WO2015096636A1 - Procédé de mesure de paquets perdus et dispositif réseau - Google Patents

Procédé de mesure de paquets perdus et dispositif réseau Download PDF

Info

Publication number
WO2015096636A1
WO2015096636A1 PCT/CN2014/093824 CN2014093824W WO2015096636A1 WO 2015096636 A1 WO2015096636 A1 WO 2015096636A1 CN 2014093824 W CN2014093824 W CN 2014093824W WO 2015096636 A1 WO2015096636 A1 WO 2015096636A1
Authority
WO
WIPO (PCT)
Prior art keywords
data packet
node
group
packet group
data
Prior art date
Application number
PCT/CN2014/093824
Other languages
English (en)
Chinese (zh)
Inventor
彭敏
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
Publication of WO2015096636A1 publication Critical patent/WO2015096636A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0823—Errors, e.g. transmission errors
    • H04L43/0829—Packet loss

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a method and network device for packet loss measurement.
  • Multiprotocol Label Switching uses short, fixed-length labels to encapsulate packet packets.
  • a tag is a short identifier with a fixed length and only local significance. It is used to uniquely identify the Forwarding Equivalence Class (FEC) to which a packet belongs.
  • FEC Forwarding Equivalence Class
  • the transmitting end may send a Loss Measurement Message (LMM) with the request information to the receiving end, where the LMM includes the local sending packet count value of the sending end of the data packet, and the receiving end receives the packet.
  • LMM Loss Measurement Message
  • LMR Lost Measurement Reply
  • the LMR message includes a packet count value when the LMM is received.
  • the sender can perform packet loss measurement according to the packet count value and the packet count value.
  • multiple data packets sent by the sender may be different from the path that is passed, causing the data packet to arrive at the receiving end later than the LMM, causing an out-of-order problem, which may result in a packet loss measurement result. Inaccurate.
  • the embodiment of the invention provides a method for measuring packet loss and a network device, which is beneficial to improving the accuracy of the packet loss measurement result in a load balancing scenario.
  • a first aspect of the present invention provides a method for packet loss measurement, including:
  • the first data packet includes a sequence number of the first data packet group and a first data packet number, where the first data packet number is used to identify a quantity of data packets included in the first data packet group;
  • the second node determines that the second data packet does not belong to the first data packet group, the number of received second data packets belonging to the first data packet group is counted;
  • the second node determines the number of lost packets of the first data packet group according to the first data packet number and the second data packet number.
  • the method further includes:
  • the second node When the second node receives the data packet of the first data packet group, counting and receiving the received data packet belonging to the first data packet group, and correspondingly, according to the statistics, the first data The number of packets of the packet group determines the number of the second packet.
  • each data packet in the first data packet group is encapsulated with a special label, and the special label includes the first data.
  • the method further includes:
  • the second node determines that the second data packet does not encapsulate the special label, determining that the second data packet does not belong to the first data packet group, or
  • the second node determines that the sequence number of the second data packet group to which the second data packet belongs is different from the sequence number of the first data packet group, determining that the second data packet does not belong to the first data packet group.
  • the method further includes:
  • the second node starts a specific duration timer
  • the second node counts the received number of the second data packets belonging to the first data packet group.
  • the two nodes determine, according to the number of the first data packet and the number of the second data packet, the number of packet loss of the first data packet group, which specifically includes:
  • the X represents the number of first data packets
  • the Y represents the number of second data packets
  • the Z represents the number of lost packets.
  • a second aspect of the present invention provides a method for packet loss measurement, including:
  • the first node determines a quantity of the first data packet included in the first data packet group
  • the first node After the data packet in the first data packet group is sent, the first node sends a second data packet to the second node, where the second data packet does not belong to the first data packet group.
  • the first node before the sending, by the first node, the first data packet in the first data packet group to the second node, the first node further includes:
  • the first node encapsulates the first data packet by using a special label, where the special label includes a sequence number of the first data packet group and the first data packet number.
  • each data packet in the first data packet group is encapsulated with the special label.
  • a third aspect of the present invention provides a second node, including:
  • a receiving unit configured to receive a first data packet sent by the first node, where the first data packet belongs to a first data packet group, and the first data packet group is sent by the first node to the second node a data packet group, the first data packet includes a sequence number of the first data packet group and a first data packet number, where the first data packet number is used to identify a data packet included in the first data packet group Quantity
  • the receiving unit is further configured to receive a second data packet sent by the first node
  • a statistic unit configured to determine, when the second data packet does not belong to the first data packet group, the number of second data packets that are received by the first data packet group;
  • a calculating unit configured to determine, according to the first data packet quantity and the second data packet quantity, a number of packet loss of the first data packet group.
  • the method further includes:
  • a counting unit configured to count statistics of the received data packets belonging to the first data packet group, and correspondingly, determine the number of the second data packet according to the statistics, the number of data packets of the first data packet group .
  • each data packet in the first data packet group is encapsulated with a special label, where the special label includes the first data The serial number of the packet group and the number of the first data packet.
  • the method further includes:
  • a determining unit configured to: when it is determined that the second data packet does not encapsulate the special label, determine that the second data packet does not belong to the first data packet group, or
  • the method further includes:
  • timing unit for starting a specific duration timer a timing unit for starting a specific duration timer
  • the statistic unit is specifically configured to determine, after the timer started by the timing unit expires, the received number of the second data packets belonging to the first data packet group.
  • the calculating The unit is specifically used to:
  • the X represents the number of first data packets
  • the Y represents the number of second data packets
  • the Z represents the number of lost packets.
  • a fourth aspect of the present invention provides a first node, including:
  • a determining unit configured to determine a quantity of the first data packet included in the first data packet group
  • a sending unit configured to send, to the second node, the first data packet in the first data packet group, where the first data packet includes a sequence number of the first data packet group and a first data packet number, where The first number of data packets is used to identify the number of data packets included in the first data packet group, where a data packet group is a data packet group sent by the first node to the second node;
  • the sending unit is further configured to: after the data packet in the first data packet group is sent, send a second data packet to the second node, where the second data packet does not belong to the first data packet group.
  • the method further includes:
  • An encapsulating unit configured to encapsulate the first data packet by using a special label, where the special label includes the first data, before the sending unit sends the first data packet in the first data packet group to the second node The serial number of the packet group and the number of the first data packet.
  • the encapsulating unit is further configured to encapsulate the special label for each data packet in the first data packet group.
  • the second node receives the first data packet sent by the first node, the first data packet belongs to the first data packet group, and the first data packet includes a sequence of the first data packet group. Number and the first number of data packets, the first number of data packets is used to identify the number of data packets included in the first data packet group, when the second node receives not belonging to the first data packet group And the second data packet, the second data packet that belongs to the first data packet group is counted, and the second node determines the number according to the first data packet quantity and the second data packet quantity.
  • the number of packet loss in a packet group is beneficial to improve the accuracy of packet loss measurement results in a load balancing scenario.
  • FIG. 1 is a flowchart of a method for packet loss measurement according to an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of a method for packet loss measurement according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of interaction between a first node and a second node according to an embodiment of the present disclosure
  • FIG. 4 is a schematic diagram of a format of an MPLS special label according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of state switching of a second node according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a second node according to an embodiment of the present disclosure.
  • FIG. 7 is a schematic structural diagram of a second node according to an embodiment of the present disclosure.
  • FIG. 8 is a schematic structural diagram of a second node according to an embodiment of the present disclosure.
  • FIG. 9 is a schematic structural diagram of a second node according to an embodiment of the present disclosure.
  • FIG. 10 is a schematic structural diagram of a first node according to an embodiment of the present disclosure.
  • FIG. 11 is a schematic structural diagram of a first node according to an embodiment of the present disclosure.
  • FIG. 12 is a schematic structural diagram of a second node according to an embodiment of the present disclosure.
  • FIG. 13 is a schematic structural diagram of a first node according to an embodiment of the present invention.
  • FIG. 1 is a schematic flowchart diagram of a method for packet loss measurement according to an embodiment of the present invention. As shown in FIG. 1 , the method for packet loss measurement in this embodiment includes the following steps:
  • Step S100 The second node receives the first data packet sent by the first node, where the first data packet belongs to the first data packet group, and the first data packet group is sent by the first node to the second node.
  • a data packet group the first data packet includes a sequence number of the first data packet group and a first data packet number, where the first data packet number is used to identify a data packet included in the first data packet group quantity.
  • Step S101 The second node receives the second data packet sent by the first node.
  • Step S102 When the second node determines that the second data packet does not belong to the first data packet group, the number of received second data packets belonging to the first data packet group is counted.
  • Step S103 The second node determines the number of lost packets of the first data packet group according to the first data packet number and the second data packet number.
  • the executor of each step in this embodiment is a second node.
  • the second node may be a Label Switched Router (LSR) in the MPLS domain, and the second section
  • LSR Label Switched Router
  • the point can be any node other than the head node on the Label Switched Path (LSP), that is, the intermediate node or the tail node on the LSP.
  • LSP Label Switched Path
  • the second node may determine, according to the sequence number of the first data packet group, the first data packet to belong to.
  • each data packet in the first data packet group is encapsulated with a special label, where the special label includes a sequence number of the first data packet group and the first data packet number.
  • the second node may obtain the sequence number of the first data packet group and the first data packet number by parsing the special label encapsulated on the first data packet.
  • the special label may be an MPLS label, and the second node may parse the MPLS label by using a prior art method, and obtain a sequence number of the first packet group and the first data packet included in the special label. Quantity.
  • the second node when the second node receives the data packet of the first data packet group, performing counting statistics on the received data packets belonging to the first data packet group, and correspondingly, according to the statistics
  • the number of packets of the first packet group determines the number of the second packet.
  • the second node may further read the special identifier directly carried in the first data packet, and obtain the serial number of the first data packet group and the first data packet by using the special identifier. Quantity. It should be noted that the specific manner in which the second node obtains the special identifier from the first data packet is not limited herein. For example, the second node may agree with the first node and read the first data at the agreed location. The special identifier of the packet can also directly traverse the first data packet to obtain a special identifier.
  • the first node continuously sends the data packet in the first data packet group, that is, after the first node sends the first data packet in the first data packet group, the first node continuously sends the first data packet. Other packets in a packet group.
  • the second node after receiving the first data packet of the first data packet group sent by the first node, the second node continuously receives other data packets in the first data packet group.
  • the second node After the second node receives the second data packet sent by the first node, the second node needs to determine whether the second data packet belongs to the first data packet group. When the second node determines that the second data packet does not encapsulate the special label, determining that the second data packet does not belong to the first data packet group, or when the second node determines the first data packet The second packet group to which the second data packet belongs When the sequence number is different from the sequence number of the first packet group, it is determined that the second data packet does not belong to the first data packet group.
  • the second node may collect, according to the sequence number of the first data packet group, the received second data of the data packet belonging to the first data packet group. The number of packages.
  • the second node may determine, according to the first data packet quantity and the second data packet quantity, that the data packet in the first data packet group is lost on the second node. The number of packages.
  • the second node receives the first data packet sent by the first node, the first data packet belongs to the first data packet group, and the first data packet includes the sequence of the first data packet group.
  • Number and the first number of data packets when the second node receives the second data packet that does not belong to the first data packet group, the number of received second data packets belonging to the first data packet group is counted Determining, by the second node, the number of packet loss of the first data packet group according to the first data packet quantity and the second data packet quantity, because the second node only receives the When the second data packet of the data packet group is used, the packet loss detection is performed, which is advantageous for improving the accuracy of the packet loss measurement result in the load sharing scenario.
  • FIG. 2 is a schematic flowchart diagram of a method for packet loss measurement according to an embodiment of the present invention. As shown in FIG. 2, the method for packet loss measurement in this embodiment includes the following steps:
  • Step S200 The first node determines a quantity of the first data packet included in the first data packet group.
  • Step S201 the first node sends a first data packet in the first data packet group to a second node, where the first data packet includes a sequence number of the first data packet group and a first data packet number.
  • the first number of data packets is used to identify the number of data packets included in the first data packet group, and the first data packet group is a data packet group that is sent by the first node to the second node.
  • Step S202 after the data packet in the first data packet group is sent, the first node sends a second data packet to the second node, where the second data packet does not belong to the first data packet. group.
  • the executor of each step in this embodiment is the first node.
  • the first node may be an LSR in the MPLS domain, which may be any node except the tail node on the LSP, that is, it may be determined on the LSP. Head node or intermediate node.
  • the first node may not send an LMM message to the second node to measure the packet loss. Instead, the first data packet to which the data packet belongs and the first data packet number in the data packet group can be identified directly by the first data packet sent.
  • the method before the sending, by the first node, the first data packet in the first data packet group to the second node, the method further includes:
  • the first node encapsulates the first data packet by using a special label, where the special label includes a sequence number of the first data packet group and the first data packet number. Each of the first packet group is encapsulated with the special tag.
  • the first node continuously sends the data packet in the first data packet group, that is, after the first node sends the first data packet in the first data packet group, the first node continuously sends the first data packet. Other packets in a packet group.
  • the second node after receiving the first data packet of the first data packet group sent by the first node, the second node continuously receives other data packets in the first data packet group.
  • the second node After the second node receives the second data packet sent by the first node, the second node needs to determine whether the second data packet belongs to the first data packet group. When the second node determines that the second data packet does not encapsulate the special label, determining that the second data packet does not belong to the first data packet group, or when the second node determines the first data packet When the sequence number of the second data packet group to which the two data packets belong is different from the sequence number of the first data packet group, it is determined that the second data packet does not belong to the first data packet group.
  • the first node is sending In the case of the second data packet, only step S200 and step S201 may be repeatedly performed; if the second data packet is a normal data packet, that is, the second data packet does not include the sequence number of the first data packet group and the The first number of data packets, at which time the first node can perform forwarding operations on normal data packets in the prior art.
  • the second node may determine, according to the received first data packet and the second data packet, the received The second packet number of the first packet group, and determining the number of packet loss of the first packet group according to the first packet number and the second packet number.
  • the first data packet including the sequence number and the first data packet number is sent, so that the second node can acquire the first data packet group and the first data packet group where the first data packet is located.
  • the first number of data packets, and determining the number of packet loss of the first data packet group according to the received first data packet and the second data packet, because the second node only receives the first data packet group The packet loss detection is performed only when the second data packet is used, which is beneficial to improving the accuracy of the packet loss measurement result in the load sharing scenario.
  • FIG. 3 is a schematic flowchart of interaction between a first node and a second node according to an embodiment of the present invention. As shown in FIG. 3, the interaction process includes the following steps:
  • Step S300 The first node determines a quantity of the first data packet included in the first data packet group.
  • the execution subject is the first node, and in practice, the node may be an LSR in the MPLS domain.
  • the first data packet may include a sequence number of the first data packet group to which it belongs and the first data packet number.
  • the first number of packets can be identified by:
  • the first data packet includes a transmission count x, then, optionally, the number of packets sent by the first node in each round of statistics, that is, the number of the first data packet can be expressed as 10 x +1 , and if 8 bits are used to represent x, then The first node needs to send at least 10 packets per round of statistics, and at most 10256 packets are sent.
  • the number of packets sent in the statistics of each round of the first node is the number of first packets.
  • the first node can automatically calculate the transmission count included in the first data packet sent in each round of statistics according to the traffic size and the statistical duration configured by the system.
  • the first node may count the first data packet sent in each round of statistics to record the number of first data packets sent by the first node, when the number of the first data packet reaches 10 x+ When one is available, the first node can determine that the first data packet of the round is sent.
  • the number of the first data packet is 10 x +1 , but in practice, the first data packet that the first node may send is less than 10 x+1 , that is, The first data packet sent by the first node includes a transmission count of x, but the number of first data packets actually sent is n, and the value of n is less than 10 x+1 , and if the first data packet is received, The second node determines, according to the foregoing sending count, the number of first data packets sent by the first node in the round of statistics, and calculates the number of lost packets of the first data packet group, and mistakes the 10 x+1 that is not sent by the first node.
  • the data packet is such that the total number of data packets transmitted is 1 million, and the above 1 million data packets are transmitted.
  • 10 x+1 is only a counting method agreed by the first node and the second node, and the formula is not necessarily used in actual operation, and a calculation formula known by other transceivers may also be used.
  • Step S301 The first node encapsulates the first data packet by using a special label.
  • the first node may encapsulate the first data packet by using a special label
  • the special label may include a sequence number of the first data packet group and a first data packet number.
  • the first node may encapsulate the special label for each data packet in the first data packet group.
  • Different types of special tags can have different formats. Take the MPLS label as an example: the MPLS label length is fixed, and is used to uniquely identify a Forwarding Equivalence Class (FEC) to which a packet belongs. Usually the length of the tag is 4 bytes, ie 32 bits. In the embodiment of the present invention, the 32-bit bit included in the existing MPLS type special label may be redefined.
  • FIG. 4 is a schematic diagram of the MPLS special label format provided by the embodiment of the present invention. As shown in FIG. 4, the special label may include:
  • Send count (Tx count), used to identify the first number of packets.
  • the special label value can be used to indicate whether the current label is a special label, and its length is 20 bits. In actual use, the value can be taken as an MPLS label stack code, that is, RFC3032. Among the several reserved labels specified in 4-15, one label that has not been occupied by other protocols. It should be noted that, as long as the data packets sent by the first node to the second node are encapsulated with the above special labels, their special label values are the same regardless of whether they belong to the same data packet group; the serial number is used to identify the first The first packet group to which the data packet belongs is 3 bits long.
  • the MPLS labels encapsulated in the data packets of different data packet groups may correspond to different serial numbers.
  • the first node After the first node sends a group of data packets, the first node sends the data packet.
  • the serial number is incrementally processed to distinguish different packet groups in which different data packets are located; the bottom mark can be the same as the prior art definition, indicating that the MPLS supports the hierarchical structure of the label, that is, supports the multi-layer label, and its length is 1 Bit, when the value is 1, indicates the lowest level label; the transmission count can be used to identify the first number of data packets of the first data packet group, and its length is 8 bits.
  • Step S302 The first node sends the first data packet in the first data packet group to the second node.
  • step S201 This step is the same as the description of step S201, and details are not described herein again.
  • Step S303 The first node sends a second data packet to the second node.
  • the first node may further send the second data packet to the second node, where the second data packet does not belong to the first data packet group.
  • the first node may perform two different processing on the second data packet according to actual needs. One is to terminate the special label for the second data packet in the second data packet group, but to complete the forwarding according to the determined LSP path, and the second data packet at this time is a normal data packet without the special label encapsulated.
  • the processing mode is applicable to on-demand statistics, that is, one-time statistics, that is, the system only needs to count the number of packets lost in the first packet group. In another case, if the statistical requirements of the system configuration are continuous statistics, that is, If multiple rounds of statistics are needed, the first node prepares for the next statistics after sending the first packet group.
  • the first node may perform a cyclic increment process on the sequence number included in the second data packet of the second data packet group, and repeat the foregoing steps S300-S302 to send the second data packet.
  • the first node encapsulates a special label with a sequence number of 0 for the first data packet in the first data packet group, and then the second data packet is encapsulated when the special data label is encapsulated for the data packet of the second data packet group.
  • the sequence number of the data packet in the group is incremented to 1.
  • the subsequent each data packet is similarly incremented. It should be noted that since the serial number of the special label is 3 bits, only 8 different numbers of 0-7 can be represented.
  • the first node when the serial number is incremented to 7, the first node is in the data of the subsequent data packet group.
  • the column number is reset to 0, and the above process is the cyclic increment process of the serial number.
  • the number of data packet groups sent by the first node is not limited herein.
  • the second node may repeat the above steps S300-S302 for the data packets in each data packet group. operating.
  • Step S304 The second node receives the first data packet sent by the first node.
  • Step S305 The second node receives the second data packet sent by the first node.
  • Step S306 The second node determines whether the second data packet belongs to the first data packet group.
  • the second node determines that the second data packet does not encapsulate the special label, determining that the second data packet does not belong to the first data packet group, or when the second node determines the first data packet When the sequence number of the second data packet group to which the two data packets belong is different from the sequence number of the first data packet group, it is determined that the second data packet does not belong to the first data packet group.
  • Step S307 The second node starts a specific duration timer. After the timer expires, the second node counts the received second data packet belonging to the first data packet group.
  • the second node Since different data packets sent by the first node do not necessarily arrive at the second node in the order of transmission, the second node needs to wait for a period of time to receive the delayed data packets in the same data packet group, so as to more accurately measure the number of lost packets.
  • the second node may start a timer of a specific duration, and if the data packet in the first packet group to which the first data packet belongs is not received before the timer expires, after the timer expires, Counting the number of second packets of the received first packet group.
  • the second node first receives some data packets with sequence number 0, that is, the first data packet belonging to the first data packet group. Then, a second data packet with the sequence number 1 arrives, that is, the data packet belonging to the second data packet group, or another second data packet without the special label is encapsulated, according to the description of step S306, the second The node needs to be the same as the sequence number of the first data packet received by the node, and if the second data packet is the data packet with the sequence number 1, since the sequence number of the first data packet is 0, therefore, The two nodes can start a timer. When the timer expires, the second node counts the received second packet number belonging to the first packet group.
  • the timer is for the second node to wait for a certain period of time. If the data packet of the first data packet group is not received within the specific time, the timer is read in step S304 when the timer expires.
  • the statistical count value of the first data packet group is the second data packet number in the first data packet group received by the second node.
  • the second node receives the data packet with the same sequence number as the first data packet, for example, the second node receives the second data packet with the sequence number 1 or After the second data packet of the special label is not encapsulated, the first data packet with the sequence number 0 is received.
  • the second node counts the first data packet, that is, The statistics counter value of the first packet group is incremented by 1, and the timer can be turned off.
  • the first node still repeats the foregoing counting operation until receiving a data packet that is not in the first data packet group or There is no normal data packet encapsulating the special label. At this time, the second node starts the above timer again, and repeats the operation of step S307.
  • duration of the timer is set according to actual conditions, such as actual network congestion conditions, for example, 60 seconds or 120 seconds.
  • Step S308 The second node determines the number of lost packets of the first data packet group according to the first data packet number and the second data packet number.
  • the second node may obtain the first data packet quantity included in the first data packet group from the transmission count of the first data packet, and after the operation of step S307, the second node may determine the first data packet that is received by the second node.
  • the second packet amount of the group, therefore, the corresponding packet loss number of the first packet group is:
  • Z X-Y, where X represents the number of first packets, Y represents the number of second packets, and Z represents the number of packets lost.
  • the embodiment provides a state switching diagram of the second node. As shown in FIG. 5, there are four states of the second node: an initial state, a statistical state, a waiting state, and an ending state.
  • the initial state indicates an initial state when the second node does not enable the packet loss statistics function, and when the second node receives the first packet encapsulated with the special label of the first packet group, it starts the counter. And entering the statistical state, when the second node is in the statistical state, if the second node receives the same data packet with the serial number and the first data packet, the receiving counter corresponding to the first data packet group is incremented by one; If the second received second packet sequence number is different from the sequence number of the first packet group, a timer for the first packet group is started, and a waiting state is entered, where the waiting state refers to the second node Wait a while to receive the delayed packets in the first packet group.
  • the second node When the second node is in the waiting state, if the timer has not expired, and the second node receives again After the data packet in the first packet group, it switches from the waiting state to the statistical state, and then counts the received data of the first packet group. If the timer expires, the second node enters the end state from the waiting state, and reports the statistical result in the state, that is, reports the number of data packets of the first data packet group actually received. It should be noted that the statistical state, the waiting state, and the ending state can be restored to the initial state by artificially enabling the operation of the statistical function of the node.
  • any node except the first node on the LSP can be used as the second node, and thus the node can be
  • the measurement of the number of packet loss is performed, that is, the solution of the embodiment of the present invention can not only know the number of lost packets from the first node to any node on the LSP path, but also the second reported by any two nodes. The number of packets is subtracted, and the number of packets lost between any two nodes is obtained.
  • the second node may be reported by adding a device as a public server in the network area where the LSP path between the first node and the second node is located, or by using an existing network management system. Specifically, when all the second nodes enabled with the statistics function finish the statistics of the current round, the second data packet counted by each is summarized on the public server or the network management. This feature enhancement allows the user to locate the faulty segment of a specific packet loss.
  • Step S309 the second node decapsulates the special label of the first data packet.
  • the second node can enable the statistics function to perform statistical reporting, and can also enable the deletion function to decapsulate the special label for the first data packet.
  • the second node determines the path end point of the LSP, that is, when the first data packet with the special label reaches the second node, the second The node not only performs statistical reporting, but also removes the special label of the first data packet, so that the data packet that is subsequently forwarded along the determined LSP is restored to the normal encapsulation and no special label is carried.
  • the second node receives the first data packet sent by the first node, the first data packet belongs to the first data packet group, and the first data packet includes the first data packet group.
  • a serial number and a first number of data packets when the second node receives the second data packet that does not belong to the first data packet group, the received second data packet that belongs to the first data packet group is counted
  • the number, the second node is determined according to the number of the first data packet and the number of the second data packet
  • the number of packet loss of the first data packet group is that the second node only performs packet loss detection when receiving the second data packet that does not belong to the first data packet group, which is beneficial to the load. Improve the accuracy of packet loss measurement results in a shared scenario.
  • FIG. 6 is a schematic structural diagram of a second node according to an embodiment of the present invention.
  • the second node 600 includes a receiving unit 610, a statistic unit 611, and a calculating unit 612.
  • the second node 600 can be the second node in the embodiment corresponding to FIG. 1 or FIG. 2 or FIG. 3.
  • the receiving unit 610 is configured to receive the first data packet sent by the first node, where the first data packet belongs to the first data packet group, and the first data packet group is the data packet group sent by the first node to the second node, where the first data is The packet includes a sequence number of the first packet group and a first packet number, and the first packet number is used to identify the number of data packets included in the first packet group;
  • the receiving unit 610 is further configured to receive the second data packet sent by the first node;
  • the statistics unit 611 is configured to determine, when the second data packet does not belong to the first data packet group, the number of the second data packets that belong to the first data packet group.
  • the calculating unit 612 is configured to determine the number of lost packets of the first data packet group according to the first data packet number and the second data packet number.
  • each data packet in the first data packet group is encapsulated with a special label, and the special label includes a serial number of the first data packet group and a first data packet number.
  • the statistics unit 611 is specifically configured to determine, after the timer started by the timing unit expires, the received second data packet belonging to the first data packet group.
  • calculating unit 612 is specifically configured to:
  • X represents the number of first packets
  • Y represents the number of second packets
  • Z represents the number of lost packets
  • the second node 600 may further include:
  • the counting unit 710 is configured to count statistics of the received data packets belonging to the first data packet group, and accordingly, determine the second data packet quantity according to the statistical data packet number of the first data packet group.
  • the second node 600 may further include:
  • the determining unit 810 is configured to: when determining that the second data packet does not encapsulate the special label, determine that the second data packet does not belong to the first data packet group, or determine the sequence number of the second data packet group to which the second data packet belongs When the sequence number of the first packet group is different, it is determined that the second packet does not belong to the first A packet group.
  • the second node 600 may further include:
  • the timing unit 910 is configured to start a specific duration timer, and when the timer expires, determine the number of the second data packets that belong to the first data packet group.
  • the first data packet including the first data packet group serial number and the first data packet number is sent, so that the second node 600 can acquire the first data packet group where the first data packet is located. And determining a number of packet loss of the first packet group according to the first packet number of the first packet group, and determining, according to the received first packet and the second packet, that the second node 600 only receives the packet When the second data packet belongs to the first data packet group, the packet loss detection is performed, which is beneficial to improving the accuracy of the packet loss measurement result in the load balancing scenario.
  • FIG. 10 is a schematic structural diagram of a first node according to an embodiment of the present invention.
  • the first node 1000 includes a determining unit 1010 and a transmitting unit 1020.
  • the first node 1000 may be the first node in the embodiment corresponding to FIG. 1 or FIG. 2 or FIG.
  • a determining unit 1010 configured to determine a quantity of the first data packet included in the first data packet group
  • the sending unit 1020 is configured to send, to the second node, the first data packet in the first data packet group, where the first data packet includes a sequence number of the first data packet group and a first data packet number, where the first data packet number is used. Identifies the number of packets included in the first packet group;
  • the sending unit 1020 is further configured to send the second data packet to the second node, where the second data packet does not belong to the first data packet group.
  • the first node 1000 further includes:
  • the encapsulating unit 1110 is configured to encapsulate the first data packet by using a special label before the sending unit 1020 sends the first data packet in the first data packet group to the second node, where the special label includes the The serial number of the first packet group and the first packet number.
  • the encapsulating unit 1110 is further configured to encapsulate the special label for each data packet in the first data packet group.
  • the first node 1000 sends the first data packet including the first packet group serial number and the first data packet number, so that the second node can obtain the first data packet. a first packet number of the data packet group and the first data packet group, and determining the number of packet loss of the first data packet group according to the received first data packet and the second data packet, since the second node is only in the Receiving the second data packet that does not belong to the first data packet group,
  • the packet loss detection is beneficial to improve the accuracy of packet loss measurement results in a load balancing scenario.
  • an embodiment of the present invention further provides a schematic diagram of a second node.
  • the second node 1200 may be the second node in the embodiment corresponding to FIG. 1 or FIG. 2 or FIG.
  • the second node 1200 includes a memory 1201 and a receiver 1202, and a processor 1203 connected to the memory 1201 and the receiver 1202, respectively.
  • the memory 1201 is configured to store a set of program instructions, and the processor 1203
  • the program instructions for invoking the memory 1201 to store perform the following operations:
  • the trigger receiver 1202 receives the first data packet sent by the first node, where the first data packet belongs to the first data packet group, and the first data packet group is a data packet group sent by the first node to the second node, where the first data packet includes a sequence number of the first packet group and a first packet number, wherein the first packet number is used to identify the number of data packets included in the first packet group;
  • the processor 1203 determines that the second data packet does not belong to the first data packet group
  • the received second data packet number belonging to the first data packet group is counted, and according to the first data packet number and the second data packet number. Determine the number of lost packets in the first packet group.
  • each data packet in the first data packet group is encapsulated with a special label, and the special label includes a serial number of the first data packet group and a first data packet number.
  • the processor 1203 is specifically configured to determine, after the timer started by the timing unit expires, the received second data packet belonging to the first data packet group.
  • processor 1203 is specifically configured to:
  • X represents the number of first packets
  • Y represents the number of second packets
  • Z represents the number of lost packets
  • the processor 1203 may be a central processing unit (CPU), and the memory 1201 may be an internal memory of a random access memory (RAM) type, and the receiver 1202
  • the physical interface may be an Ethernet (Ethernet) interface or an Asynchronous Transfer Mode (ATM) interface.
  • the processor 1203, the transmitter 1202, and the memory 1201 may be integrated into one or more independent circuits or hardware, such as an Application Specific Integrated Circuit (ASIC).
  • ASIC Application Specific Integrated Circuit
  • FIG. 13 is a schematic structural diagram of a first node according to an embodiment of the present invention.
  • the first node 1300 may be the first node in the embodiment corresponding to FIG. 1 or FIG. 2 or FIG.
  • the first node 1300 includes a memory 1301 and a transmitter 1302, and a processor 1303 connected to the memory 1301 and the transmitter 1302, respectively.
  • the memory 1301 is configured to store a set of program instructions, and the processor 1303
  • the program instructions for invoking the memory 1301 to store perform the following operations:
  • the trigger transmitter 1302 sends the first data packet in the first data packet group to the second node, where the first data packet includes a sequence number of the first data packet group and a first data packet number, where the first data packet number is used to identify the first data packet.
  • the number of data packets included in a packet group, and the second data packet is sent to the second node, and the second data packet does not belong to the first data packet group.
  • the processor 1303 is further configured to: after the transmitter 1302 sends the first data packet in the first data packet group to the second node, encapsulate the first data packet by using a special label, where the special label The sequence number of the first packet group and the first packet number are included.
  • the processor 1303 is further configured to encapsulate the special label for each data packet in the first data packet group.
  • the processor 1303 may be a CPU
  • the memory 1301 may be a RAM type internal memory
  • the transmitter 1302 may include a common physical interface, and the physical interface may be an Ethernet interface or an ATM interface.
  • the processor 1303, the transmitter 1302, and the memory 1301 may be integrated into one or more separate circuits or hardware, such as an ASIC.
  • the foregoing storage method may be at least one of the following: a read-only memory (ROM), a RAM, a magnetic disk, or an optical disk, and the like may store the program code. Medium.

Landscapes

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

Abstract

La présente invention concerne un procédé de mesure de paquets perdus et un dispositif réseau, qui comprennent les étapes suivantes : réception, par un second nœud, de premiers paquets de données envoyés par un premier nœud, les premiers paquets de données appartenant à un premier groupe de paquets de données, le premier groupe de paquets de données étant un groupe de paquets de données envoyé au second nœud par le premier nœud, les premiers paquets de données contenant un numéro de série du premier groupe de paquets de données et le nombre des premiers paquets de données, le nombre des premiers paquets de données étant utilisé afin d'identifier le nombre de paquets de données contenus dans le premier groupe de paquets de données ; réception, par le second nœud, de seconds paquets de données envoyés par le premier nœud ; lors de la détermination que les seconds paquets de données n'appartiennent pas au premier groupe de paquets de données, comptage, par le second nœud, du nombre de seconds paquets de données reçus appartenant au premier groupe de paquets de données ; et, en fonction du nombre des premiers paquets de données et du nombre des seconds paquets de données, détermination, par le second nœud, du nombre de paquets perdus du premier groupe de paquets de données. Grâce à la solution technique du mode de réalisation de la présente invention, il est bénéfique d'augmenter la précision de mesure des paquets perdus dans un cas de partage de charge.
PCT/CN2014/093824 2013-12-24 2014-12-15 Procédé de mesure de paquets perdus et dispositif réseau WO2015096636A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310722868.8A CN103684923B (zh) 2013-12-24 2013-12-24 一种丢包测量的方法及网络设备
CN201310722868.8 2013-12-24

Publications (1)

Publication Number Publication Date
WO2015096636A1 true WO2015096636A1 (fr) 2015-07-02

Family

ID=50321302

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/093824 WO2015096636A1 (fr) 2013-12-24 2014-12-15 Procédé de mesure de paquets perdus et dispositif réseau

Country Status (2)

Country Link
CN (1) CN103684923B (fr)
WO (1) WO2015096636A1 (fr)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103684923B (zh) * 2013-12-24 2017-10-17 华为技术有限公司 一种丢包测量的方法及网络设备
CN106656644B (zh) * 2015-10-28 2020-01-10 华为技术有限公司 丢包统计的方法和网络设备
CN108345621B (zh) * 2017-01-24 2022-03-08 腾讯科技（深圳）有限公司 一种数据对账方法、***及数据***
CN111371634B (zh) * 2018-12-26 2022-01-18 华为技术有限公司 一种通信方法、装置及***
CN110278058B (zh) * 2019-06-05 2020-11-03 华为技术有限公司 一种冗余信息反馈方法及通信装置
WO2021051418A1 (fr) * 2019-09-21 2021-03-25 Huawei Technologies Co., Ltd. Procédés et nœuds de réseau pour la mesure de la fiabilité
CN111158300B (zh) * 2019-11-20 2021-03-23 福建星云电子股份有限公司 一种数据采集及还原方法
CN112737884B (zh) * 2020-12-28 2022-11-11 北京达佳互联信息技术有限公司 数据的转发及检测网络性能的方法、相关装置
WO2024065550A1 (fr) * 2022-09-29 2024-04-04 北京小米移动软件有限公司 Procédé et appareil de gestion de perte de paquets

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101699786A (zh) * 2009-10-15 2010-04-28 华为技术有限公司 一种丢包检测的方法、装置和***
CN101719850A (zh) * 2009-11-04 2010-06-02 中兴通讯股份有限公司 对以太网丢包率进行统计的装置、方法及报文交换设备
CN102868576A (zh) * 2012-09-26 2013-01-09 电子科技大学 宽带网用户接入链路下行丢包率测量方法
CN102904775A (zh) * 2012-09-29 2013-01-30 华为技术有限公司 网络丢包测量方法、设备和***
CN103684923A (zh) * 2013-12-24 2014-03-26 华为技术有限公司 一种丢包测量的方法及网络设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6662223B1 (en) * 1999-07-01 2003-12-09 Cisco Technology, Inc. Protocol to coordinate network end points to measure network latency
CN102546081B (zh) * 2010-12-21 2015-09-16 中兴通讯股份有限公司 丢包检测方法、***和媒体客户端
CN102404166B (zh) * 2011-10-27 2015-04-08 华为技术有限公司 丢包检测方法和***、发送设备及接收设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101699786A (zh) * 2009-10-15 2010-04-28 华为技术有限公司 一种丢包检测的方法、装置和***
CN101719850A (zh) * 2009-11-04 2010-06-02 中兴通讯股份有限公司 对以太网丢包率进行统计的装置、方法及报文交换设备
CN102868576A (zh) * 2012-09-26 2013-01-09 电子科技大学 宽带网用户接入链路下行丢包率测量方法
CN102904775A (zh) * 2012-09-29 2013-01-30 华为技术有限公司 网络丢包测量方法、设备和***
CN103684923A (zh) * 2013-12-24 2014-03-26 华为技术有限公司 一种丢包测量的方法及网络设备

Also Published As

Publication number Publication date
CN103684923A (zh) 2014-03-26
CN103684923B (zh) 2017-10-17

Similar Documents

Publication Publication Date Title
WO2015096636A1 (fr) Procédé de mesure de paquets perdus et dispositif réseau
US10819605B2 (en) Method, apparatus, and system for implementing delay measurement
US11228515B2 (en) Data packet detection method, device, and system
KR101492510B1 (ko) 다중 전달 라우트 패킷 정렬
US7668107B2 (en) Hardware implementation of network testing and performance monitoring in a network device
US7730230B1 (en) Floating frame timing circuits for network devices
WO2019056899A1 (fr) Procédé et dispositif de transmission de message oam, et support de stockage
CN110545241B (zh) 一种报文处理方法和装置
US8270314B2 (en) Synthetic frame loss ratio
US20220407791A1 (en) Network performance detection method and apparatus, and network device
US8774008B2 (en) Real-time network measurement
CN109428782B (zh) 网络监控的方法和设备
CN103155488A (zh) 延迟测量***和延迟测量方法以及延迟测量设备和延迟测量程序
US20150036510A1 (en) Method and device for measuring ethernet performance
EP3188413B1 (fr) Procédé, appareil et système de mise en uvre de détection de perte de paquets
KR101429991B1 (ko) 클럭 게이팅을 위해 플릿들을 사용하는 기술들
US9331927B2 (en) Communication system, communication device, and communication method
US20090310508A1 (en) Real-time network measurement
WO2021057756A1 (fr) Procédé de mesure de retard, système et support de stockage
CN111371634A (zh) 一种通信方法、装置及***
EP2768187B1 (fr) Procédé pour mesurer un indicateur de performance de trafic, et dispositif sur le côté transmission
CN114157595B (zh) 一种通信***、数据处理方法以及相关设备
CN113708985B (zh) 一种流量的检测方法、装置及***
CN104243226A (zh) 流量统计方法及装置
CN112702218B (zh) 数据传输质量的检测方法、***、网络设备及存储介质

Legal Events

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

Ref document number: 14874803

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: 14874803

Country of ref document: EP

Kind code of ref document: A1