format of the performance measurement message can also be a fixed format or a TLV (Type, Length, Value, Type, Length, Value) format.
  • the fixed format may be in the format shown in Figure 16, where:
  • Version used to indicate different versions of the measurement methods used for performance measurement messages.
  • the operation type is used to indicate that the type of the performance measurement packet is a delay measurement packet.
  • the transmission period/packet number is used to indicate the transmission period of the preset performance measurement message, or to set the number of packets to be sent in advance, or to send the performance measurement message mode as required.
  • the TLV offset is used to indicate the position of the TLV in the message.
  • the TLV is used for the extension of the function and is optional.
  • the serial number is used to indicate the order in which the transmission delay measurement message is sent.
  • the timestamp is sent to record the time when the delay measurement message is sent.
  • Receive time stamp used to record the time delay measurement. 4 The time when the text is received is optional.
  • TLV end flag used to indicate the end of TLV, encountered this flag, indicating the end of all TLV, is optional.
  • the TLV format may be in the format shown in Figure 17, where:
  • TYPE used to indicate that the type of the packet is a delay measurement packet, which can be extended.
  • LENGTH used to identify the length of the data in the message, can be calculated in bytes.
  • Version used to indicate different versions of the measurement methods used for performance measurement messages.
  • the transmission period/packet number is used to indicate the transmission period of the preset performance measurement message, or to set the number of packets to be sent in advance, or to send the performance measurement message mode as required.
  • the serial number is used to indicate the order in which the transmission delay measurement message is sent.
  • the timestamp is sent to record the time when the delay measurement message is sent.
  • Receive time stamp used to record the time delay measurement. 4 The time when the text is received is optional.
  • FIG. 14 to FIG. 17 are only schematic diagrams of several message formats.
  • the format of the performance measurement message is not limited thereto.
  • some specific values may be set in some fields of the message.
  • the protocol field of the performance measurement message can be set to 255.
  • the performance measurement packet can be identified according to the field, and the protocol field of the performance measurement packet is different from the protocol field of the data flow, and the packet of the performance measurement packet and the data flow does not affect the same forwarding path. And forwarding rules, etc.
  • it may be based on a five-tuple or a seven-tuple to distinguish different data streams.
  • a suitable field other than a five-tuple or a seven-tuple may be selected to carry a specific value to enable reception.
  • the end can identify the performance measurement message.
  • the specific configuration is determined by a specific network, and the embodiment of the present invention is not specifically limited.
  • the performance measurement packet may be a unicast packet or a multicast packet.
  • the foregoing performance measurement packet may also be an IPv6 packet.
  • the embodiment of the present invention further provides a method for performance measurement. As shown in FIG. 18, the method includes: B100, the receiving end receives at least one performance measurement message, where the performance measurement message carries the performance measurement message at the transmitting end. Performance measurement parameters;
  • B120 Obtain a performance measurement parameter of the sending end, and determine a performance measurement parameter of the receiving end; B130, measure network performance between the sending end and the receiving end according to the performance measurement parameter of the transmitting end and the performance measurement parameter of the receiving end.
  • the order of B110 and B120 is interchangeable.
  • the implementation of the embodiments of the invention is not affected.
  • the receiving end may copy the received performance measurement packet, and then obtain the performance measurement parameter of the sending end from the copied performance measurement packet.
  • the packet is forwarded again without affecting the forwarding of the performance measurement packet.
  • the performance measurement packet is sent by the sending end according to the first data stream.
  • the B110 may be specifically:
  • the receiving end forwards to the downstream network node based on the first data stream, or copies and forwards the performance measurement message.
  • the first data stream may be a unicast stream or a multicast stream.
  • the receiving end forwards the performance measurement message to the downstream network node based on the first data stream; when the first data stream is a multicast stream, the receiving end is the same
  • the performance measurement packet is forwarded to the downstream network node based on the foregoing first data flow.
  • the performance measurement packet is also copied and forwarded on the network node where the first data flow is copied.
  • the intermediate network node after receiving the performance measurement packet, performs measurement of the corresponding network performance, and forwards the performance measurement packet, so that the downstream network node of the intermediate network node can also receive the performance measurement packet.
  • Performance measurement packets which not only enable end-to-end network performance measurement, but also enable segment-to-segment network performance measurement, which further enables network fault location.
  • the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware.
  • the technical solution of the embodiment of the present invention may be embodied in the form of a software product, where the computer software product may be stored in a storage medium, such as R ⁇ M/RAM, a magnetic disk, an optical disk, etc., including a plurality of instructions.
  • a computer device, or server, or other network device is caused to perform the methods described in various embodiments of the present invention or in some portions of the embodiments.

Landscapes

  • Engineering & Computer Science (AREA)
  • Environmental & Geological Engineering (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Description

网络性能测量的方法、 装置和系统 技术领域 本发明涉及通信领域, 特别涉及一种网络性能测量的方法、 装置和系统。 背景技术 业务的发展不断给网络提出更多的挑战, 例如, 在视频组播业务中, 如果数 据流在传输过程中出现丟包、 时延超长等情况, 视频图像就会出现马赛克、 不 流畅等故障现象, 直接影响了用户的体验。 因此, 如何对网络的丟包、 时延等 性能进行测量、 对出现的故障进行定位, 成为迫切需要解决的问题。
现有技术中, 通常可以利用 LMM ( Loss Measurement Message, 损耗则 量报文)、 CCM( Continuity Check Message,连续性检测报文)、 1 DM( One-way Delay Measurement, 单向时延测量)、 以及 DMM/DMR ( Delay Measurement Message/ Delay Measurement Reply, 时延测量报文 /时延测量回复)等 4艮文, 分别对单端丟包、 双端丟包、 单程时延以及双程时延等性能进行检测。
以单程时延为例, 如图 1 所示, NE ( Network Element, 网元) 1 和 NE2 为一个维护域的两端。 当需要对 NE1和 NE2之间的时延进行测量时, 可以由 NE1 向 NE2发送 1 DM报文。 其中, 该 1 DM报文中携带 NE1 端时间戳 t1。 NE2收到该 1 DM报文后,获取上述时间戳 t1;并根据收到报文时本地(即 NE2 端) 时间戳 t2, 就可以得到 NE1和 NE2之间的时延为 (t2 - t1 )。
但是,在实现本发明的过程中,本发明的发明人发现现有技术至少存在以下 问题: 上述方法是面向二层网络的性能测量方法, 无法实现基于流的网络性能 测量方法。
发明内容 有鉴于此,本发明的实施例提供了以下技术方案, 能够解决现有技术中无法 实现对基于流的网络性能进行检测的问题。
一种网络性能测量报文的方法, 包括:
接收端接收至少一个性能测量报文,该性能测量报文携带该性能测量报文的 发送端的性能测量参数, 该发送端的性能测量参数与第一数据流对应;
获取该发送端的性能测量参数,确定接收端的性能测量参数,接收端的性能 测量参数与所述第一数据流对应;
根据发送端的性能测量参数和接收端的性能测量参数,测量发送端和所述接 收端之间的网络性能, 该网络性能与第一数据流对应;
其中, 性能测量报文是发送端基于第一数据流发送的。
一种网络性能测量的方法, 包括:
接收端接收至少一个性能测量报文,性能测量报文携带性能测量报文的发送 端的性能测量参数;
向接收端的下游网络节点转发该性能测量报文;
获取发送端的性能测量参数, 确定接收端的性能测量参数;
根据发送端的性能测量参数和接收端的性能测量参数,测量发送端和接收端 之间的网络性能。
一种网络性能测量的装置, 包括;
接收模块,用于接收至少一个性能测量报文,该性能测量报文携带发送端性 能测量参数, 发送端的性能测量参数与第一数据流对应, 该性能测量报文是所 述发送端基于所述第一数据流发送的;
管理模块, 用于获取发送端性能测量参数, 确定该装置的性能测量参数, 该 装置的性能测量参数与第一数据流对应;
测量模块,用于根据发送端的性能测量参数和装置的性能测量参数, 测量发 送端和装置之间的网络性能, 该网络性能与第一数据流对应。
一种网络性能测量的系统, 包括:
第一设备,用于基于第一数据流发送性能测量报文,性能测量报文中携带第 一设备的性能测量参数, 第一设备的性能测量参数与第一数据流对应; 第二设备,用于接收性能测量报文, 获取性能测量报文中携带的第一设备的 性能测量参数, 确定第二设备的性能测量参数; 第二设备的性能测量参数与所 述第一数据流对应;
根据第一设备的性能测量参数和第二设备的性能测量参数,测量第一设备和 第二设备之间的网络性能。
采用本发明的实施例提供的性能测量的方法、装置和系统, 能够检测到具体 每一个业务流、 每一跳的丟包、 时延等, 实现基于流的性能检测。 附图说明 图 1为现有技术中进行时延检测的一个实例示意图;
图 2为本发明的一个具体实施例示意图;
图 3为本发明的又一个具体实施例示意图;
图 4为本发明的实施例提供的一种性能测量的方法流程图;
图 5为本发明的实施例提供的又一种性能测量的方法流程图;
图 6为本发明的实施例提供的又一种性能测量的方法流程图;
图 7为本发明的又一个具体实施例示意图;
图 8为本发明的实施例提供的一种性能测量的装置示意图;
图 9为本发明的实施例提供的又一种性能测量的装置示意图;
图 10为本发明的实施例提供的又一种性能测量的装置示意图;
图 11为本发明的实施例提供的又一种性能测量的装置示意图;
图 12为本发明的实施例提供的一种性能测量的系统示意图;
图 13为 IPv4报文的格式示意图;
图 14为本发明的实施例中丟包测量报文的一种固定格式示意图; 图 15为本发明的实施例中丟包测量报文的一种 TLV格式示意图; 图 16为本发明的实施例中时延测量报文的一种固定格式示意图; 图 17为本发明的实施例中时延测量报文的一种 TLV格式示意图; 图 18为本发明的实施例提供的又一种性能测量的方法流程图。 具体实施方式 为使本发明的实施例的目的、技术方案及优点更加清楚明白,以下参照附图, 对本发明的实施例作进一步地详细说明。
为便于说明, 下面介绍本发明的一个具体实施例。 在图 2 所示的实例中,
NE1和 NE2分别为一个维护域的两端。
假设在本实施例中, 该维护域承载了三个数据流, 分别是视频流 1、 视频流 2以及语音流 3。 其中, 视频流 1 由 NE1通过转发路径 A转发至 NE2, 视频流 2由 NE1通过转发路径 B转发至 NE2,语音流 3由 NE1通过转发路径 C转发 至 NE2。 可以理解, 对于不同的转发路径, 丟包、 时延等性能也是不相同的。 下面分别对丟包和时延两种情况进行具体介绍。
1、 丟包
一方面, 当需要对视频流 1 的丟包进行检测时, NE1 可以根据一定的周期 向 NE2发送性能测量报文。 例如, 可以设定性能测量报文发送的周期, 在该周 期内, NE1对发送的视频流 1的数据报文进行计数, 当该周期结束时, NE1生 成第一性能测量报文,第一性能测量报文携带有 NE1发送的视频流 1的数据报 文的计数值 a。 然后, NE1基于视频流 1发送该第一性能测量报文。 类似的, 当下一周期开始时, NE1继续对发送的视频流 1的数据报文进行计数, 周期结 束后生成第二性能测量报文, 并基于视频流 1发送该第二性能测量报文, 第二 性能测量报文同样携带有 NE1发送的视频流 1的数据报文的计数值 b。
或者, NE1 也可以定包向 NE2发送性能测量报文。 例如, 可以设定定包 发送报文数, NE1对发送的视频流 1 的数据报文进行计数, 当 NE1发送的视 频流 1的数据报文数等于预先设定的定包发送报文数时, NE1生成第一性能测 量报文,第一性能测量报文携带有 NE1发送的视频流 1的数据报文的计数值 a。 然后, NE1基于视频流 1发送该第一性能测量报文。 然后, NE1继续对发送的 视频流 1的数据报文进行计数, 当发送的视频流 1的数据报文数再次等于预先 设定的定包发送报文数时, NE1生成第二性能测量报文, 并基于视频流 1发送 该第二性能测量报文,第二性能测量报文携带有 NE1发送的视频流 1的数据报 文的计数值 b。
或者, NE1也可以根据实际的需要向 NE2发送性能测量报文。 例如, NE1 对发送的视频流 1的数据报文进行计数, 在其他信号的触发下, 生成第一性能 测量报文,第一性能测量报文携带有 NE1发送的视频流 1的数据报文的计数值 a。 然后, NE1基于视频流 1发送该第一性能测量报文。 然后, NE1继续对发 送的视频流 1 的数据报文进行计数, 当再次收到其他信号的触发时, NE1 生 成第二性能测量报文, 并基于视频流 1发送该第二性能测量报文, 第二性能测 量报文携带有 NE1发送的视频流 1的数据报文的计数值 b。
可选的, 在本发明的实施例中, NE1 可以利用视频流 1 的数据报文, 得到 上述性能测量报文。 例如, NE1可以复制一个视频流 1的数据报文, 然后用上 述 NE1发送的视频流 1 的数据报文的计数值 a或 b承载替换掉该数据报文的 负载中原来携带的信息, 从而得到上述第一性能测量报文或者第二性能测量报 文。 这样, 上述第一性能测量报文或者第二性能测量报文的转发路径和转发规 则等, 与上述视频流 1 的转发路径和转发规则是相同的, 从而能够保证 NE1 基于视频流 1发送上述第一性能测量报文或者第二性能测量报文。
可选的, 在本发明的实施例中, 可以在性能测量报文的某一个字段, 携带特 定的数值, 以使得 NE2能够识别出该报文是性能测量报文。
另一方面, NE2也对本地接收到的视频流 1 的数据报文进行计数, 当 NE2 接收到 NE1基于视频流 1发送的第一性能测量报文时,确定本地接收到的视频 流 1的数据报文计数值 a', 并根据接收到的第一性能测量报文, 获取报文中携 带的 NE1发送的视频流 1的数据报文计数值 a。 当 NE2接收到 NE1基于视频 流 1发送的第二性能测量报文时, 确定本地接收到的视频流 1的数据报文计数 值 b', 并根据接收到的第二性能测量报文,获取报文中携带的 NE1发送的视频 流 1的数据报文计数值 b。
NE2就可以根据上述性能测量参数 a、 b、 a'、以及 b'得到视频流 1在由 NE1 转发至 NE2 的丟包数为 [ ( b-a ) - ( b'-a' ) ]。 其中, (b-a )为两次发送性能测 量报文的间隔内, NE1 发送的报文数, (b'-a' ) 为两次接收到性能测量报文的 间隔内, NE2接收到的报文数;二者之差,即为视频流 1在由 NE1转发至 NE2 的丟包数。
在实际应用中, NE1和 NE2可以利用本地计数器,对发送 /接收到的视频流 1的数据报文进行计数。
类似的, 采用上述方案也可以分别对视频流 2和语音流 3在由 NE1转发至 NE2的视频流 1的丟包数进行测量。
2、 时延
当需要对视频流 1 的时延进行检测时, 与测量丟包时的情况类似, NE1基 于视频流 1发送性能测量报文。 当然, NE1 可以根据一定的周期向 NE2发送 性能测量报文, 或者定包向 NE2发送性能测量报文, 或者根据实际的需要向 NE2发送性能测量报文。 例如, 可以设定性能测量报文发送周期、 或者设定定 包发送报文数, 当一个周期结束、或者 NE1发送的报文数等于预先设定的定包 发送报文数时, NE1生成并基于视频流 1发送第一性能测量报文, 该第一性能 测量 4艮文携带有 NE1发送该第一性能测量 ·^艮文时的本地时间 t。 又例如, 可以 其他信号的触发下, 生成并基于视频流 1发送第一性能测量报文, 第一性能测 量报文携带有 NE1发送该第一性能测量报文时的本地时间 t。
NE2接收到 NE1基于视频流 1发送的第一性能测量报文时, 确定接收到该 第一性能测量报文时的本地时间 t', 并根据接收到的第一性能测量报文, 获取 报文中携带的时间参数
NE2根据上述性能测量参数 以及 t', 就能够得到视频流 1在由 NE1转发 至 NE2的时延为 (t'-t )。
可选的, 在本发明的实施例中, NE1和 NE2的本地时间是同步的。
可选的, 在本发明的实施例中, NE1和 NE2的本地时间还可以存在一定的 时间差 A t, 此时, 在测量视频流由 NE1转发至 NE2的时延性能时, 需要考虑 到相应的时间差 A t。 当然,上述介绍的两种情况是分别利用性能测量报文对数据流的丟包和时延 进行测量。 在实际中, 可以对数据流的丟包和时延同时进行测量。 例如, 性能 测量报文中可以同时携带 NE1发送的视频流 1 的数据报文计数值和 NE1 的本 地时间参数。 当然,报文同时携带了 NE1发送的视频流 1的数据报文计数值和 NE1 的本地时间参数时, 如果只需要测量丟包或者时延, 则只需要读取 NE1 发送的视频流 1的数据报文计数值或者 NE1的本地时间参数就可以了。过程与 上述两种情况类似, 这里就不再赘述了。 在实际的应用中, 还可以利用上述性 能测量报文测量丟包率、 时延抖动、 吞吐量等网络性能。
以吞吐量为例, 在开始时, 可以将视频流 1的数据发送量限制在 X, 此时如 果没有检测到丟包, 则适当将视频流 1 的数据发送量增加至 Y, 此时再检测是 否出现丟包。 如此反复调整, 直到视频流 1的数据发送量增加至 Z时, 检测出 现了丟包, 此时, Z就是经过测量得到的网络吞吐量。
可选的,本发明的实施例中,承载上述数据流的可以是 IP( Internet Protocol , 网际协议), 或者是 ETH ( Ethernet, 以太网), 或者是 MPLS ( Multi-Protocol Label Switch , 多协议标签交换), 或者是 VPLS ( Virtual Private LAN Service, 虚拟私有局域网服务)等网络。
可选的, 在本发明的实施例中, 从通信模式上来说, 上述数据流可以是如上 介绍的单播流, 也可以是组播流; 从业务上来说, 上述数据流可以是如上介绍 的视频流 1、 视频流 2、 语音流 3等多媒体流, 当然, 也可以是普通的数据流。
可选的, 在本发明的实施例中, 如果只需要对特定的数据流进行性能检测, 可以根据数据流的信息, 选择需要进行测量的数据流, 然后将性能测量报文基 于该数据流中发送。 例如, 可以根据数据流的二元组(即: 源 IP地址、 目的 IP 地址), 或者三元组(即: 源 IP地址, 源端口、 协议号), 或者五元组(即: 源、 目的 IP地址, 源、 目的端口, 协议号), 或者七元组(即: 接口索引, 源、 目 的 IP地址, 源、 目的端口号, 协议号和服务类型)选择进行测量的流。 当然, 其他信息, 例如流标识也是可以的。
可选的, 在本发明的实施例中, 如果需要对所有的数据流进行性能检测, 也 可以逐流进行选择, 并将性能测量报文基于数据流发送。
可选的,在本发明的实施例中,性能测量报文中还可以携带性能测量报文的 序列号, 在发送顺序上相邻的两个性能测量 ·^艮文, 携带的序列号可以是顺序编 排的。 以上述实施例为例, 可以设定在发送顺序上相邻的两个性能测量报文, 携带的序列号可以是顺序加 1 的。 若第一性能测量报文中携带的报文序列号为
1 , 第二性能测量报文中携带的报文序列号应该为 2。 若 ΝΕ2接收到第二性能 测量报文后, 发现第二性测量报文中携带的序列号为 3, 则 ΝΕ2知道, 在第一 性能测量报文和第二性能测量报文之间, 还有一个性能测量报文没有收到。
在本实施例中, ΝΕ1基于具体的数据流发送性能测量报文, ΝΕ2接收与 数据流对应的性能测量报文, 能够检测到具体每一个流的丟包、 时延等, 实现 基于流的性能检测。
在图 2介绍的实施例中, ΝΕ1和 ΝΕ2分别是一个维护域的两端, 实际应用 中, 并不局限于维护域的两端, 报文转发路径上的中间网络设备, 也可以进行 性能测量。
如图 3所示, 图 3示出了又一种具体的实现, 能够对每一跳的物理链路和 每一跳设备进行性能检测。在图 3所示的实施例中, ΝΕ3是转发路径 Α上的一 个中间网络设备。 在端口 1先后基于视频流 1发送第一性能测量报文和第二性 能测量报文,其中,第一性能测量报文中携带 NE1发送的视频流 1的数据报文 的计数值 a1 , 第二性能测量报文中携带 NE1发送的视频流 1 的数据报文的计 数值 b1。 NE3的端口 2和端口 3、 NE2的端口 4和端口 5分别接收到与视频 流 1对应的第一性能测量报文和第二性能测量报文后, 分别确定本地接收到的 视频流 1的数据报文计数值, 就可以得到视频流 1的丟包数, 如下表所示: 表 1
本地接收到的视频流 1的数据报文计数值
端口 丢包数
第一性能测量报文 第二性能测量报文
2 a2 b2 ( b1 -a1 )-( b2-a2 )
3 a3 b3 ( b1 -a1 )-( b3-a3 ) 4 a4 b4 ( b1 -a1 )-( b4-a4 )
5 a5 b5 ( b1 -a1 )-( b5-a5 ) 进一步的, 根据上述丟包数, 还能够对 NE1至 NE2之间可能出现的丟包故 障进行定位。
例如, 假设端口 4检测到丟包, 同样, 端口 5也检测到相同数量的丟包。 但端口 2和端口 3没有检测到丟包, 可以判断端口 3〜端口 4之间的链路出现 性能故障, 导致丟包。
又例如, 端口 3、 4、 5检测到相同数量的丟包, 而端口 2没有检测到丟包, 可以判断 N E3内部转发出现性能故障, 导致丟包。
类似的, 还可以利用检测出的丟包数对可能出现的多处丟包故障进行定位, 此处不再赘述。
当然, 图 3所示的实例中,也可以对每一跳的物理链路和每一跳设备的时延 性能进行检测。 在端口 1基于视频流 1发送第一性能测量报文, 其中, 第一性 能测量报文中携带发送第一性能测量报文时的本地时间参数 t1。 NE3的端口 2 和端口 3、 NE2的端口 4和端口 5分别接收到与视频流 1对应的第一性能测量 报文后, 分别确定本地接收到第一性能测量报文时本地的时间参数, 就可以得 到视频流 1的时延, 如下表所示:
表 2
Figure imgf000011_0001
进一步的, 根据上述时延, 还能够对 NE1至 NE2之间可能出现的时延故障 进行定位。 例如, H没端口 4检测到时延异常, 同样, 端口 5也检测到时延异常, 且 时延异常值是一致的, 即均出现了相同值的增加或减少。 但端口 2、 端口 3的 时延是正常的, 可以判断到端口 3〜端口 4之间的链路出现了故障, 导致时延 异常变化。
又例如, 端口 3、 4、 5检测到时延异常, 且时延异常值是一致的, 即均出 现了相同值的增加或减少。 而端口 2的时延是正常的,可以判断 NE3内部转发 出现性能故障, 导致时延异常变化。
类似的, 还可以利用检测出的时延, 对可能出现的多处时延故障进行定位, 此处不再赘述。
可选的,在本发明的实施例中,接收到性能测量报文的中间网络设备对性能 测量报文不对性能测量报文进行写操作, 不影响性能测量报文的复制和转发。
可选的,若接收到性能测量报文的网络设备发现已经不需要继续转发该性能 测量报文时, 可以将该性能测量报文丟弃。
在本实施例中, NE1基于具体的数据流发送性能测量报文, NE2和 NE3接 收与该数据流对应的性能测量报文, 不仅能够实现基于流的性能检测, 而且还 能够对每一跳的物理链路和每一跳设备进行性能检测, 实现对网络中出现故障 的地方进行定位。
如图 4所示, 图 4为本发明的实施例提供的一种性能测量的方法流程图, 包括:
400,接收第一性能测量报文,该第一性能测量报文携带第一性能测量参数, 该第一性能测量参数与第一数据流对应;
可选的,本发明的实施例中, 上述第一性能测量参数为发送端的性能测量参 数, 至少包括发送端发送的第一数据流的报文计数值。
可选的,本发明的实施例中, 第一性能测量报文是第一性能测量报文的发送 端 (以下筒称为发送端 )基于上述第一数据流发送的。
可选的, 本发明的实施例中, 上述第一性能测量报文是发送端周期发送的, 或者定包发送的, 或者根据实际需要发送的。 可选的,本发明的实施例中, 上述第一性能测量报文还携带有性能测量报文 的序列号。
410, 获取上述第一性能测量参数, 确定第二性能测量参数;
可选的, 本发明的实施例中, 第二性能测量参数与上述第一数据流对应, 为 第一性能测量报文的接收端 (以下筒称为接收端) 的性能测量参数, 至少包括 接收端接收的第一数据流的数据报文计数值。
420,接收第二性能测量报文,该第二性能测量报文携带第三性能测量参数, 该第三性能测量参数与第一数据流对应;
可选的,本发明的实施例中, 上述第三性能测量参数为发送端的性能测量参 数, 至少包括发送端发送的第一数据流的数据报文计数值。
可选的,本发明的实施例中, 第二性能测量报文是发送端基于上述第一数据 流发送的。
可选的, 本发明的实施例中, 上述第二性能测量报文是发送端周期发送的, 或者定包发送的, 或者根据实际需要发送的。
可选的,本发明的实施例中, 上述第二性能测量报文还携带有性能测量报文 的序列号。
430, 获取上述第三性能测量参数, 确定第四性能测量参数;
可选的, 本发明的实施例中, 第四性能测量参数为接收端的性能测量参数, 至少包括接收端接收的第一数据流的数据报文计数值。
440,根据上述第一性能测量参数, 第二性能测量参数, 第三性能测量参数, 以及第四性能测量参数, 测量发送端和接收端之间的丟包性能。
可选的,本发明的实施例中, 当上述第一性能测量参数和第三性能测量参数 为发送端发送的第一数据流的数据报文计数值, 第二性能测量参数和第四性能 测量参数为接收端接收的第一数据流的数据报文计数值时, 上述第一数据流由 发送端转发至接收端的丟包数为: [ (第三性能测量参数 -第一性能测量参数) - (第四性能测量参数-第二性能测量参数) ]。
可选的,本发明的实施例中, 当上述第一性能测量参数和第三性能测量参数 至少包括发送端发送的第一数据流的数据报文计数值, 第二性能测量参数和第 四性能测量参数至少包括接收端接收的第一数据流的数据报文计数值时, 上述 第一数据流由发送端转发至接收端的丟包数为: [ (第三性能测量参数中包括的 报文计数值-第一性能测量参数中包括的报文计数值) - (第四性能测量参数中 包括的报文计数值-第二性能测量参数中包括的报文计数值 ) ]。
可选的,在本发明的实施例中,接收到上述第一性能测量报文和第二性能测 量报文的中间网络节点, 对上述性能测量报文不进行写操作, 不影响上述性能 测量 4艮文的复制和转发。
可选的,若接收到性能测量报文后,发现已经不需要继续转发该性能测量报 文时, 可以将该性能测量 ·^艮文丟弃。
如图 5所示, 图 5为本发明的实施例提供的又一种性能测量的方法流程图, 包括:
500,接收第一性能测量报文,该第一性能测量报文携带第一性能测量参数, 该第一性能测量参数与第一数据流对应;
可选的,本发明的实施例中, 上述第一性能测量参数为发送端的性能测量参 数, 至少包括发送端的本地时间参数。
可选的,本发明的实施例中, 第一性能测量报文是发送端基于上述第一数据 流发送的。
可选的, 本发明的实施例中, 上述第一性能测量报文是发送端周期发送的, 或者定包发送的, 或者根据实际需要发送的。
可选的,本发明的实施例中, 上述第一性能测量报文还携带有性能测量报文 的序列号。
510, 获取上述第一性能测量参数, 确定第二性能测量参数;
可选的, 本发明的实施例中, 第二性能测量参数与上述第一数据流对应, 为 接收端的性能测量参数, 至少包括接收端的本地时间参数。
520, 根据上述第一性能测量参数、 以及第二性能测量参数, 测量发送端和 接收端之间的时延性能。 可选的,本发明的实施例中, 当上述第一性能测量参数为发送端的本地时间 参数, 第二性能测量参数为接收端的本地时间参数时, 上述第一数据流由发送 端转发至接收端的时延为 (第二性能测量参数-第一性能测量参数)。
可选的,本发明的实施例中, 当上述第一性能测量参数至少包括发送端的本 地时间参数, 第二性能测量参数至少包括接收端的本地时间参数时, 上述第一 数据流由发送端转发至接收端的时延为 (第二性能测量参数中包括的接收端的 本地时间参数-第一性能测量参数中包括的发送端的本地时间参数)。
可选的,在本发明的实施例中,接收到上述第一性能测量报文的中间网络节 点, 对上述性能测量报文不进行写操作, 不影响上述性能测量报文的复制和转 发。
可选的,若接收到性能测量报文后,发现已经不需要继续转发该性能测量报 文时, 可以将该性能测量报文丟弃。
如图 6所示, 图 6为本发明的实施例提供的又一种性能测量的方法流程图, 包括:
600 , 接收至少一个性能测量报文, 该性能测量报文携带发送端的性能测量 参数, 该发送端的性能测量参数与第一数据流对应;
可选的,本发明的实施例中, 性能测量报文是发送端基于上述第一数据流发 送的。
可选的, 本发明的实施例中, 上述性能测量报文是发送端周期发送的, 或者 定包发送的, 或者根据实际需要发送的。
可选的, 本发明的实施例中, 当对上述第一数据流的丟包进行测量时, 上述 发送端性能测量参数至少包括发送端发送的第一数据流的数据报文计数值; 当 对上述第一数据流的时延进行测量时, 上述第一性能测量参数至少包括发送端 的本地时间参数。
610 , 获取发送端性能测量参数, 确定接收端性能测量参数;
可选的, 本发明的实施例中, 接收端性能测量参数与第一数据流对应。 当对 上述第一数据流的丟包进行测量时, 上述接收端性能测量参数至少包括接收端 接收的第一数据流的数据报文计数值;当对上述第一数据流的时延进行测量时, 上述接收端性能测量参数至少包括接收端的本地时间参数。
620, 根据上述发送端性能测量参数和接收端性能测量参数, 测量发送端和 接收端之间的网络性能。
可选的, 在本发明的实施例中, 接收到上述性能测量报文的中间网络节点, 对上述性能测量报文不进行写操作, 不影响上述性能测量报文的复制和转发。
可选的,若接收到性能测量报文后,发现已经不需要继续转发该性能测量报 文时, 可以将该性能测量报文丟弃。
在本实施例中,发送端基于具体的数据流发送性能测量报文,接收端接收与 数据流对应的性能测量报文, 能够检测到具体每一个流的丟包、 时延等, 实现 基于流的性能检测。
为了更好的理解本发明的实施例, 下面再介绍一个组播流的具体实施例。 如图 7所示, 第一组播流由 NE1的端口 1进入, 在 NE1 的端口 2, 对第一 组播流进行复制, 转发给 NE2和 NE3; 在 NE2的端口 4, 对第一组播流进行 复制, 转发给 NE4和 NE5 (当然, 可以理解, 为了使说明筒便, 图 7中并未 示出第一组播流的完整转发路径)。
为了对第一组播流的丟包、 时延等性能进行测量, NE1 可以周期发送、 或 者定包发送、 或者根据实际需要发送性能测量报文, 该性能测量报文中携带发 送端(本具体实施例中, 即为 NE1 )的性能测量参数。 该发送端的性能测量参 数可以是发送端发送的第一组播流的数据报文计数值, 或者是发送端的本地时 间参数,或者是发送端发送的报文计数值和本地时间参数等,本具体实施例中, 就不做具体的限定了。
与图 2所示的实施例中类似的, 发送端基于第一组播流发送性能测量报文。 如, 从转发路径、 转发规则方面来看, 性能测量报文是等同于第一组播流的, 在第一组播流发生复制的中间网络节点,性能测量报文也同样进行复制并转发。
接收端(本具体实施例中, 即为 NE2、 NE3、 NE4、 NE5 )接收到第一组播 流对应的性能测量报文后,就能够对第一组播流由 NE1转发至该接收端本地的 丟包、 时延等性能进行检测。 具体的实现与图 2所示的实施例中的描述类似, 此处就不再赘述了。
当然, 与图 3所示的具体实施例相类似的, 本具体实施例中, 也能够对每一 跳的物理链路和每一条设备进行性能检测。 具体的实现与图 3所示的实施例中 的描述类似, 此处也不再赘述了。
在本实施例中, NE1 基于具体的组播流发送性能测量报文, 接收端接收与 组播流对应的性能测量报文, 能够检测到具体每一个组播流的丟包、 时延等, 实现基于组播流的性能检测。 进一步的, 还能够对每一跳的物理链路和每一跳 设备进行性能检测, 实现对网络中出现故障的地方进行定位。
本发明的实施例还提供了一种性能测量的装置, 如图 8 ( a )所示, 该装置 包括:
接收模块 800, 用于接收至少一个性能测量报文, 该性能测量报文携带发送 端性能测量参数, 该发送端的性能测量参数与第一数据流对应, 该性能测量报 文是所述发送端基于上述第一数据流发送的;
管理模块 810, 用于获取发送端性能测量参数, 确定接收端性能测量参数, 该装置的性能测量参数与上述第一数据流对应;
测量模块 820, 用于根据上述发送端性能测量参数和接收端性能测量参数, 测量发送端和接收端之间的网络性能, 该网络性能与上述第一数据流对应。
本实施例中, 接收端即为上述性能测量的装置。
可选的, 如图 9所示, 上述接收模块 800可以包括:
丟包测量报文接收单元 801 , 用于接收至少一个性能测量报文, 该性能测量 报文携带发送端性能测量参数, 该性能测量参数至少包括发送端发送的第一数 据流的数据报文计数值;
和 /或,
时延测量报文接收单元 802, 用于接收至少一个性能测量报文, 该性能测量 报文携带发送端性能测量参数, 该性能测量参数至少包括发送端发送该性能测 量报文时的本地时间参数。 为便于说明, 图 9中仅示出了接收模块 800同时包括丟包测量报文接收单 元 801和时延测量报文接收单元 802的情况。
可选的, 如图 10所示, 上述管理模块 810可以包括:
丟包管理单元 811 ,用于获取丟包测量报文接收单元 801接收到的性能测量 报文中携带的发送端发送的第一数据流的数据报文计数值, 确定接收端接收到 的第一数据流的数据报文计数值;
和 /或,
时延管理单元 812, 用于获取时延测量报文接收单元 802接收到的性能测 量报文中携带的发送端发送该性能测量报文时的本地时间参数, 确定接收端接 收到该性能测量报文时的本地时间参数。
为便于说明, 图 10 中仅示出了管理模块 810 同时包括丟包管理单元 811 和时延管理单元 812的情况。
可选的, 如图 11所示, 上述测量模块 820可以包括:
丟包测量单元 821,用于根据丟包管理单元 811获取到的发送端发送的第一 数据流的数据报文计数值和确定的接收端接收到的第一数据流的数据报文计数 值, 测量第一数据流由发送端转发至接收端的丟包性能;
和 /或,
时延测量单元 822, 用于根据时延管理单元 812获取到的发送端发送该性 能测量报文时的本地时间参数和确定的接收端接收到该性能测量报文时的本地 时间参数, 测量第一数据流由发送端转发至接收端的时延性能。
为便于说明,图 11 中仅示出了测量模块 820同时包括丟包测量单元 821和 时延测量单元 822的情况。
可选的, 本发明的实施例中, 如图 8 ( b )所示, 上述性能测量装置还可以 包括:
发送模块 830, 用于转发接收模块 800接收到的性能测量报文; 或者复制 转发接收模块 800接收到的性能测量报文。
在实施例中,发送端基于具体的数据流发送性能测量报文,接收端的性能测 量装置接收与数据流对应的性能测量报文, 能够检测到具体每一个流的丟包、 时延等, 实现基于流的性能检测。
本发明的实施例还提供了一种性能测量的系统,如图 12所示,该系统包括: 第一设备,用于基于第一数据流发送性能测量报文,该性能测量报文中携带 第一设备的性能测量参数,该第一设备的性能测量参数与上述第一数据流对应; 第二设备,用于接收上述性能测量报文,获取性能测量报文中携带的第一设 备的性能测量参数, 确定第二设备的性能测量参数; 该第二设备的性能测量参 数与上述第一数据流对应;
根据上述第一设备的性能测量参数和第二设备的性能测量参数,测量第一设 备和第二设备之间的网络性能。
可选的,本发明的实施例中,上述第二网络设备接收到上述性能测量报文后, 转发或者复制转发上述性能测量 ·^艮文。
在实施例中, 第一设备基于具体的数据流发送性能测量报文, 第二设备接收 与数据流对应的性能测量 ·^艮文, 能够检测到具体每一个流的丟包、 时延等, 实 现基于流的性能检测。
如前面所介绍的,承载上述数据流的可以是 IP,或者是 MPLS,或者是 VPLS 等网络, 因此, 本发明的实施例中, 性能测量报文也可以根据实际的网络协议 来构造。
下面以上述性能测量报文是 IP报文为例, 介绍几种性能测量报文的具体格 式。
如图 13所示, 图 13示出了 IPv4报文(以下筒称为 IP报文) 的基本格式。 其中, 本发明实施例中的性能测量参数可以承载在 IP报文的数据部分。
当需要对丟包进行测量时,性能测量报文的格式可以是固定格式,也可以是 TLV ( Type, Length, Value, 类型, 长度, 值)格式。
例如, 固定格式可以是如图 14所示的格式, 其中:
级别, 用于表明性能测量报文所适用的维护域。
版本, 用于表明性能测量报文所适用测量方法的不同版本。 操作类型, 用于表明性能测量报文的类型为丟包测量报文。
发送周期 /包数, 用于表明预先设定的性能测量报文的发送周期, 或定预先 设定的定包发送报文数, 或按实际需要方式发送性能测量报文模式, 或按实际 需要方式发送性能测量报文模式, 为可选项。
TLV偏移, 用来指示 TLV在报文中的位置, TLV可用于功能的扩展, 为可 选项。
序列号, 用于表明丟包测量报文的顺序, 为可选项。
发送数据报文计数值, 用于记录发送的报文计数值。
接收数据报文计数值, 用于记录接收的报文计数值, 为可选项。
TLV结束标志, 用来表示 TLV结束, 遇到此标志, 表示所有 TLV的结束, 为可选项。
又例如, TLV格式可以是如图 15所示的格式, 其中:
TYPE, 用于表示报文的类型是丟包测量报文, 可以扩展。
LENGTH , 用于标识报文中数据的长度, 可以以字节计算。
级别, 用于表明性能测量报文所适用的维护域。
版本, 用于表明性能测量报文所适用测量方法的不同版本。
发送周期 /包数, 用于表明预先设定的性能测量报文的发送周期, 或定预先 设定的定包发送报文数,或按实际需要方式发送性能测量报文模式, 为可选项。
序列号, 用于表明丟包测量报文的顺序, 为可选项。
发送数据报文计数值, 用于记录发送的报文计数值。
接收数据报文计数值, 用于记录接收的报文计数值, 为可选项。
类似的, 当需要对时延进行测量时, 性能测量报文的格式同样可以是固定格 式, 也可以是 TLV ( Type, Length, Value, 类型, 长度, 值)格式。
例如, 固定格式可以是如图 16所示的格式, 其中:
级别, 用于表明性能测量报文所适用的维护域。
版本, 用于表明性能测量报文所适用测量方法的不同版本。
操作类型, 用于表明性能测量报文的类型为时延测量报文。 发送周期 /包数, 用于表明预先设定的性能测量报文的发送周期, 或定预先 设定的定包发送报文数,或按实际需要方式发送性能测量报文模式, 为可选项。
TLV偏移, 用来指示 TLV在报文中的位置, TLV用于功能的扩展, 为可选 项。
序列号, 用于表明发送时延测量报文的顺序, 为可选项。
发送时戳, 用来记录时延测量报文发送时的时间。
接收时戳, 用来记录时延测量 4艮文接收时的时间, 为可选项。
TLV结束标志, 用来表示 TLV结束, 遇到此标志, 表示所有 TLV的结束, 为可选项。
又例如, TLV格式可以是如图 17所示的格式, 其中:
TYPE, 用于表示报文的类型是时延测量报文, 可以扩展。
LENGTH , 用于标识报文中数据的长度, 可以以字节计算。
级别, 用于表明性能测量报文所适用的维护域。
版本, 用于表明性能测量报文所适用测量方法的不同版本。
发送周期 /包数, 用于表明预先设定的性能测量报文的发送周期, 或定预先 设定的定包发送报文数,或按实际需要方式发送性能测量报文模式, 为可选项。
序列号, 用于表明发送时延测量报文的顺序, 为可选项。
发送时戳, 用来记录时延测量报文发送时的时间。
接收时戳, 用来记录时延测量 4艮文接收时的时间, 为可选项。
当然, 图 14〜图 17示出的仅仅是几种报文格式的示意图,本发明的实施例 中, 性能测量报文的格式并不局限于此。
可选的,在本发明的实施例中, 为了使得接收到性能测量报文的设备能够识 别出该报文是性能测量报文, 可以在报文的某些字段设置一些特定的数值。
例如, 在一些场景中, 是基于二元组来区分不同的数据流, 此时, 可以将性 能测量报文的协议字段, 设置为 255。 此时, 根据该字段就能够识别出性能测 量报文, 同时性能测量报文的协议字段与数据流的协议字段不相同, 并不影响 性能测量报文和数据流的报文遵循相同的转发路径和转发规则等。 在另一些场景中,可能是基于五元组或者七元组来区分不同的数据流,此时, 可以选择五元组或者七元组之外的合适字段, 来承载特定的值, 以使得接收端 能够识别出性能测量报文。
当然, 具体的配置应视具体网络而定, 本发明的实施例不做具体的限定。 可选的, 上述性能测量报文可以是单播报文, 也可以是组播报文。
可选的, 上述性能测量报文也可以是 IPv6报文。
本发明的实施例还提供了一种性能测量的方法,如图 18所示,该方法包括: B100, 接收端接收至少一个性能测量报文, 该性能测量报文携带性能测量 报文的发送端的性能测量参数;
B110, 向该接收端的下游网络节点转发所述性能测量报文;
B120 , 获取发送端的性能测量参数, 确定该接收端的性能测量参数; B130, 根据发送端的性能测量参数和该接收端的性能测量参数, 测量所述 发送端和该接收端之间的网络性能。
可选的, 在本发明的实施例中, B110和 B120的顺序可调换。 不影响本发 明实施例的实现。
可选的, 在本发明的实施例中, 接收端接收到性能测量报文后, 可以复制接 收到的性能测量报文, 然后从复制的性能测量报文中获取发送端的性能测量参 数。 而对于该接收到的性能测量报文, 则继续转发出去, 不影响该性能测量报 文的转发。
可选的,本发明的实施例中, 该性能测量报文是上述发送端基于第一数据流 发送的, 此时, B110具体可以为:
该接收端基于上述第一数据流向下游网络节点转发,或者复制转发该性能测 量报文。
可选的, 在本发明的实施例中, 与图 2〜图 17所示的实施例中描述的情况 类似的, 上述第一数据流可以是单播流, 也可以是组播流。
例如, 当上述第一数据流是单播流时, 该接收端基于上述第一数据流向下游 网络节点转发该性能测量报文; 当上述第一数据流是组播流时, 该接收端同样 基于上述第一数据流向下游网络节点转发该性能测量报文, 当然, 在第一数据 流发生复制的网络节点, 性能测量报文也进行复制并转发。
当然, 图 2〜图 17所示的实施例中描述的其他情况, 均适用于本实施例, 此处不再赘述。
在本实施例中, 中间网络节点接收到性能测量报文后,进行相应的网络性能 的测量, 同时又将该性能测量报文转发下去, 使得该中间网络节点的下游网络 节点也能接收到该性能测量报文, 从而不仅能够实现端到端的网络性能测量, 还能够实现段到段的网络性能测量, 进一步能够实现网络故障的定位。
通过以上的实施方式的描述,本领域的普通技术人员可以清楚地了解到本发 明实施例可借助软件加必需的通用硬件平台的方式来实现, 当然也可以通过硬 件来实现。 基于这样的理解, 本发明实施例的技术方案可以以软件产品的形式 体现出来, 该计算机软件产品可以存储在存储介质中, 如 R〇M/RAM、 磁碟、 光盘等, 包括若干指令用以使得一台计算机设备、 或者服务器、 或者其他网络 设备执行本发明各个实施例或者实施例的某些部分所述的方法。
以上仅为本发明的较佳实施例, 并非用于限定本发明的保护范围。凡在本发 明的精神和原则之内, 所作的任何修改、 等同替换、 改进等, 均应包含在本发 明的保护范围之内。

Claims

权 利 要 求 书
1、 一种网络性能测量的方法, 其特征在于, 包括:
接收端接收至少一个性能测量报文, 所述性能测量报文携带所述性能测 量报文的发送端的性能测量参数, 所述发送端的性能测量参数与第一数据流对 ;
获取所述发送端的性能测量参数, 确定所述接收端的性能测量参数, 所 述接收端的性能测量参数与所述第一数据流对应;
根据所述发送端的性能测量参数和所述接收端的性能测量参数,测量所述发 送端和所述接收端之间的网络性能, 所述网络性能与所述第一数据流对应; 其中, 所述性能测量报文是所述发送端基于所述第一数据流发送的。
2、 如权利要求 1所述的方法, 其特征在于, 所述接收端接收至少一个性 能测量报文之后, 转发、 或者复制转发所述性能测量报文, 其中, 所述接收 端为中间网络节点。
3、 如权利要求 1 -2中任一项所述的方法, 其特征在于, 所述发送端的性 能测量参数至少包括: 所述发送端发送的所述第一数据流的报文计数值; 所述接收端的性能测量参数至少包括: 所述接收端接收的所述第一数据 流的报文计数值。
4、 如权利要求 3所述的方法, 其特征在于,
所述接收端接收至少一个性能测量报文, 具体为:
所述接收端先后接收第一性能测量报文和第二性能测量报文, 所述第一 性能测量报文携带所述发送端的第一性能测量参数,所述第二性能测量报文 携带所述发送端的第三性能测量参数;
所述获取所述发送端的性能测量参数,确定所述接收端的性能测量参数, 具体为:
获取所述第一性能测量参数, 确定所述接收端的第二性能测量参数; 获 取所述第三性能测量参数, 确定所述接收端的第四性能测量参数;
所述根据所述发送端的性能测量参数和所述接收端的性能测量参数, 测 量所述发送端和所述接收端之间的网络性能, 具体为:
根据所述第一性能测量参数、 第二性能测量参数、 第三性能测量参数、 以及第四性能测量参数, 测量所述发送端和所述接收端之间的丟包性能。
5、 如权利要求 1 -2中任一项所述的方法, 其特征在于, 所述发送端的性 能测量参数至少包括: 所述发送端发送所述性能测量报文时的本地时间参 数;
所述接收端的性能测量参数至少包括: 所述接收端接收所述性能测量报 文时的本地时间参数;
所述根据所述发送端的性能测量参数和所述接收端的性能测量参数, 测 量所述发送端和所述接收端之间的网络性能, 包括:
根据所述发送端发送所述性能测量报文时的本地时间参数和所述接收端 接收所述性能测量报文时的本地时间参数测量所述发送端和所述接收端之 间的时延性能。
6、 一种网络性能测量的方法, 其特征在于, 包括:
接收端接收至少一个性能测量报文, 所述性能测量报文携带所述性能测 量报文的发送端的性能测量参数;
向所述接收端的下游网络节点转发所述性能测量报文;
获取所述发送端的性能测量参数, 确定所述接收端的性能测量参数; 根据所述发送端的性能测量参数和所述接收端的性能测量参数, 测量所 述发送端和所述接收端之间的网络性能。
7、 如权利要求 6所述的方法, 其特征在于, 所述性能测量报文是所述发 送端基于第一数据流发送的;
所述向所述接收端的下游网络节点转发所述性能测量报文, 具体为: 所述接收端基于所述第一数据流向所述接收端的下游网络节点转发、 或 者复制转发所述性能测量报文。
8、 一种网络性能测量的装置, 其特征在于, 包括;
接收模块, 用于接收至少一个性能测量报文, 所述性能测量报文携带发 送端性能测量参数, 所述发送端的性能测量参数与第一数据流对应, 所述性 能测量报文是所述发送端基于所述第一数据流发送的;
管理模块, 用于获取所述发送端性能测量参数, 确定所述装置的性能测 量参数, 所述装置的性能测量参数与所述第一数据流对应;
测量模块, 用于根据所述发送端的性能测量参数和所述装置的性能测量 参数, 测量所述发送端和所述装置之间的网络性能, 所述网络性能与所述第 一数据流对应。
9、 如权利要求 8所述的装置, 其特征在于, 所述接收模块包括: 丟包测量报文接收单元, 用于接收所述性能测量报文, 所述性能测量报 文携带所述发送端性能测量参数,所述性能测量参数至少包括所述发送端发 送的所述第一数据流的报文计数值;
和 /或,
时延测量报文接收单元, 用于接收所述性能测量报文, 所述性能测量报 文携带所述发送端性能测量参数,所述性能测量参数至少包括所述发送端发 送所述性能测量报文时的本地时间参数。
1 0、 如权利要求 9所述的装置, 其特征在于, 所述管理模块包括: 丟包管理单元, 用于获取所述丟包测量报文接收单元接收到的性能测量 报文中携带的所述发送端发送的所述第一数据流的报文计数值,确定所述装 置接收到的所述第一数据流的报文计数值;
和 /或,
时延管理单元, 用于获取所述时延测量报文接收单元接收到的性能测量 报文中携带的所述发送端发送所述性能测量报文时的本地时间参数,确定所 述装置接收到所述性能测量报文时的本地时间参数。
1 1、 如权利要求 1 0所述的装置, 其特征在于, 所述测量模块包括: 丟包测量单元, 用于根据所述丟包管理单元获取到的所述发送端发送的 所述第一数据流的报文计数值和确定的所述装置接收到的所述第一数据流 的报文计数值,测量所述第一数据流由所述发送端转发至所述装置的丟包性 和 /或,
时延测量单元, 用于根据所述时延管理单元获取到的所述发送端发送所 述性能测量报文时的本地时间参数和确定的所述装置接收到所述性能测量 报文时的本地时间参数,测量所述第一数据流由所述发送端转发至所述装置 的时延性能。
12 , 如权利要求 8至 1 1 中任一项所述的装置, 其特征在于, 所述装置 还包括:
发送模块, 用于转发所述接收模块接收到的所述性能测量报文; 或者复 制转发所述接收模块接收到的所述性能测量报文。
13、 一种网络性能测量的系统, 其特征在于, 包括:
第一设备, 用于基于第一数据流发送性能测量报文, 所述性能测量报文 中携带所述第一设备的性能测量参数,所述第一设备的性能测量参数与所述 第一数据流对应;
第二设备, 用于接收所述性能测量报文, 获取所述性能测量报文中携带 的所述第一设备的性能测量参数, 确定所述第二设备的性能测量参数; 所述 第二设备的性能测量参数与所述第一数据流对应;
根据所述第一设备的性能测量参数和所述第二设备的性能测量参数, 测 量所述第一设备和所述第二设备之间的网络性能。
PCT/CN2009/071811 2009-05-15 2009-05-15 网络性能测量的方法、装置和系统 WO2010130102A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
CN2009801552232A CN102301651A (zh) 2009-05-15 2009-05-15 网络性能测量的方法、装置和系统
BRPI0924567A BRPI0924567A2 (pt) 2009-05-15 2009-05-15 método, aparelho e sistema para a medição de desempenho de rede
EP09844514A EP2432160A1 (en) 2009-05-15 2009-05-15 Method, device and system for measuring network performance
PCT/CN2009/071811 WO2010130102A1 (zh) 2009-05-15 2009-05-15 网络性能测量的方法、装置和系统
US13/296,782 US20120057497A1 (en) 2009-05-15 2011-11-15 Method, apparatus, and system for measuring network performance

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2009/071811 WO2010130102A1 (zh) 2009-05-15 2009-05-15 网络性能测量的方法、装置和系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US13/296,782 Continuation US20120057497A1 (en) 2009-05-15 2011-11-15 Method, apparatus, and system for measuring network performance

Publications (1)

Publication Number Publication Date
WO2010130102A1 true WO2010130102A1 (zh) 2010-11-18

Family

ID=43084620

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/071811 WO2010130102A1 (zh) 2009-05-15 2009-05-15 网络性能测量的方法、装置和系统

Country Status (5)

Country Link
US (1) US20120057497A1 (zh)
EP (1) EP2432160A1 (zh)
CN (1) CN102301651A (zh)
BR (1) BRPI0924567A2 (zh)
WO (1) WO2010130102A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103856369A (zh) * 2012-11-30 2014-06-11 雅马哈株式会社 用于网络音乐会话的延迟测量设备和延迟测量方法

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102685014B (zh) 2012-05-29 2014-06-25 华为技术有限公司 用于测量业务流的性能指标的方法和发送端设备
CN103401716B (zh) * 2013-08-06 2016-11-23 华为技术有限公司 网络性能测量的方法、网管设备和网元设备
US10135704B2 (en) * 2014-06-20 2018-11-20 Microsoft Technology Licensing, Llc Identification of candidate problem network entities
US9602374B2 (en) * 2014-07-21 2017-03-21 Ciena Corporation Systems and methods for collecting and analyzing data to determine link quality and stability in layer two networks
WO2021051418A1 (en) * 2019-09-21 2021-03-25 Huawei Technologies Co., Ltd. Methods and network nodes for reliability measurement

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102006045050A1 (de) * 2006-09-21 2008-03-27 Adva Ag Optical Networking Verfahren und Vorrichtung zur Bestimmung der Latenzzeit eines digitalen Übertragungssystems, insbesondere eines digitalen optischen Übertragungssystems
US20080276001A1 (en) * 2007-05-02 2008-11-06 Spirent Communications Of Rockville, Inc. Quality of experience indicator for network diagnosis
US20080291891A1 (en) * 2007-05-23 2008-11-27 Broadcom Corporation Synchronization Of A Split Audio, Video, Or Other Data Stream With Separate Sinks

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5793976A (en) * 1996-04-01 1998-08-11 Gte Laboratories Incorporated Method and apparatus for performance monitoring in electronic communications networks
CN100583784C (zh) * 2005-03-30 2010-01-20 华为技术有限公司 在多协议标签交换网络中监测丢包率的方法
CN100502324C (zh) * 2005-08-31 2009-06-17 华为技术有限公司 一种获取链路评价参数的方法
CN101247288B (zh) * 2008-03-14 2010-08-18 华为技术有限公司 链路质量检测方法和链路质量检测设备

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102006045050A1 (de) * 2006-09-21 2008-03-27 Adva Ag Optical Networking Verfahren und Vorrichtung zur Bestimmung der Latenzzeit eines digitalen Übertragungssystems, insbesondere eines digitalen optischen Übertragungssystems
US20080276001A1 (en) * 2007-05-02 2008-11-06 Spirent Communications Of Rockville, Inc. Quality of experience indicator for network diagnosis
US20080291891A1 (en) * 2007-05-23 2008-11-27 Broadcom Corporation Synchronization Of A Split Audio, Video, Or Other Data Stream With Separate Sinks

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2432160A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103856369A (zh) * 2012-11-30 2014-06-11 雅马哈株式会社 用于网络音乐会话的延迟测量设备和延迟测量方法

Also Published As

Publication number Publication date
EP2432160A4 (en) 2012-03-21
EP2432160A1 (en) 2012-03-21
US20120057497A1 (en) 2012-03-08
CN102301651A (zh) 2011-12-28
BRPI0924567A2 (pt) 2018-10-09

Similar Documents

Publication Publication Date Title
KR101459252B1 (ko) Traceroute_delay 진단 커맨드
Frost et al. Packet loss and delay measurement for mpls networks
Sharma et al. Enabling fast failure recovery in OpenFlow networks
JP7468969B2 (ja) 第1のネットワークノードにおいて使用される装置及びコントローラにおいて使用される装置
JP7424725B2 (ja) マルチキャストパケット処理方法、デバイスおよびコンピュータプログラム
CN101616037B (zh) 网络通信中丢包的检测方法、装置及系统
US20050099955A1 (en) Ethernet OAM fault isolation
WO2010130102A1 (zh) 网络性能测量的方法、装置和系统
WO2006133635A1 (fr) Procede de mesure des parametres de performance du reseau a commutation multiprotocole par etiquette
US20130159806A1 (en) Packet communication apparatus and packet communication method
CA2818316A1 (en) Systems and methods for measuring available capacity and tight link capacity of ip paths from a single endpoint
US9641420B1 (en) Methods and apparatus for assessing the quality of a data path including both layer-2 and layer-3 devices
WO2008037203A1 (en) Method and node device for realizing the network topology discovery
WO2013044827A1 (zh) 一种跟踪路由测试方法、系统、装置及设备
WO2007073649A1 (fr) Procede et systeme pour obtenir une unite de transfert maximale de voie dans un reseau
JP5766882B2 (ja) 終端間多重パスネットワークシステムのためのフィードバックプロトコル
TWI492575B (zh) 快速標籤交換路徑警示機制
WO2007056915A1 (en) A method for measuring mpls network performance parameter and device and system for transmitting packet
WO2009092257A1 (zh) 运营商骨干网传输网络的故障检测方法和装置
WO2011144114A2 (zh) 探测路径信息的方法、设备及系统
Mizrahi et al. An overview of operations, administration, and maintenance (OAM) tools
CN112005527A (zh) 能够在分组交换通信网络中进行性能测量
WO2013189414A2 (zh) 网络拓扑自动获取方法及系统、网络查询及管理系统
WO2012071851A1 (zh) 根据网络抖动调整bfd发送间隔的方法及装置
WO2014008809A1 (zh) 一种丢失帧测定方法及系统

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200980155223.2

Country of ref document: CN

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

Ref document number: 09844514

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: 8678/CHENP/2011

Country of ref document: IN

WWE Wipo information: entry into national phase

Ref document number: 2009844514

Country of ref document: EP

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: PI0924567

Country of ref document: BR

ENP Entry into the national phase

Ref document number: PI0924567

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20111116