WO2021075683A1 - System, device, and method for measuring delay time of service function - Google Patents

Abstract

The present invention relates to a system, device, and method for measuring a delay time of a service function, and a delay time measurement device may comprise: a service function unit for receiving a packet, processing a service function for the packet, and outputting the processed packet; and a transfer processing unit for transferring a packet to the service function unit, recording a first transmission time at which the packet is transferred to the service function unit, and recording a first reception time at which the processed packet is received from the service function unit, wherein the service function unit records a second reception time at which the packet is received, and records a second transmission time at which the service function-processed packet is transferred to the transfer processing unit, and a processing delay time for the packet is calculated on the basis of the second reception time and the second transmission time.

Description

System, apparatus and method for measuring latency of service functions

The present invention relates to a system, apparatus, and method for measuring delay time of a service function (SF).

When an external terminal device is communicatively connected with the network device and transmits at least one packet to the network device, the network device processes at least one function for the received packet and provides the network service requested by the terminal device to the terminal. It can be provided to the device. Here, the function that the network device processes for at least one packet is referred to as a service function (SF can be referred to as a network service function).

Service Function Chaining (SFC) technology (which can also be referred to as Network Service Chaining technology) is a combination of multiple service functions. It refers to the technology to be processed accordingly. That is, the service function chaining technology classifies packets according to the network policy and allows a group of service functions to be sequentially processed for them.

Conventionally, in order to determine whether the processing delay of each service function under the service function chaining (ie, whether there is a failure), a detection packet is used periodically. In detail, a separately prepared controller or monitoring engine periodically generates detection packets and delivers them to the service function, and the corresponding service function transmits feedback on whether or not there is a failure to the controller or engine in response, and receives the feedback received by the controller or engine. Based on the determination of the failure, the delay in processing was diagnosed. Such a monitoring method is commonly referred to as out-of-band monitoring. However, out-of-band monitoring has a problem of causing network congestion because packets for monitoring are periodically generated. In addition, there is a problem that detailed monitoring becomes difficult because it is impossible to measure the processing delay time of the service function in this manner.

An object of the present invention is to provide a system, apparatus, and method for measuring a delay time of a service function capable of monitoring at least one network service function without additional bandwidth, and to grasp a specific situation in more detail when processing is delayed.

In order to solve the above problems, a system, apparatus, and method for measuring delay time of a service function are provided.

The delay time measurement system receives a packet, processes a service function on the packet, delivers the packet to the service function unit and the service function unit, which outputs the processed packet, and delivers the packet to the service function unit. And a transfer processing unit for recording a first transmission time and recording a first reception time at which the processed packet is received from the service function unit, wherein the service function unit records a second reception time at which the packet is received, The second transmission time of transmitting the packet processed by the service function to the transfer processing unit is recorded, and the processing delay time for the packet may be calculated based on the second reception time and the second transmission time.

The delay time measurement method includes the steps of: a delivery processing unit delivering a packet to a service function unit and recording a first transmission time at which the packet is delivered, a service function unit receiving a packet, and determining a second reception time at which the packet is received. Recording, the service function transferring the packet processed with the service function to the transfer processing unit, and recording a second transmission time when the packet processed with the service function is transferred to the transfer processing unit, the transfer processing unit Receiving the processed packet from the functional unit, recording a first reception time at which the processed packet was received, and calculating a processing delay time for the packet based on the second reception time and the second transmission time It may include the step of.

The delay time measurement apparatus includes a time recording unit for recording a first transmission time for transmitting a packet to a service function unit and a first reception time for receiving a packet from the service function unit, and a second reception at which the service function recorded in the packet receives the packet. A delay processing unit that reads out a second transmission time at which the time and the service function unit transmits the packing, and calculates a processing delay time by using a difference between the second transmission time and the second reception time.

According to the above-described service function delay time measurement system, apparatus, and method, it is possible to more appropriately monitor whether or not a delay or failure of at least one service function occurs without additional bandwidth.

In addition, it is possible to confirm and analyze in more detail where the processing delay of at least one service function has occurred, so that the cause of the delay can be accurately determined.

In addition, in the service function chaining technology, it is possible to measure the delay time of the service function or determine whether a failure occurs without generating and inputting a separate detection packet, so that separate hardware or software such as a controller or monitoring engine for monitoring can be built. In addition to eliminating the need, network congestion such as bandwidth overhead due to detection packets can be prevented in advance, so that the efficiency and economics of monitoring network devices can be reconsidered.

1 is a block diagram of an embodiment of a delay time measurement system.

2 is a diagram for explaining an embodiment of an operation of a delivery processing unit and a service function unit.

3 is a block diagram of an embodiment of a network service header.

4 is a first diagram for explaining an embodiment of an operation of a delay processing unit.

5 is a second diagram for explaining an embodiment of an operation of a delay processing unit.

6 is a first flowchart of an embodiment of a method for measuring a delay time of a service function.

7 is a second flowchart of an embodiment of a method for measuring a delay time of a service function.

8 is a third flowchart of an embodiment of a method for measuring a delay time of a service function.

9 is a flowchart of an embodiment of an operation of a transfer processing unit.

In the following specification, the same reference numerals refer to the same elements unless otherwise specified. The term "unit" used below may be implemented in software or hardware, and according to an embodiment, one "unit" is implemented as one physical or logical part, or a plurality of "units" are implemented as one It may be implemented as a physical or logical part, or one'unit' may be implemented as a plurality of physical or logical parts.

When a part is said to be connected to another part throughout the specification, it may mean a physical connection depending on the part and another part, or it may mean that it is electrically connected. In addition, when a part includes another part, this does not exclude another part other than the other part unless specifically stated to the contrary, and it means that another part may be further included according to the designer's choice. do.

Terms such as first or second are used to distinguish one part from another, and unless otherwise specified, they do not mean sequential expressions. In addition, expressions in the singular may include plural expressions unless there is a clear exception in the context.

Hereinafter, an embodiment of a system for measuring a delay time of a service function and a device for measuring a delay time will be described with reference to FIGS. 1 to 5.

1 is a block diagram of an embodiment of a delay time measurement system.

1, the delay time measurement system 100 of a predetermined network (for example, a software defined network (SDN)) acquires a packet 20 from an external device 10 and receives a packet ( It is possible to process and perform the service function for 20) and/or monitor the operation and failure of the service function. More specifically, the packet 20 is received from the external device 10, and the received network It may be provided to classify the packets 20 into one or more according to a policy, and to sequentially or randomly process network service functions for all or part of the classified at least one packet 20.

According to an embodiment, the delay time measurement system 100 includes a control unit 110 that manages or controls the overall operation of the delay time measurement system 100, and a service processing unit that processes a service function for each packet 20 ( 120). The control unit 110 and the service processing unit 120 may be physically separated. In this case, the control unit 110 and the service processing unit 120 may be implemented using at least one physically separated processor or at least one computing device. For example, the control unit 110 may be implemented as a computing device or a server named as a controller of SDN, and each component of the service processing unit 120 may be implemented as a node that performs a predetermined operation. have.

The controller 110 may control the packet 20 and/or control or manage the service processing unit 120. For example, the control unit 110 may designate or change a setting for the packet 20 and/or transmit a corresponding response to the service processing unit 120 in response to a request from the service processing unit 120. More specifically, for example, in the former case, the control unit 110 uses a method such as time stamping on the input packet 20 to transmit the transmission time or the reception time (t11, t12 in FIG. 2, It can also be set to record t21, t22), and the like. In the latter case, the control unit 110 generates a processing rule (112 in FIG. 5) requested by the service processing unit 120 or a separately provided storage medium (for example, a semiconductor memory device or a magnetic disk storage device provided in the control unit 110). ), and/or modifying an existing processing rule to obtain a new processing rule 112, and then transmitting it to the service processing unit 120. The control unit 110 may also be referred to as a control plane.

The service processing unit 120 may process at least one service function for each packet 20 transmitted by the external device 10. According to an embodiment, the service processing unit 120 may be implemented based on a service function chaining technology as needed. In this case, the service processing unit 120 may be referred to as a service function chaining domain (SFC domain). In addition, the service processing unit 120 may be implemented as a data plane. Depending on the embodiment, the service processing unit 120 may be implemented based on a programmable data plane (PDP).

The service processing unit 120 includes a service classification unit (SC: Service Classifier) 121, a first to nth forwarding processing unit (SFF: Service Function Forwarder, 130-1 to 130-n, n is a natural number of 1 or more), and a first To n-th service function units 135-1 to 135-n. In FIG. 1, the first to nth service functional units 135-1 to 135-n are shown to be provided corresponding to each of the first to nth transfer processing units 130-1 to 130-n. Depending on the situation, it may be provided in response to a many-to-one or one-to-many relationship. For example, the first delivery processing unit 130-1 corresponds to the first service function unit 135-1 to the kth service function unit 135-k, where k is an arbitrary natural number greater than 1 and less than or equal to n. Packets may be delivered to a corresponding service function unit so that a predetermined service function is performed.

All or part of the service classification unit 121, the first to nth delivery processing units 130-1 to 130-n, and the first to nth service function units 135-1 to 135-n, respectively, are physically It may be distinguished, and/or may be logically divided. In other words, the service processing unit 120 may be implemented using a plurality of physical devices. In this case, each of the service classification unit 121, the first to nth transfer processing units 130-1 to 130-n, and the first to nth service function units 135-1 to 135-n is at least one It can be implemented using a physical device. Here, the physical device may include a server computer device, a router, an Ethernet hub device, a gateway device, a desktop computer, a laptop computer, a smartphone, a tablet PC, or a set-top box, but is not limited thereto, and various It may include an electronic device. More specifically, for example, the service classification unit 121, the first to nth transfer processing units 130-1 to 130-n, and the first to nth service function units 135-1 to 135-n are Each may be implemented using separate communication equipment. In this case, the service classification unit 121 is implemented by a switch or a router, and the first to nth transfer processing units 130-1 to 130-n and the Each of the first to nth service function units 135-1 to 135-n may be implemented by a separate server computer device. Depending on the embodiment, any one delivery processing unit (130-1 to 130-n) and any one or two or more service processing units (135-1 to 135-n) corresponding thereto are It is also possible to implement based on a computer device, etc.).

The packet 20 transmitted by the external device 10 may be sequentially or non-sequentially delivered to at least one of the first to nth transfer processing units 130-1 to 130-n through the service classification unit 121. have. If necessary, the packet 20 is transmitted to at least one of the first to nth service function units 135-1 to 135-n, and then at least one of the service processing units 130-2 to 130-n in the next order. ). In this case, the packet 20 may include an encapsulation packet (30 in FIG. 2). Specifically, the transfer processing units 130-1 to 130-n determine whether to transfer the packets 20 and 30 to the service function units 135-1 to 135-n, and transmit the packets 20 and 30 according to the determination. It may be transmitted to the service function units 135-1 to 135-n and/or to the delivery processing units 130-2 to 130-n or the external device 90 in the following order. Operations and functions of the service classification unit 121, the delivery processing units 130-1 to 130-n, and the service function units 135-1 to 135-n will be described later.

The external devices 10 and 90 transmit the packet 20 to the service processing unit 120 of the delay time measurement system 100 and/or the service processing from the service processing unit 120. Can receive packets. In this case, the external device 10 transmitting the packet and the external devices 10 and 90 receiving the processed packet may be the same or different from each other. The external devices 10 and 90 may be physical media or virtual media. In the case of a physical medium, the external devices 10 and 90 may include at least one of a wired communication network and a wireless communication network (which may include mobile communication and/or may include short-range communication such as Wi-Fi or Bluetooth). It is possible to include an electronic device provided to be communicatively connected to the service processing unit 120 through the transmission and reception of the packet 20, wherein the electronic device is, for example, a smartphone, a tablet PC, a desktop computer, and a laptop. It may include a computer, a server computer, a home appliance, a robot, a vehicle, a black box, a wearable device, a mechanical device, and/or a construction machine.

Each configuration of the service processing unit 120 shown in FIG. 1, that is, a service classification unit 121, a first to nth transfer processing unit 130-n, and a first to nth service function unit and an external device 10, 90) Each can also be expressed as a node. In this case, the external device 10 is a source node, a service classifying unit is a service classifier (SC) node, a forwarding processing unit is a service function forwarder (SFF) node, a service function unit is a service function (SF) node, and an external device. Reference numeral 90 may be referred to as a destination node.

2 is a diagram for explaining an embodiment of an operation of a delivery processing unit and a service function unit. 2 shows only one delivery processing unit 130 and one service function unit 135 corresponding thereto, but this is for convenience of description, and does not exclude the presence of other delivery processing units and other service function units. Other transfer processing units of FIG. 1, for example, second to nth transfer processing units 130-2 to 130-n, and corresponding second to nth service functions 135-2 to 135-n. ) Also, it may be designed to operate in the same manner as described below or through a partially modified method.

The service classification unit 121 may receive the packet 20 and classify the received packet 20 according to a predefined classification rule. In addition, the service classification unit 121 may decrypt the received packet 20 or further encrypt the packet to be transmitted, if necessary.

According to an embodiment, the service classification unit 121 performs encapsulation on the packet 20 by adding a network service header 31 to all or part of the received packet 20 according to the classification result. By doing so, the network service header 31 is added to obtain an encapsulated packet 30 (hereinafter, an encapsulated packet). Accordingly, the identifier can be marked on the packet in the traffic flow. The encapsulation packet 30 may be delivered to at least one of the at least one delivery processing unit 130-1 to 130-n. For example, the encapsulation packet 30 is initially transferred to the first transfer processing unit 130-1, and then sequentially or non-sequentially, the second transfer unit to the n-th transfer unit 130-1 to 130-n. It may be delivered to at least one of.

3 is a block diagram of an embodiment of a network service header.

Referring to FIG. 3, the network service header 31 may include a basic header 32, a service path header 34, and a context header 36. The basic header 32 may include various types of information related to the network service header 31. For example, the basic header 32 is the version of the network service header 31, the type of the packet 20, the maximum number of transfer processing units 130-1 to 130-n, and the total length of the network service header 31 , Meta data type and/or protocol type of encapsulated data may be included. The service path header 34 includes a service path identifier 34-1, which is information for identifying a service function path (SFP) for processing the received packet 20, and a service index for identifying a service function to be processed ( 34-2). The service index 34-2 may include a number of 1 or more. In this case, the service index 34-2 is decreased by 1 each time the service function unit 135 processes the packet 30, so that it is possible to check how many times the service function has been processed for the packet 30. I can do it. Subtraction of the service index 34-2 may be performed by the service function unit 135. At least one metadata such as context data may be marked and recorded in the context header 36. Depending on the embodiment, the context header 36 may be a Fixed-Length Context Header, a Variable-Length Context Header, or Variable-Length Metadata. have. According to an embodiment, the context header 36 may be marked with a second reception time t22, which is a time when the service function unit 135 receives the packet 32. Depending on the situation, at least one of the basic header 32, the service path header 34, and the context header 36 may be omitted.

The service classification unit 121 may also remove the added network service header 31. Specifically, a packet encapsulated by the first to nth transfer processing units 130-1 to 130-n and the corresponding first to nth service functions 135-1 to 135-n. When all the scheduled service functions for (30) are processed, the packet may be transmitted to the external devices 10 and 90 of the packet 30, and the service classifying unit 121 transmits a packet to the external devices 10 and 90. It is also possible to remove the network service header 31 from the packet 30 before it is delivered. Depending on the embodiment, it is also possible for elements other than the service classification unit 121 to remove the network service header 31. For example, the n-th transfer processing unit 130-n may remove the network service header 31, and another network device physically or logically separately provided may remove the network service header 31.

As shown in FIG. 2, the transfer processing unit 130 may include a selection unit 131 and a first time recording unit 132, and may further include a delay processing unit 140 if necessary.

The selection unit 131 determines whether to transmit the encapsulation packet 30 to the corresponding service function unit 135 based on the information in the network service header 31 of the received encapsulation packet 30, and Accordingly, the packet 30 can be delivered to the service function unit 135. Depending on the situation, the selection unit 131 may transmit the encapsulation packet 30 to other transfer processing units 130-2 to 130-n according to the information of the network service header 31. In this case, the transfer of the packet 30 to the other transfer processing units 130-2 to 130-n may be performed regardless of the transfer of the packet 30 to the service function unit 135, or the service function unit 135 It may be performed in conjunction with the delivery of the packet 30 to. In the latter case, the transfer of the packet 30 to the other transfer processing units 130-2 to 130-n may be performed simultaneously with the transfer of the packet 30 to the service function unit 135 or may be performed sequentially.

According to an embodiment, the selection unit 131 may determine whether the received packets 30 and 32 are packets transmitted from the service classification unit 121 or returned via the service function unit 135. . That is, the selection unit 131 determines whether the received packet 30, 32 is the first packet 30 or the previously received packet 32, and as a result of the determination, the first received packet 30 In the case of, the packet 30 is transferred to the service function unit 135 and/or to the next transfer processing unit 130-2 to 130-n, and in the case of the previously received packet 32, the packet 32 is transferred to the next It can be transferred to the transfer processing units 130-2 to 130-n.

When the encapsulation packet 30 is transmitted to the service function unit 135 by the selection unit 131, the first time recording unit 132 is the time at which the packet 30 is transmitted to the service function unit 135 ( t11, hereinafter, the first transmission time) may be determined and recorded temporarily or non-temporarily. In this case, the first time recording unit 132 may store the first transmission time t1 in the storage units 150 and 151, or may record it in the network service header 31, for example, the context header 36. have. According to an embodiment, the storage unit 150 may be provided in the transfer processing unit 130. For example, the storage unit 150 may include a semiconductor memory device or a magnetic disk storage device mounted on the transfer processing unit 130 implemented as a programmable switch. In addition, the storage unit 151 may be provided separately from the transfer processing unit 130. In this case, the storage unit 151 may include a computer device or an external storage medium (such as a universal serial bus memory device or an external hard disk device) separately constructed for storing data. According to an embodiment, the first time recording unit 132 stores the first transmission time t11 in the storage units 150 and 151 and/or the network service header 31 using a time stamping function of a programmable data plane. , For example, it can be recorded in the context header 36.

In addition, the first time recording unit 132, when receiving the processed packet 32 from the service function unit 135, stores the reception time (t12, hereinafter, the first reception time) of the processed packet 32 ( It may be temporarily or non-temporarily stored in at least one of 150 and 151 and the network service header 31. Even in this case, the first time recording unit 132 may store the first reception time t12 in the storage units 150 and 151 and/or the network service header 31 using a time stamping function of a programmable data plane. .

The delay processing unit 140 may determine whether a service processing is delayed or a failure occurs. This will be described later.

The service function unit 135 may process a predefined service function for the encapsulation packet 30 and transmit the packet 32 (hereinafter, processed packet) in which the service function is processed to the transfer processing unit 130 again. Here, the service functions provided by the service function unit 135 are firewall, packet filtering, deep packet inspection (DPI), network address translation (NAT), transmission control protocol (TCP) optimization, and intrusion. Detection, load balancer, or virtualization may be included, but are not limited thereto.

According to an embodiment, the service function unit 135 may include a second time recording unit 136 for recording time. The second time recording unit 136 stores the time at which the encapsulation packet 30 is received (t22, hereinafter, the second reception time), a network service header 31, a storage unit 150 of the transfer processing unit 130, and a storage unit provided separately. It can be recorded in at least one of (151). For example, if the received packet 30 is a packet set to perform time stamping by the control unit 110, the second time recording unit 136 uses the time stamping function to determine the second reception time t22 to the network service. It may be marked and recorded on the context header 36 of the header 31 and/or marked on the storage units 150 and 151 and stored. In addition, when the service function processing is terminated and the packet 32 processed by the delivery processing unit 30 is transmitted, the second time recording unit 136 stores the transmitted time (t21, hereinafter the second transmission time) in the network service header. (31), it may be recorded in at least one of the storage unit 150 of the transfer processing unit 130 and the storage unit 151 provided separately. For example, the second time recording unit 136 stores the second reception time t22 in at least one of the context header 36 and the storage units 150 and 151 of the network service header 31 by using a time stamping function. It can be marked and recorded. At least one of the second reception time t22 and the second transmission time t21 may be temporarily stored or may be stored non-temporarily.

FIG. 4 is a first diagram illustrating an operation of a delay processing unit according to an exemplary embodiment, and FIG. 5 is a second diagram illustrating an operation of a delay processing unit according to an exemplary embodiment.

4 and 5, the delay processing unit 140 of the delivery processing unit 130 includes a time recorded in at least one of the storage units 150 and 151 and the network service header 31 (t11 to t22). The delay times d1 to d3 are calculated and measured based on the values, and it is possible to determine whether a service processing is delayed or a failure occurs based on the calculation and measurement results. By such an operation of the delay processing unit 140, the transfer processing unit 130 can operate as a delay time measuring device.

According to an embodiment, the delay time measurement system 100 may include a delay processing unit 145 provided separately from the transmission processing unit 130, as shown in FIG. 4. In this case, the delay processing unit 140 of the transfer processing unit 130 may be omitted. The delay processing unit 145 may be physically separated from the transfer processing unit 130 and may be implemented using, for example, a computing device. Depending on the embodiment, the delay processing unit 145 provided separately may be implemented by the above-described control unit 110 or the service processing unit 120. In addition, the delay processing unit 145 provided separately may be logically separated from the transfer processing unit 130. The delay processing unit 145 provided separately from the transmission processing unit 130 may operate in the same manner as the operation of the delay processing unit 140 of the transmission processing unit 130 or partially modified as described below.

According to an embodiment, the delay processing unit 140 may include a delay time calculating unit 141, a comparison unit 143, and a result processing unit 143, as shown in FIG. 5.

First, the delay time calculator 141 may obtain at least two of the first to fourth transmission times t11, t12, t21, and t22. More specifically, for example, the delay time calculation unit 141 calls the first transmission time t11 and the first reception time t12 from the storage units 150 and 151, and the network service header 31, for example, The second transmission time t21 and the second reception time t22 recorded in the raw context header 36 may be read. Depending on the embodiment, the delay time calculating unit 141 may call all of these times (t11 to t22) from the storage units 150 and 151, or all of these times (t11 to t22) may be used in the network service header 31 It is also possible to read from ).

The delay time calculating unit 141 sequentially performs an operation based on at least two of the acquired first to fourth transmission times (t11, t12, t21, t22), and the round-trip delay (d1) ), processing delay (d2), and transmission delay (Propagation Delay, d3).

The round trip delay time d1 refers to a time required for the packet 30 transmitted from the transfer processing unit 130 to return to the corresponding transfer processing unit 130 through the service function unit 135. The delay time calculation unit 141 includes a first transmission time t11, which is a time when the transmission processing unit 130 transmits the encapsulated packet 30, and a time when the transmission processing unit 130 receives the processed packet 32. By calculating the difference between the first reception time t12 which is the time, the round trip delay time d1 can be calculated. Depending on the embodiment, the round trip delay time d1 may be calculated by adding a predetermined weight or correction value to the difference between the first transmission time t11 and the first reception time t12.

The processing delay time d2 refers to a time that has elapsed while the service function unit 135 processes the service function. The delay time calculation unit 141 calculates the difference between the second reception time t22 and the second transmission time t21 stored in the storage units 150 and 151 or marked on the network service header 31, so that the processing delay time ( d2) can be calculated. The processing delay time d2 may also be obtained by adding a weight or a correction value to the difference between the second reception time t22 and the second transmission time t21.

The transmission delay time d3 refers to a time taken for the packets 30 and 32 to be transmitted between the transfer processing unit 130 and the service function unit 135. In other words, the transmission delay time d3 means the elapsed time in the link between the transfer processing unit 130 and the service function unit 135. The delay time calculator 141 may calculate the transmission delay time d3 by subtracting the processing delay time d2 from the round trip delay time d1.

As described above, the delay time measurement system 100 can measure not only the round trip delay time d1, but also the processing delay time d2 or the transmission delay time d3. It is possible to diagnose and determine in more detail whether a failure has occurred or whether a failure has occurred in the link between the delivery processing unit 130 and the service function unit 135.

According to an embodiment, at least one of the calculated round trip delay time d1, processing delay time d2, and transmission delay time d3 may be transmitted to the comparison unit 142.

The comparison unit 142 determines at least one of the round trip delay time d1, the processing delay time d2, and the transmission delay time d3 received from the delay time calculating unit 141, and a threshold value v1, v2, v3) and output the comparison result. For example, the comparison unit 142 compares the round trip delay time d1 with a preset first threshold value v1, and compares the processing delay time d2 with a preset second threshold value v2, And/or the transmission delay time d3 is compared with a preset third threshold value v3 to obtain a comparison result. These threshold values v1, v2, and v3 may be previously defined by a designer or a user of the delay time measurement system 100. The comparison result may be transmitted to the result processing unit 143.

The result processing unit 143 may perform a predetermined operation according to the comparison result of the comparison unit 142. The result processing unit 143 may determine whether there is a delay or failure according to the comparison result, and determine whether a processing rule 152 corresponding to the determined delay or failure exists. If there is no pre-defined processing rule 152 corresponding to the generated delay, the result processing unit 143 may transmit a message requesting a new processing rule 112 to the control unit 110. In this case, at least one of a round trip delay time (d1), a processing delay time (d2), and a transmission delay time (d3) together with the request message so that the control unit 110 can appropriately create or modify the processing rule 112 It can also be transmitted to (110). In this case, the control unit 110 transmits the newly acquired processing rule 112 to the delivery processing unit 130 and/or the service function unit 135 according to a network policy, and the like, so that the delivery processing unit 130 and/or the service function unit ( 135) can be properly operated in response to delays or failures. In this case, the newly acquired processing rule 112 may include a failure response table rule, and the failure response table rule is stored in the storage unit 150 of the delivery processing unit 130 in a manner that is stored in the delivery processing unit 130. It is provided and/or recorded in a separate storage unit 151 and transmitted to the delivery processing unit 130 according to the request of the delivery processing unit 130, so that the delivery processing unit 130 can operate appropriately in response to a failure. . Specifically, for example, the delivery processing unit 130 cancels the transmission of the packet 30 to the service function 135 according to the table rule and delivers the packet 30 to the other delivery processing units 130-2 to 130-n. You can do it.

Conversely, if there is a pre-defined processing rule 152 for the delay or failure that has occurred, the result processing unit 143 is the transfer processing unit 130 and/or the service function unit 135 as a pre-defined processing rule 152 It can be made to perform the operation according to.

In one embodiment, if the processing delay time d2 is greater than the second threshold value v2, the result processing unit 143 determines that a problem has occurred in the corresponding service function unit 135, and transfers corresponding thereto. Such a result can be transmitted to the processing unit 130. The transfer processing unit 130 may perform an operation according to a processing rule 152 stored in the storage units 150 and 151 or previously inserted in the transfer processing unit 130. For example, the transmission processing unit 130 may transmit the received or to-be-received packet 30 to another transmission processing unit corresponding to the other service function unit so that another service function unit processes the packet 30. In this case, the other service function unit may be a service function unit that processes the same service function as the service function unit 135 in which a delay or failure occurs. In other words, when the processing delay time d2 of the service function unit 135-i corresponding to the i-th transmission processing unit 130-i, i is a natural number equal to or greater than 1) exceeds the threshold value v2, the i-th transmission processing unit (130-i) transfers all or part of the previously received or continuously received packet 30 to another (i+k) transfer processing unit (130-i+k, where k is a natural number greater than or equal to 1), and , The (i+k)th transfer processing unit 130-i+k transmits the received packet 30 to the corresponding (i+k)th service function unit 135-i+k to the packet 30. Service functions can be handled. In this way, the load of the service function unit 135 having a long processing delay time d2 can be distributed.

According to another embodiment, if the transmission delay time d3 is greater than the third threshold value v3, there is a high possibility that there is a failure in the link between the delivery processing unit 130 and the service function unit 135, and the result The processing unit 143 transmits this to the corresponding delivery processing unit 130, and the delivery processing unit 130 determines whether there is another path (another link) connecting the delivery processing unit 130 and the corresponding service function unit 135. It is determined, and if a different path exists, the packets 30 and 32 may be bypassed and transmitted to the other path. If there is no other path, the packet 30 may be transmitted to another delivery processing unit (at least one of 130-2 to 130-n) for processing by another service function unit (at least one of 135-2 to 135-n). have. According to this method, it is possible to immediately execute the rule to bypass the link under the programmable data plane without the intervention of the control unit 110, so that the system 100 or each of the transfer processing units 140-1 to 140-n This will automatically allow you to respond to failures more quickly.

Hereinafter, various embodiments of a method for measuring a delay time of a service function will be described with reference to FIGS. 6 to 9.

6 is a first flowchart illustrating a method of measuring a delay time of a service function, and FIG. 7 is a second flowchart illustrating a method of measuring a delay time of a service function. 8 is a third flowchart of an embodiment of a method for measuring a delay time of a service function.

As shown in FIG. 6, first, a packet is input from an external physical device or a virtual device, and accordingly, a delay time measurement system, for example, a service classifier, obtains the packet (300). Here, the delay time measurement device may be implemented by including at least one electronic device (for example, a server computer device, a network switch, and/or a network hub, etc.) capable of physically or virtually transmitting and receiving packets.

The delay time measurement system, for example, the service classification unit may classify the received packet and encapsulate the packet by adding a network service header to the packet (302). The encapsulated packet to which the network service header is added may be transmitted from the service classification unit to at least one transmission processing unit, for example, the first transmission processing unit (303, 304).

The first transfer processing unit may determine whether to process a service function for the encapsulated packet (306 ).

If there is a need to perform processing of the service function (example of 306), the first delivery processing unit delivers the encapsulated packet to the service function unit corresponding to the first delivery processing unit, for example, the first service function unit, and at the same time Alternatively, the time at which packets encapsulated to the first service function unit are sequentially delivered (ie, the first transmission time) may be recorded. The first transmission time may be stored in a separately provided storage unit (eg, a memory space allocated for the first transmission processing unit), and/or a partial space of the network service header (eg, a context header), depending on the embodiment. It can also be stored in.

The first service function unit may receive the encapsulation packet and process a service function on the encapsulation packet (312). Simultaneously or sequentially, the first service function may record the reception time (ie, the second reception time) of the encapsulated packet. Depending on the embodiment, the first service function unit may store the second reception time in a separate storage unit, and/or may be marked and stored in a space of the network service header (for example, a context header).

When the processing of the service function is terminated, the first service function unit may return the encapsulated packet back to a corresponding transmission processing unit, that is, the first transmission processing unit (316). Prior to or concurrently with the transmission of the encapsulated packet, the first service function may record the transmission time (i.e., the second transmission time). Depending on the embodiment, the first service function unit may store the second reception time in a separate physical storage unit, and/or may be marked and stored in a space of the network service header (for example, a context header).

As shown in FIG. 7, when the first transfer processing unit receives the encapsulated packet in which the service function has been processed from the first service function unit, it may record the time at which the packet is received, that is, the first reception time ( 318). According to an embodiment, the first reception time may be recorded in a separate storage unit and/or a location of a network service header.

The delay time measurement system, for example, the first transmission processing unit may calculate at least one delay time using at least one of a first transmission time, a first reception time, a second transmission time, and a second reception time (320). . Depending on the embodiment, the calculation of the delay time may be performed by a separate logical element or a physical device.

The at least one delay time may include at least one of a round trip delay time, a processing delay time, and a transmission delay time. The round trip delay time can be obtained by calculating the difference between the first transmission time and the first reception time, the processing delay time can be obtained by calculating the difference between the second reception time and the second transmission time, and the transmission delay time is It can be obtained by calculating the difference between the round trip delay time and the processing delay time. Depending on the embodiment, it is also possible to calculate these delay times by adding a correction value or a weight to the difference.

The delay time and the threshold value may be sequentially compared with each other (322). The threshold value may be a dictionary predefined by a user or a designer, and may be stored, for example, in a storage unit. The threshold value may be predefined corresponding to each delay time. That is, according to an embodiment, the threshold value may include at least one of a first threshold value corresponding to a round trip delay time, a second threshold value corresponding to a processing delay time, and a third threshold value corresponding to a transmission delay time. . Accordingly, the round trip delay time may be compared with the first threshold value, the processing delay time may be compared with the second threshold value, and/or the transmission delay time may be compared with the third threshold value. Depending on the embodiment, the first transfer processing unit may perform comparison on all of the aforementioned round trip delay time, processing delay time, and transmission delay time, or may perform comparison only on one or both of them.

As a result of comparing the delay time and the threshold value, if the delay time is greater than the threshold value (YES in 322), it may be sequentially determined whether or not a processing rule exists (324). Specifically, for example, when the processing delay time is greater than the second threshold, it may be determined whether or not a corresponding processing rule exists, and/or when the transmission delay time is greater than the third threshold, the corresponding It may be determined whether or not there is a processing rule to be processed. If a plurality of delay times (for example, a processing delay time and a transmission delay time) are implemented to be compared with a threshold value, the first transfer processing unit or the first service function unit, depending on the embodiment, among these delay times If any one is greater than the threshold value, it may be designed to determine the presence or absence of the processing rule, or if all of these delay times are greater than the threshold value, it may be designed to determine the presence or absence of the processing rule.

If a previously defined or stored processing rule exists (example of 324), a delivery processing unit or a service function unit operates according to the processing rule (326). For example, if the processing delay time is greater than the second threshold value, the first delivery processing unit is not the first service function unit, but another service function unit that performs the same operation according to the corresponding processing rule (for example, the second service function unit). The encapsulated packet may be delivered to another delivery processing unit (for example, a second delivery processing unit, etc.) so that the packet is transmitted to each other. For another example, when the transmission delay time is greater than the third threshold value, the first transmission processing unit searches for another link and transmits the packet to the first service function unit through the searched other link according to a processing rule corresponding thereto. Alternatively, the packet may be delivered to another delivery processing unit corresponding to the other service function unit so that the other service function unit processes the packet. In addition, the delivery processing unit, the service function unit, and the like may deliver and process the encapsulated packet according to a predetermined processing rule.

Conversely, if the defined or stored processing rule does not exist (NO in 324), a new processing rule may be provided (328). In detail, for example, the first delivery processing unit or the first service processing unit may directly create or modify a processing rule according to the absence of a processing rule, or may request a processing rule from a portion capable of creating or modifying the processing rule. In this case, the part capable of generating or modifying the processing rule may include a first delivery processing unit or another part logically or physically separated from the first service processing unit, and may include, for example, a control unit.

When a new processing rule is provided, the first transfer processing unit, the first service function unit, etc. operate according to the new processing rule (326). If the first delivery processing unit or another part logically or physically separated from the first service processing unit (eg, a control unit) creates or modifies a new processing rule, the first delivery processing unit or the first service processing unit It receives the processing rule to be transmitted, and operates according to the received processing rule.

Conversely, if there is no need to perform processing of the service function (No in 306), the first transfer processing unit may transfer the encapsulated packet to the next transfer processing unit, for example, the second transfer processing unit (308, 304). This may be performed until the encapsulation packet is delivered to all of the predetermined delivery processing units. For example, when N number of transmission processing units are set, the operation may be performed until finally the encapsulation packet is finally delivered to the Nth transmission processing unit (307).

On the other hand, if the delay time is less than the threshold value (No in 322), the first transfer processing unit or the first service function unit may continue to perform the service function processing process as previously designed.

As shown in FIG. 8, if all service function processing for a packet is finished (example of 332), the network service header encapsulated in the packet is removed (334), and the packet from which the network service header is removed is output to the outside. Can be (336). Here, the completion of all service function processing may include that packets are processed sequentially or in an arbitrary order by all of the predetermined service function units. In other words, among all the transmission processing units, all of the transmission processing units for which the packet is scheduled to be delivered. It may include that the encapsulation packet is delivered at least once. The removal of the network service header may be performed by the service classification unit, or may be performed by another part separately provided for removal of the network service header. Here, the service classification unit may include a service classification unit that attaches a network service header to the packet.

If the service function processing for the packet is not finished (No in 332), the encapsulated packet may be delivered to the next delivery processing unit (for example, the second delivery processing unit) (339, 304). Thereafter, the next delivery processing unit may check the delay time through the same process as described above, and when a delay occurs, use a pre-defined delay or failure processing rule or create and use a new processing rule (304 to 326). ). If no failure has occurred, the encapsulated packet may be delivered to the next delivery processing unit (for example, a third transmission processing unit), or the network service header may be arranged and the packet may be output to the outside. The above-described process may be repeatedly performed until a packet is delivered to all required transfer processing units (ie, at least one of the first to n-th transfer processing units) (338).

On the other hand, as shown in Fig. 6, the service function corresponding to the delivery processing unit (for example, the first transmission processing unit) that has received the packet does not need to perform processing on the packet, so that the delivery processing unit is When the encapsulated packet is delivered to the second delivery processing unit) (No in 306), the next delivery processing unit transmits the packet to the next delivery processing unit (for example, the third delivery processing unit) depending on whether or not the service function is processed (306, 308, 310) Or the same or a partially modified method as described above, determines whether there is a delay or failure, and delivers a packet to the corresponding service function unit (for example, a third service function unit) according to the determination result, or exists. Or, it may operate according to a newly created processing rule (312 to 336). The k-th transfer processing unit (k is a natural number greater than or equal to 1) may also operate in the same or modified manner as described above (304 to 336).

9 is a flowchart of an embodiment of an operation of a transfer processing unit.

As shown in FIG. 9, the transfer processing unit may perform different operations according to whether the input encapsulation packet is a previously input packet or a new packet.

Specifically, first, the transfer processing unit may obtain a packet (340), and determine whether the acquired packet is a packet that has been previously input or a new packet (342). Here, the previously input packet may include a packet transmitted by the delivery processing unit to the service function unit.

If it is a new packet (YES in 342), the transfer processing unit may transmit the packet to the service function unit, and simultaneously or sequentially record the time at which the packet is transmitted, that is, the first transmission time (344). The service function unit performs service function processing in response to reception of the packet. As described above, the service function unit may mark the packet reception time (ie, the second reception time) in addition to the packet reception on the storage unit or the network service header added to the packet.

Conversely, if it is not a new packet (No in 342), the transfer processing unit records the time at which the packet is received (i.e., the first reception time), and if necessary, the delay time (e.g., round trip delay time, processing delay time, and / Or a transmission delay time) may be calculated or the received packet may be transmitted to another transmission processing unit (346).

The method for measuring a delay time of a service function according to the above-described embodiment may be implemented in the form of a program that can be driven by a computer device. Here, the program may include a program command, a data file, a data structure, or the like alone or in combination. The program may be designed and produced using machine code or high-level language code. The program may be specially designed to implement the above-described method, or may be implemented using various functions or definitions that are known and available to those skilled in the computer software field. In addition, here, the computer device may be implemented including a processor or a memory that enables the function of a program to be realized, and may further include a communication device if necessary.

The above-described program can be recorded on a computer-readable recording medium. The computer-readable recording medium includes, for example, a semiconductor storage device such as a solid state drive (SSD), ROM, RAM, or flash memory, a magnetic disk storage medium such as a hard disk or a floppy disk, and a compact disk or a DVD. It may include at least one type of physical device capable of storing a specific program executed according to a call such as a computer, such as a magnetic-optical recording medium such as an optical recording medium, a floptic disk, and the like, and a magnetic tape.

Although various embodiments of the service function delay time measurement system, apparatus, and method have been described above, the service function delay time measurement system, apparatus, and method are not limited to the above-described embodiments. Various devices or methods that can be implemented by modifying and modifying based on the above-described embodiment by a person of ordinary skill in the art may also be an example of the system, device, and method for measuring the delay time of the service function described above. For example, the described techniques are performed in an order different from the described method, and/or components such as systems, structures, devices, and circuits described are combined or combined in a form different from the described method, or other components or Even if it is replaced or substituted by an equivalent, it may be an example of a system, apparatus, and method for measuring the delay time of the service function described above.

Claims (15)

A time recording unit for recording a first transmission time of transmitting a packet to a service function (SF) node and a first reception time of receiving the packet from the SF node; And The SF node recorded in the packet reads a second reception time at which the packet was received and a second transmission time at which the SF node transmitted the packet, and a difference between the second transmission time and the second reception time Delay time measuring device including a delay processing unit for calculating the processing delay time by using. The method of claim 1, The delay time measuring device is a service function forwarder (SFF) node, The delay processing unit calculates a round trip delay time for the packet based on the first transmission time and the first reception time. The method of claim 2, The delay processing unit calculates a transmission delay time for the packet by using a difference between the round trip delay time and the processing delay time. The method of claim 3, When at least one of the processing delay time, the round trip delay time, and the transmission delay time exceeds a predefined threshold, the delay processing unit operates according to a predefined processing rule or requests a processing rule from an external device. and, The external device is a controller of a Software Defined Network (SDN), Delay time measurement device. A service function unit for receiving a packet, processing a service function on the packet, and outputting the processed packet; And A transfer processing unit that delivers a packet to the service function unit, records a first transmission time of transmitting the packet to the service function unit, and records a first reception time at which the processed packet is received from the service function unit; Including, The service function unit records a second reception time at which the packet is received, and records a second transmission time at which the packet processed by the service function is transmitted to the transfer processing unit, The processing delay time for the packet is calculated based on the second reception time and the second transmission time. The method of claim 5, The round trip delay time for the packet is calculated based on the first transmission time and the first reception time, and the transmission delay time for the packet is calculated based on a difference between the round trip delay time and the processing delay time. Latency measurement system. The method of claim 6, When at least one of the processing delay time, the round trip delay time, and the transmission delay time exceeds a predefined threshold value, the transmission processing unit operates according to a predefined processing rule. The method of claim 7, The delay time measurement system further includes a control unit for generating a new processing rule, The delivery processing unit, when the predefined processing rule does not exist, requests a new processing rule from the control unit, and operates according to a new processing rule generated by the control unit. The method of claim 5, The packet is a delay time measurement system including an encapsulated packet to which a network service header is added. The method of claim 9, The service function unit marks and records the second reception time and the second transmission time on the network service header. The method of claim 1, The delay time measurement system is an SDN system, The service function unit is a service function (SF) node, The forwarding processing unit is a Service Function Forwarder (SFF) node, Latency measurement system. Transmitting a packet to a service function unit by a delivery processing unit and recording a first transmission time at which the packet is delivered; Receiving, by a service function unit, a packet, and recording a second reception time at which the packet is received; Recording, by the service function unit, a second transmission time when the service function-processed packet is transferred to the transfer processing unit and the service function-processed packet is transferred to the transfer processing unit; Receiving the processed packet from the service function unit by the transfer processing unit and recording a first reception time at which the processed packet is received; And And calculating a processing delay time for the packet based on the second reception time and the second transmission time. The method of claim 12, Calculating a round trip delay time for the packet based on the first transmission time and the first reception time; And Calculating a transmission delay time for the packet by using a difference between the round trip delay time and the processing delay time. The method of claim 13, When at least one of the processing delay time, the round trip delay time, and the transmission delay time exceeds a predefined threshold value, the transfer processing unit operates according to a predefined processing rule; . The method of claim 14, If the predefined processing rule does not exist, generating a new processing rule, and operating the transfer processing unit according to the new processing rule.

Images