WO2015184890A1

WO2015184890A1 - Packet loss measurement method and apparatus

Info

Publication number: WO2015184890A1
Application number: PCT/CN2015/074646
Authority: WO
Inventors: 窦战伟; 郭俊; 赵俊
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-10-29
Filing date: 2015-03-19
Publication date: 2015-12-10
Also published as: CN105634840A; CN105634840B

Abstract

A packet loss measurement method and apparatus. The method includes: a first node in a TWAMP test session sends to a second node a test message recording the number of preset test messages; the second node performs received packets statistics for the received test messages, and obtains the number of the preset test messages received by the second node; the second node preprocesses the preset test messages received by the second node, obtains the number of preset test messages to be fed back to the first node, constructs a response message by adding in the test message the number of the preset test messages received by the second node and the number of the preset test messages to be fed back to the first node, and feeds back the response message to the first node; the first node performs received packets statistics for the received response message, and obtains the number of the preset test messages received by the first node; the first node acquires the number of lost packets in the TWAMP test session according to the number of the preset test messages. The above technical solution can improve the accuracy of the packet loss measurement.

Description

Method and device for measuring packet loss

Technical field

The present invention relates to the field of communications technologies, and in particular, to a method and an apparatus for measuring packet loss.

Background technique

TWAMP (Two-Way Active Measurement Protocol) is a protocol for IP performance measurement. It is mainly used for performance measurement such as IP network link delay and packet loss rate. The TWAMP protocol consists of a two-part protocol: TWAMP Control Protocol (TWAMP-Control) and TWAMP Test Protocol (TWAMP-Test). TWAMP-Control is mainly used to initialize, start and stop test sessions. TWAMP-Test is mainly used to test test packets between endpoints of the test, while measuring IP performance.

TWAMP usually consists of 4 logical entities. The simple architecture is shown in Figure 1:

● Client Control-Client: The TWAMP test initiator sends a control connection request to the server-side server entity, the communication mode of the negotiation packet, and the port number of the reflection-node Session-Reflector end to receive the test packet. The client Control-Client controls the start and end of the TWAMP-Test session.

● Server: The server receives the connection request sent from the client Control-Client, and negotiates the packet communication mode with the client Control-Client and the port number of the test node that receives the test packet from the Session-Reflector. The server side manages one or more TEAMP-Test sessions.

● The sending node Session-Sender: The node that sends the test packet to the Session-Reflector in the TWAMP-Test session, and receives the test packet from the reflection session of the reflection node, and collects the performance information and statistically measures the measurement result.

The reflection node Session-Reflector: in the TWAMP-Test session, receives the test packet from the Session-Sender end of the transmitting node, and sends a response message.

As shown in FIG. 2, in the test phase, the test node first sends a test message to the reflection node Session-Reflector, and the test message includes the sequence of message transmission. Column number and time stamp. After receiving the test packet sent by the sending node Session-Sender, the reflection node reflects the test packet back to the transmitting node Session-Sender, and adds the own receiving timestamp in the reflected test packet. The packet timestamp and the packet sequence number are as shown in FIG. 3. After receiving the test packet sent by the reflection node Session-Reflector, the sending node Session-Sender collects the packet information and performs IP performance measurement. The TTL in the figure indicates the time to live of the message.

Suppose we define the timestamp of the sending node Session-Sender as T1 in the above process, the receiving timestamp is defined as T4, the receiving timestamp of the reflecting node Session-Reflector is defined as T2, and the sending timestamp is positioned as T3, then we can pass The following method calculates the delay of the link.

Link delay = (T4-T1) - (T3-T2).

Reflector processing delay = T3-T2.

Forward delay link = T2-T1 (meaningful in the case of network clock synchronization, meaningless in other cases).

Reverse link delay = T4-T3 (meaningful in the case of network clock synchronization, meaningless in other cases).

Suppose we define the number of packets sent by the sending node Session-Sender as TxC during the test, and the number of packets sent by the receiving reflection session Session-Reflector is defined as RxC, then I can calculate the whole through TxC-RxC. The number of lost packets on the link back. Although this method can calculate the packet loss of the loopback link, the following problems exist:

1. The granularity of packet loss statistics is relatively coarse. When there is packet loss on the link, we cannot accurately determine that there is packet loss in that direction of the link.

2. It is impossible to accurately count the packet loss rate in each direction;

3. Although both the Sender-Test test message and the Reflector-Test test message carry the sequence number of the transmitted message, the two serial numbers are generated independently, and its generation depends on the transmission rules of each end. Therefore, we believe that using serial numbers to calculate the packet loss statistics in each direction is unreliable;

4. When the TWAMP protocol performs packet loss measurement, only the TWAMP test packet can be counted, and the user data cannot be counted.

Summary of the invention

The embodiment of the invention provides a method and a device for measuring packet loss, so as to achieve the purpose of accurately measuring the packet loss on the link to be tested without changing the specification and flow of the related bidirectional active measurement protocol TWAMP.

In order to achieve the above object, an embodiment of the present invention provides a method for measuring packet loss, which is applied to a test session of a bidirectional active measurement protocol TWAMP, including:

The first node in the test session of the TWAMP sends a test message with the number of preset test messages to the second node;

The second node performs the packet collection statistics on the received test packet, and obtains the number of preset test packets received by the second node;

The second node performs pre-processing on the preset test packet received by the second node, and obtains the number of preset test packets that are fed back to the first node, and receives the received by the second node. The number of the preset test packets and the number of the preset test packets fed back to the first node are added to the test message to form a response message, and the response message is fed back to the first node;

Receiving, by the first node, the response packet, and performing the packet collection statistics on the received response packet, to obtain the number of preset test packets received by the first node;

The first node acquires the test session of the TWAMP according to the number of preset test packets sent and received by the first node and the number of preset test packets received and fed back by the second node. The amount of lost packets.

The number of preset test packets sent and received by the first node and the number of preset test packets received and fed by the second node are stored in a type-length-value TLV format;

The TLV format includes a type field, a length field, and a content field, wherein the type field is set to define a type of the TLV format, the length field is set to define a length of the content field, and the content field is set to And storing the number of preset test messages sent and received by the first node and the number of preset test messages received and fed by the second node.

The content field includes a first counter configured to store the number of preset test messages sent by the first node, and a second counter configured to store the number of preset test messages received by the second node, Set to store the third number of preset test messages fed back by the second node to the first node And a counter configured to store the number of preset test messages received by the first node.

The number of the preset test packets received by the second node and the number of preset test packets to be fed back to the first node are added to the test packet to form a response packet. The step of the first node feeding back the response message includes:

The second node writes the content of the first counter of the test message into the first counter of the response message;

The second node writes the number of preset test messages received by the second node into the second counter of the response message;

The second node writes the number of preset test messages that need to be fed back to the first node into the third counter of the response message;

The second node clears the fourth counter of the response message, and feeds back the response message to the first node.

The time at which the first node measures the amount of packet loss is less than or equal to the minimum flip time of the first counter, the second counter, the third counter, and the fourth counter.

The test packet further includes: a sequence number of the test packet, a sending time of the test packet, and a padding data packet of the test packet;

The response message further includes: a time when the second node receives the test packet, a time when the second node feeds back the response message to the first node, and a sequence number of the response message.

The method further includes:

Before the first node sends a test packet to the second node,

The padding field length in the request session message in the test session of the TWAMP is set to be greater than 23 bytes.

The embodiment of the invention further includes a measurement device for packet loss, which is applied to a test session of the bidirectional active measurement protocol TWAMP, including:

a sending module, configured to send, by the first node in the test session of the TWAMP, a test message that records the number of preset test messages to the second node;

The first receiving module is configured to: the second node collects the received test packet, and obtains the number of the preset test packets received by the second node;

a feedback module, configured to: the second node pre-processes a preset test packet received by the second node, and obtains a number of preset test packets that are fed back to the first node, where the number is The number of the preset test packets received by the two nodes and the number of the preset test packets fed back to the first node are added to the test message to form a response message, and the response is fed back to the first node. Message

The second receiving module is configured to receive, by the first node, the response packet, and perform packet collection statistics on the received response packet, to obtain the number of preset test packets received by the first node. ;

Obtaining, the acquiring, by the first node, the number of the preset test packets sent and received by the first node, and the number of preset test packets received and fed by the second node, The amount of packet loss in the TWAMP test session.

The content field includes a first counter configured to store the number of preset test messages sent by the first node, and a second counter configured to store the number of preset test messages received by the second node, And a third counter configured to store the number of preset test messages fed back by the second node to the first node, and a fourth counter set to store the number of preset test messages received by the first node.

The feedback module includes:

a first writing unit, configured to: the second node writes a content of the first counter of the test message into a first counter of the response message;

The second writing unit is configured to write, by the second node, the number of preset test messages received by the second node into the second counter of the response message;

a third writing unit, configured to: the second node writes the number of the preset test messages that are to be fed back to the first node into a third counter of the response message;

And a feedback unit, configured to: the second node clears the fourth counter of the response message, and feeds back the response message to the first node.

The time at which the first node measures the amount of packet loss is less than or equal to the first count. The minimum flip time of the second counter, the third counter, and the fourth counter.

Wherein, the measuring device further comprises:

The setting module is set to set the padding field length in the request session message in the test session of the TWAMP to be greater than 23 bytes.

Embodiments of the present invention also include a computer storage medium having stored therein computer executable instructions for performing the above method.

The above technical solution has at least the following beneficial effects:

In the method and device for measuring packet loss, the test packet sent by the first node and the response message fed back by the second node are added to the preset test packet sent and received by the corresponding node. Number, to achieve accurate measurement of the packet loss on the link to be tested; this method does not need to change the relevant TWAMP protocol specifications and procedures, no need to add a new mode, only need to extract the corresponding preset test message from the response message Count, you can get the packet loss or packet loss rate in each direction, improve the accuracy of packet loss measurement.

BRIEF abstract

1 shows a simple architecture showing the TWAMP protocol in the related art;

2 shows a package format of a test message Sender-Test in an unauthenticated mode in the related art;

FIG. 3 is a diagram showing an encapsulation format of a response message Reflector-Test in a non-authentication mode in the related art;

4 is a schematic diagram showing the basic steps of a method for measuring packet loss according to an embodiment of the present invention;

FIG. 5 shows a TLV format for packet loss measurement in an embodiment of the present invention;

6 is a schematic diagram showing a process of constructing a response message in an embodiment of the present invention;

7 is a schematic diagram showing the operation of measuring a TLV in a non-authentication mode according to an embodiment of the present invention;

8 is a block diagram showing a package format of a test message Sender-Test used in a non-authentication mode according to an embodiment of the present invention;

FIG. 9 is a diagram showing an encapsulation format of a response message Reflector-Test used in a non-authentication mode according to an embodiment of the present invention;

FIG. 10 is a diagram showing a package format of a test message Sender-Test used in a non-authentication mode and a symmetric mode according to an embodiment of the present invention;

11 is a diagram showing an encapsulation format of a response message Reflector-Test used in a non-authentication mode and a symmetric mode according to an embodiment of the present invention;

Figure 12 is a view showing the basic structure of the first embodiment of the present invention;

Figure 13 is a view showing the basic structure of a second embodiment and a third embodiment of the present invention;

FIG. 14 is a block diagram showing the structure of a packet loss measuring apparatus according to an embodiment of the present invention.

Preferred embodiment of the invention

The detailed description will be made below in conjunction with the accompanying drawings and embodiments.

In the embodiment of the present invention, the packet loss statistics are relatively coarse in the related art. When there is a packet loss on the link, the problem of packet loss in the direction of the link cannot be accurately determined, and a method and a device for measuring packet loss are provided. The number of preset test packets sent and received at the corresponding node is added to the test packet sent by the first node and the response message sent by the second node to accurately measure the packet loss on the link to be tested. The purpose of this method is that the method does not need to change the relevant TWAMP protocol specifications and processes, and no need to add a new mode, only need to extract the number of corresponding preset test messages from the response message, and then obtain packet loss in each direction or The packet loss rate improves the accuracy of packet loss measurement.

As shown in FIG. 4, an embodiment of the present invention provides a method for measuring packet loss, which is applied to a test session of a bidirectional active measurement protocol TWAMP, including:

Step 41: The first node in the test session of the TWAMP sends a test packet with the number of preset test packets to the second node.

Step 42: The second node performs packet collection statistics on the received test packet, and obtains the number of preset test packets received by the second node.

Step 43: The second node performs pre-processing on the preset test packet received by the second node, and obtains the number of preset test packets fed back to the first node, where the second node is The number of the received preset test packets and the number of the preset test packets fed back to the first node are added to the test message to form a response message, and the response message is fed back to the first node. ;

Step 44: The first node receives the response packet and collects the received response message, and obtains the number of preset test packets received by the first node.

Step 45: The first node acquires the TWAMP according to the number of preset test messages sent and received by the first node, and the number of preset test messages received and fed back by the second node. The amount of lost packets in the test session.

The foregoing embodiment of the present invention is applied to a test session of the bidirectional active measurement protocol TWAMP. Optionally, the first node is a transmitting node Session-Sender, and the second node is a reflective node Session-Reflector; the test packet The method further includes: a sequence number of the test packet, a sending time of the test packet, and a padding data packet of the test packet; the response message further includes: the second node receiving the test packet The time at which the second node feeds back the response message to the first node and the sequence number of the response message. After receiving the response packet sent by the second node, the first node collects the information of the response packet and performs IP performance measurement. The technical solution adds a corresponding preset test report to both the test packet and the response packet. The number of texts is convenient for the user or the system to count the number of default test packets in different directions and in different directions, and to calculate the packet loss or packet loss rate during transmission. This method can accurately determine which link in the link. There is a packet loss in the direction, and the packet loss or packet loss rate in the direction in which the packet is lost can be accurately calculated.

The foregoing technical solution can calculate whether the sending direction (the first node to the second node) exists according to the number of preset test packets sent by the first node and the number of preset test packets received by the second node. Packet loss and packet loss; according to the number of preset test packets received by the second node and the number of preset test packets fed back to the first node by the second node, the preprocessing process of the second node can be calculated. Whether there is a packet loss or a packet loss amount; according to the number of preset test packets fed back by the second node and the number of preset test packets received by the first node, the feedback direction can be calculated (the second node to the first Whether there is a packet loss or a packet loss amount on a node; according to the number of preset test packets sent by the first node and the number of preset test packets received by the first node, a test session can be calculated. The amount of lost packets.

Optionally, as shown in FIG. 5, in the foregoing embodiment of the present invention, the number of preset test packets sent and received by the first node, and the preset test packet received and fed back by the second node. The number is stored in a type-length-value TLV format;

Optionally, the content field includes a first counter S_TxC configured to store the number of preset test messages sent by the first node, and a number configured to store the number of preset test messages received by the second node. a second counter R_RxC, a third counter R_TxC configured to store the number of preset test messages fed back by the second node to the first node, and a fourth set to store the number of preset test messages received by the first node Counter S_RxC.

Optionally, in the practical application of the embodiment of the present invention, if the receiving device of the first node has the capability of performance statistics, the fourth counter S_RxC may not be set in the content field. Optionally, the counter flipping should be considered when calculating the forward packet loss and the reverse packet loss. If a calculator is flipped, the packet loss measurement will not be performed correctly. Therefore, in the case of a given packet rate, the number of test packets and the size of the counter, it is necessary to prevent the counter from being inverted. That is, the time at which the first node measures the amount of packet loss needs to be less than or equal to the minimum flip time of the first counter, the second counter, the third counter, and the fourth counter. For example, if the packet transmission rate is 100 Gbps and the test packet size is 64 bytes, a 32 counter will be flipped within 22 seconds. Therefore, in this case, the interval of each packet loss statistics cannot be greater than 22 Seconds, the time interval is the maximum measurement interval at this rate.

Usually when performing TWAMP measurement, the measurement interval is determined in advance (usually preset configuration and less than the flip time of any counter), so it can be based on the preset test messages sent and received within the test interval. The number of bytes of the packet or packet is used to measure the capacity of the IP link to be tested.

Optionally, in the foregoing embodiment of the present invention, as shown in FIG. 6, step 43 includes:

Step 431, the second node writes the content of the first counter of the test message into the first counter of the response message;

Step 432, the second node writes the number of preset test messages received by the second node into the second counter of the response message.

Step 433, the second node writes the number of the preset test packets that need to be fed back to the first node into the third counter of the response message.

Step 434, the second node clears the fourth counter of the response message, and feeds back the response message to the first node.

In the foregoing embodiment of the present invention, the first counter, the second counter, and the third counter in the response message are all written into the corresponding number of preset test messages, and then fed back to the first node, where the first node is Writing, by the fourth counter, the number of received preset test messages, the first node can calculate the loss in any direction according to the contents of the first counter, the second counter, the third counter, and the fourth counter The amount of the package.

Optionally, in the foregoing embodiment of the present invention, before the sending, by the first node, the test packet to the second node, the method further includes:

Step 40: Set a padding field length in the request session message in the test session of the TWAMP to be greater than 23 bytes.

The first node instructs the second node to specify a minimum length of the data area in the test message in the request session message to ensure that the padding data area in the test message can accommodate the content in the TLV format. 7 is used as an example to illustrate the operation steps of the TLV in the non-authentication mode (the operation steps are the same in other modes), the node A represents the first node Session-Sender, and the node B represents the second node Session-Reflector:

Step 1: The two ends of the A and B to be tested are respectively configured to enable the function of parsing the packet loss measurement TLV, and the padding-length must be greater than 23 bytes in the request session is a request-TW-session message (testing TVL+ must have zero MBZ padding field length before testing TLV);

Step 2: At time T1, node A sends a test packet to node B (as shown in FIG. 8), and the first counter S_TxC counter of the test packet TLV carries the number of preset test packets that node A has sent. A_TxC[n], the other three counters are set to zero;

Step 3: At time T2, after receiving the test packet sent by A, the B end performs the packet collection statistics in this paragraph;

Step 4: At time T3, the B end will receive the S_TxC[n] counter in the test message TLV. From the B to the B-end response message (as shown in Figure 9), the S_TxC counter of the TLV copies the B-side packet statistics R_RxC[n] to the R_RxC counter that should be sent to the message TLV, and the B-side reflection message statistics R_TxC [n] copy to the R_TxC count of the response message TLV, wherein the S_RxC counter is cleared, and then B transmits the response message to the A end;

Step 5: At time T4, after receiving the response message, the A end performs the local receiving and collecting statistics, and simultaneously copies the collected receiving data S_RxC[n] to the S_RxC counter in the message TLV (if A The end collection device itself has the capability of performance statistics, and the S_RxC counter may not be set);

In step 6, on the A side, the performance statistics module can perform performance measurement according to the message counter of the response message TLV (including forward packet loss measurement, reverse packet loss measurement, forward packet loss rate measurement, and reverse loss). The packet rate measurement and the corresponding packet loss rate jitter measurement can also be performed under the premise of a given packet rate and measurement interval, forward path capacity measurement and reverse path capacity measurement).

According to the above steps, you can use the following algorithm to perform packet loss and capacity measurement:

Forward packet loss: TxLoss[n-1,n]=(A_TxC[n]-A_TxC[n-1])-(B_RxC[n]-B_RxC[n-1]);

Reverse packet loss: RxLoss[n-1,n]=(B_TxC[n]-B_TxC[n-1])-(A_RxC[n]-A_RxC[n-1]);

Forward packet loss rate: TxLossRadio[n-1,n]=TxLoss[n-1,n]/(A_TxC[n]-A_TxC[n-1]);

Reverse packet loss rate: RxLossRadio[n-1,n]=RxLoss[n-1,n]/(B_TxC[n]-B_TxC[n-1]). n refers to the nth packet loss statistics.

It should be noted that, in other modes, the operation steps of the TLV are the same as those in the non-authentication mode; for example, FIG. 10 shows the test message used in the embodiment of the present invention in the non-authentication mode and the symmetric mode. The encapsulation format of the Sender-Test; FIG. 11 shows the encapsulation format of the response message Reflector-Test used in the embodiment of the present invention in the non-authentication mode and the symmetric mode. Other modes, such as non-authentication mode, symmetric mode, reflection byte (Reflect Octets) mode test message Sender-Test and response message Reflector-Test package format are not listed here.

In order to better describe the measurement method of the embodiment of the present invention, the following details are combined with the application scenario. Description, as shown in FIG. 12 is a first embodiment of the present invention, which is an application for measuring an IP network drop on demand by using an TWAMP protocol in the form of an additional statistical TLV. In this example, the TWAMP is only performed. The TWAMP test packet of the control panel is counted and the user packet is not counted. The router RouterA acts as the Control-Client and the Session-Sender, and the RouterB serves as the Server and the Session-Reflector.

Step 11: First, the ability to parse the packet loss statistics TLV is enabled on RouterA and RouterB, and A initiates the establishment of the TWAMP-Control connection and the test session on the link.

Step 12: After the test command is sent on RouterA, the TWAMP process on RouterA sends the test packet to the routerA in the above manner.

In Step 13, RouterB also reflects the test packet to the A end in the foregoing implementation manner.

Step 14: After receiving the packet reflected by RouterB, RouterA performs packet statistics and performs packet loss measurement according to the above algorithm.

FIG. 13 shows a second embodiment of the present invention, which is an application for actively measuring network packet loss in the form of an additional statistical TLV using the TWAMP protocol. In the implementation, hardware participation is required. For example, the performance statistics device needs to collect TWAMP test packets and user packets at the same time, and the case is only applicable to in-band packet loss measurement, including:

Step 21 is the same as the first embodiment. Router A and Router B need to enable the ability to parse the statistical TLV.

Step 22: After RouterA initiates the test command, the hardware device (for example, the field programmable gate array FPGA) sends a test packet to RouterB according to the configured rate and the packet size on the data plane.

Step 23: After receiving the test packet sent by RouterA, the RouterB hardware device sends a test packet to RouterA and then sends a test packet to RouterA.

Step 24: After receiving the packet reflected by RouterB, RouterA performs packet statistics and performs packet loss measurement according to the above algorithm.

As shown in FIG. 13, the third embodiment is an application for performing real-time monitoring of link loss on the basis of the second embodiment, including:

Step 31: First, configure a forward packet loss rate threshold and a reverse packet loss on RouterA. Rate threshold

Step 32: Implement the steps of the second embodiment on RouterA and RouterB.

Step 33: On Router A, the packet loss rate in the two directions on the monitoring link is implemented. When the packet loss rate in the corresponding direction exceeds the configured packet loss threshold, the corresponding alarm is reported to prompt the user to perform corresponding operations.

In order to achieve the above objective, as shown in FIG. 14, the embodiment of the present invention further provides a packet loss measurement apparatus, which is applied to a test session of a bidirectional active measurement protocol TWAMP, including:

The sending module 71 is configured to send, by the first node in the test session of the TWAMP, a test message that records the number of preset test messages to the second node;

The first receiving module 72 is configured to receive, by the second node, the packet statistics of the received test packet, to obtain the number of preset test packets received by the second node;

The feedback module 73 is configured to preprocess the preset test packet received by the second node by the second node, and obtain the number of preset test packets that are fed back to the first node, and receive the second node. The number of the preset test packets and the number of the preset test packets that are fed back to the first node are added to the test message to form a response message, and the response message is fed back to the first node;

The second receiving module 74 is configured to receive, by the first node, the response packet, and perform packet collection statistics on the received response packet, to obtain a preset test packet received by the first node. number;

The obtaining module 75 is configured to obtain, according to the number of preset test packets sent and received by the first node, and the number of preset test packets received and fed by the second node, The amount of packet loss in the TWAMP test session.

In an optional embodiment of the present invention, the number of preset test packets sent and received by the first node and the number of preset test packets received and fed by the second node are type-length- Value TLV format storage;

The TLV format includes a type field, a length field, and a content field, where the type field is used to define a type of the TLV format, the length field is used to define a length of the content field, and the content field is used to And storing the number of preset test messages sent and received by the first node and the number of preset test messages received and fed by the second node.

In an optional embodiment of the present invention, the content field includes a first counter that is configured to store the number of preset test packets sent by the first node, and is configured to store the preset test packet received by the second node. a second counter of the number, a third counter configured to store the number of preset test messages fed back by the second node to the first node, and a number of preset test messages received by the first node The fourth counter.

In the foregoing embodiment of the present invention, the feedback module 73 includes:

In an optional embodiment of the present invention, the time that the first node measures the amount of packet loss is less than or equal to the minimum flip time of the first counter, the second counter, the third counter, and the fourth counter.

In an optional embodiment of the present invention, the test packet further includes: a sequence number of the test packet, a sending time of the test packet, and a padding data packet of the test packet;

In an optional embodiment of the present invention, the measuring device further includes:

In the method and device for measuring packet loss, the test packet sent by the first node and the response message fed back by the second node are added to the preset test packet sent and received by the corresponding node. Number, to achieve accurate measurement of packet loss on the link under test; this method does not need to change the relevant TWAMP protocol specifications and procedures, no need to add new modes, only need to extract corresponding from the response message The number of preset test packets can be used to obtain packet loss or packet loss rate in each direction, which improves the accuracy of packet loss measurement.

It should be noted that the device for measuring packet loss provided by the embodiment of the present invention is a device applying the above measurement method, and the above measurement method is applicable to the measurement device in all embodiments, and all of the same or similar beneficial effects can be achieved.

The above is a preferred embodiment of the present invention, and it should be noted that those skilled in the art can also make several improvements and retouchings without departing from the principles of the present invention. It should also be considered as the scope of protection of the present invention.

One of ordinary skill in the art will appreciate that all or a portion of the steps of the above-described embodiments can be implemented using a computer program flow, which can be stored in a computer readable storage medium, such as on a corresponding hardware platform (eg, The system, device, device, device, etc. are executed, and when executed, include one or a combination of the steps of the method embodiments.

Alternatively, all or part of the steps of the above embodiments may also be implemented by using an integrated circuit. These steps may be separately fabricated into individual integrated circuit modules, or multiple modules or steps may be fabricated into a single integrated circuit module. achieve.

The devices/function modules/functional units in the above embodiments may be implemented by a general-purpose computing device, which may be centralized on a single computing device or distributed over a network of multiple computing devices.

When each device/function module/functional unit in the above embodiment is implemented in the form of a software function module and sold or used as a stand-alone product, it can be stored in a computer readable storage medium. The above mentioned computer readable storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Industrial applicability

The above technical solution does not need to change the relevant TWAMP protocol specifications and processes, and can obtain packet loss or packet loss rate in each direction without adding a new mode, thereby improving the accuracy of packet loss measurement.

Claims

A method of measuring packet loss, including:

The first node in the test session of the two-way active measurement protocol TWAMP sends a test message with the number of preset test messages to the second node;

The second node performs the packet collection statistics on the received test packet, and obtains the number of preset test packets received by the second node;

The second node performs pre-processing on the preset test packet received by the second node, and obtains the number of preset test packets that are fed back to the first node, and receives the received by the second node. The number of the preset test packets and the number of the preset test packets fed back to the first node are added to the test message to form a response message, and the response message is fed back to the first node;

Receiving, by the first node, the response packet, and performing the packet collection statistics on the received response packet, to obtain the number of preset test packets received by the first node;

The first node acquires the test session of the TWAMP according to the number of preset test packets sent and received by the first node and the number of preset test packets received and fed back by the second node. The amount of lost packets.
The measurement method according to claim 1, wherein the number of preset test messages sent and received by the first node and the number of preset test messages received and fed back by the second node are of a type- Length-value TLV format storage;

The TLV format includes a type field, a length field, and a content field, wherein the type field is set to define a type of the TLV format, the length field is set to define a length of the content field, and the content field is set to And storing the number of preset test messages sent and received by the first node and the number of preset test messages received and fed by the second node.
The measurement method according to claim 2, wherein the content field comprises a first counter configured to store a number of preset test messages sent by the first node, and configured to store a preset test received by the second node. a second counter of the number of the packets, a third counter configured to store the number of preset test messages fed back by the second node to the first node, and a preset to store the preset test message received by the first node The fourth counter of the number.
The measuring method according to claim 3, wherein said receiving said second node The number of the preset test packets and the number of the preset test packets to be fed back to the first node are added to the test message to form a response message, and the response message is fed back to the first node. The steps include:

The second node writes the content of the first counter of the test message into the first counter of the response message;

The second node writes the number of preset test messages received by the second node into the second counter of the response message;

The second node writes the number of preset test messages that need to be fed back to the first node into the third counter of the response message;

The second node clears the fourth counter of the response message, and feeds back the response message to the first node.
The measuring method according to claim 4, wherein the time at which the first node measures the amount of packet loss is less than or equal to the minimum flip time of the first counter, the second counter, the third counter, and the fourth counter.
The measuring method according to claim 1,

The test packet further includes: a sequence number of the test packet, a sending time of the test packet, and a padding data packet of the test packet;

The response message further includes: a time when the second node receives the test packet, a time when the second node feeds back the response message to the first node, and a sequence number of the response message.
The measuring method according to claim 1, further comprising:

Before the first node sends a test packet to the second node,

The padding field length in the request session message in the test session of the TWAMP is set to be greater than 23 bytes.
A measuring device for packet loss, comprising:

The sending module, the first node in the test session set to the bidirectional active measurement protocol TWAMP, sends a test message recording the number of preset test messages to the second node;

The first receiving module is configured to: the second node collects the received test packet, and obtains the number of the preset test packets received by the second node;

a feedback module, configured to: the second node pre-processes a preset test packet received by the second node, and obtains a number of preset test packets that are fed back to the first node, where the number is Two The number of the preset test packets received by the node and the number of the preset test packets fed back to the first node are added to the test message to form a response message, and the response message is fed back to the first node. Text

The second receiving module is configured to receive, by the first node, the response packet, and perform packet collection statistics on the received response packet, to obtain the number of preset test packets received by the first node. ;

Obtaining, the acquiring, by the first node, the number of the preset test packets sent and received by the first node, and the number of preset test packets received and fed by the second node, The amount of packet loss in the TWAMP test session.
The measuring device according to claim 8, wherein the number of preset test messages sent and received by the first node and the number of preset test messages received and fed back by the second node are of a type - Length-value TLV format storage;

The TLV format includes a type field, a length field, and a content field, wherein the type field is set to define a type of the TLV format, the length field is set to define a length of the content field, and the content field is set to And storing the number of preset test messages sent and received by the first node and the number of preset test messages received and fed by the second node.
The measurement device according to claim 9, wherein the content field comprises a first counter configured to store a number of preset test messages sent by the first node, and configured to store a preset test received by the second node. a second counter of the number of the packets, a third counter configured to store the number of preset test messages fed back by the second node to the first node, and a preset to store the preset test message received by the first node The fourth counter of the number.
The measuring device according to claim 10, wherein the feedback module comprises:

a first writing unit, configured to: the second node writes a content of the first counter of the test message into a first counter of the response message;

The second writing unit is configured to write, by the second node, the number of preset test messages received by the second node into the second counter of the response message;

a third writing unit, configured to: the second node writes the number of preset test messages to be fed back to the first node into a third counter of the response message;

And a feedback unit, configured to: the second node clears the fourth counter of the response message, and feeds back the response message to the first node.
The measuring device according to claim 11, wherein said first node measures one time lost The amount of time of the packet is less than or equal to the minimum flip time of the first counter, the second counter, the third counter, and the fourth counter.
The measuring device according to claim 8,

The test packet further includes: a sequence number of the test packet, a sending time of the test packet, and a padding data packet of the test packet;

The response message further includes: a time when the second node receives the test packet, a time when the second node feeds back the response message to the first node, and a sequence number of the response message.
The measuring device according to claim 8, further comprising:

The setting module is set to set the padding field length in the request session message in the test session of the TWAMP to be greater than 23 bytes.
A computer storage medium having stored therein computer executable instructions for performing the method of any one of claims 1-7.