Method, apparatus, device and readable storage medium for applying quality difference analysis

Abstract

The present invention provides a method, apparatus, device, and readable storage medium for applying quality difference analysis. The method for applying quality difference analysis includes: an edge computing power board receives a second upstream data stream from a main control board, determines the priority of the second upstream data stream according to the application ID, receives a second downstream data stream from a line card, and determines the priority of the second downstream data stream according to the application ID; the edge computing power board performs application quality difference analysis and quality difference positioning according to the priority, based on the second upstream data stream and the first tag and the second tag in the second upstream data stream, and the second downstream data stream and the third tag and the fourth tag in the second downstream data stream. In the present invention, by means of the application ID field, a large amount of computing power required for the edge computing power board to identify applications is saved, and the priority queue is divided for processing, so that high-priority applications can be preferentially subjected to application quality difference analysis even when the network is congested, thereby better coping with and solving application quality differences.

Description

Technical Field

[0001] The present invention relates to the field of Passive Optical Network (PON), and particularly to a method, device, equipment and readable storage medium for applying poor quality analysis. Background Art

[0002] PON (Passive Optical Network) is a typical passive optical network, mainly composed of an OLT (Optical Line Terminal) optical line terminal device on the central office side, an ONU (Optical Network Unit) optical network unit on the user side, and an ODN (Optical Distribution Network) optical distribution network. Among them, the OLT is an important central office device, mainly including a main control board, line cards, and edge computing power boards, etc. Generally, the main control board is responsible for main switching, that is, the switching between line cards inside the device and data exchange with the upper-layer network through the uplink port. The line cards are mainly used to provide PON ports and realize data connection with the ONU through an optical splitter. By deploying computing power on the OLT and ONU, operators can build an edge intelligent network architecture with edge collaboration, giving play to the advantages of "connection + computing power" integration. Generally speaking, the upstream data stream sent by the end user to the upper-layer network enters from the PON port of the line card and is then forwarded to the upper-layer network through the uplink port of the main control board. Conversely, the downstream data stream sent by the upper-layer network to the end user enters from the uplink port of the main control board and is then forwarded to the end user through the PON port of the line card.

[0003] Due to the advantages of the PON network architecture, such as no active components, reduced cabling infrastructure, and flexible media transmission, the PON network architecture is becoming more and more widely used. Currently, the demand for multimedia services (such as applications like images, videos, and games) of end users is also rising rapidly. As a result, the end users' requirements for the quality of network transmission are getting higher and higher. For example, end users hope to meet indicators such as low latency, low packet loss, and low jitter during network transmission. Therefore, in order to meet the rapidly rising demand of the above services, it is necessary to accurately analyze whether the network devices related to the services have poor quality. The poor quality of network devices means that the current operating quality of the network devices is lower than the preset operating quality threshold, that is, the current operating quality of the network devices is poor. Currently, application identification and poor quality analysis are achieved through the edge computing power boards inside the OLT device. However, during the process of application identification by the edge computing power boards, a large amount of computing power resources are consumed, thus bringing a great load pressure to the OLT device. Summary of the Invention

[0004] The main object of the present invention is to provide a method, device, equipment and readable storage medium for application quality difference analysis, aiming to solve the technical problem of excessive load pressure on the OLT device when performing application quality difference analysis on the PON network at present.

[0005] In a first aspect, the present invention provides a method for application quality difference analysis, and the method for application quality difference analysis includes:

[0006] The line card receives the original upstream data stream, adds a first label to the data frames in the original upstream data stream to obtain a first upstream data stream, and the line card forwards the first upstream data stream to the main control board;

[0007] The main control board receives the first upstream data stream, adds a second label to the data frames in the first upstream data stream to obtain a second upstream data stream, the second label includes an application ID, and the main control board sends the second upstream data stream to the edge computing board;

[0008] The main control board receives the original downstream data stream, adds a third label to the data frames in the original downstream data stream, the third label includes an application ID, to obtain a first downstream data stream, and the main control board forwards the first downstream data stream to the line card;

[0009] The line card receives the first downstream data stream, adds a fourth label to the data frames in the first downstream data stream to obtain a second downstream data stream, and the line card sends the second downstream data stream to the edge computing board;

[0010] The edge computing board receives the second upstream data stream from the main control board, determines the priority of the second upstream data stream according to the application ID, receives the second downstream data stream from the line card, and determines the priority of the second downstream data stream according to the application ID;

[0011] The edge computing board performs application quality difference analysis and quality difference positioning according to the priority, based on the second upstream data stream and the first label and the second label in the second upstream data stream, the second downstream data stream and the third label and the fourth label in the second downstream data stream.

[0012] Optionally, the main control board receives the first upstream data stream, adds a second label to the data frames in the first upstream data stream to obtain a second upstream data stream, and the second label includes an application ID, including:

[0013] The main control board receives the first upstream data stream;

[0014] The main control board determines the application ID based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number and destination TCP port number in the first upstream data stream;

[0015] The master control board adds a second label to the data frames in the first upstream data stream to obtain a second upstream data stream, and the second label includes an application ID.

[0016] Optionally, the master control board receives an original downstream data stream and adds a third label to the data frames in the original downstream data stream. The third label includes an application ID, including:

[0017] The master control board receives the original downstream data stream;

[0018] The master control board determines the application ID based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the original downstream data stream;

[0019] The master control board adds a third label to the data frames in the original downstream data stream, and the third label includes an application ID.

[0020] Optionally, before the master control board is based on the application recognition rule, it includes:

[0021] The edge computing power board performs application recognition and calibrates the correspondence between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID;

[0022] The edge computing power board sends the calibrated correspondence between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID as the application recognition rule to the master control board.

[0023] Optionally, the application quality difference analysis and quality difference positioning based on the second upstream data stream and the first label and the second label in the second upstream data stream, the second downstream data stream and the third label and the fourth label in the second downstream data stream include:

[0024] The first label includes the upstream inflow port, upstream inflow time, and upstream packet count. The second label further includes the upstream outflow port and upstream outflow time. The third label further includes the downstream inflow port, downstream inflow time, and downstream packet count. The fourth label includes the downstream outflow port and downstream outflow time;

[0025] The edge computing power board identifies the second upstream data stream and the second downstream data stream based on the positions of the label fields in the packets, and identifies the user side and the network side based on the contents of the port fields in the packets. The second upstream data stream and the second downstream data stream are data streams communicating using the TCP protocol;

[0026] The edge computing power board marks the upstream inflow time of the packet with serial number X and length Y in the second upstream data stream as T1, and the upstream outflow time as T2. It marks the downstream inflow time of the packet with ACK as X+Y in the second downstream data stream as T3, and the downstream outflow time as T4;

[0027] The edge computing power board marks the downstream inflow time of the packet with serial number U and length V in the second downstream data stream as T5, and the downstream outflow time as T6. It marks the upstream inflow time of the packet with ACK as U+V in the second upstream data stream as T7, and the upstream outflow time as T8;

[0028] Based on T1, T2, T3, T4, T5, T6, T7 and T8, the upstream internal delay, downstream internal delay, network-side delay and user-side delay are calculated;

[0029] Based on the upstream packet count, the upstream internal packet loss quantity is calculated;

[0030] Based on the downstream packet count, the downstream internal packet loss quantity is calculated;

[0031] Based on the packet serial numbers in the downstream data stream, the network-side downstream packet loss quantity is calculated;

[0032] Based on the packet retransmission times in the downstream data stream, the network-side upstream packet loss quantity is calculated;

[0033] Based on the packet serial numbers in the upstream data stream, the user-side upstream packet loss quantity is calculated;

[0034] Based on the packet retransmission times in the upstream data stream, the user-side downstream packet loss quantity is calculated;

[0035] Based on a preset period, according to the upstream internal delay, downstream internal delay, network-side delay, user-side delay, upstream internal packet loss quantity, downstream internal packet loss quantity, network-side downstream packet loss quantity, network-side upstream packet loss quantity, user-side upstream packet loss quantity and user-side downstream packet loss quantity, application quality degradation analysis and quality degradation location are performed.

[0036] Optionally, the calculating the upstream internal delay, downstream internal delay, network-side delay and user-side delay based on T1, T2, T3, T4, T5, T6, T7 and T8 includes:

[0037] Based on T1, T2, T3, T4, T5, T6, T7 and T8, the upstream internal delay, downstream internal delay, network-side delay and user-side delay are calculated through formulas. The calculation formulas are as follows:

[0038] Upstream internal delay = T2 - T1 or T8 - T7;

[0039] Downlink internal delay = T4 - T3 or T6 - T5;

[0040] Network side delay = T3 - T2;

[0041] User side delay = T7 - T6.

[0042] Optionally, based on a preset period, according to the uplink internal delay, downlink internal delay, network side delay, user side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network side downlink packet loss quantity, network side uplink packet loss quantity, user side uplink packet loss quantity, and user side downlink packet loss quantity, performing application quality degradation analysis and quality degradation location includes:

[0043] Determining the application to be analyzed and the quality degradation threshold of the application to be analyzed according to the application ID, where the quality degradation threshold of the application to be analyzed includes the quality degradation thresholds corresponding to the uplink internal delay, downlink internal delay, network side delay, user side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network side downlink packet loss quantity, network side uplink packet loss quantity, user side uplink packet loss quantity, and user side downlink packet loss quantity;

[0044] When the uplink internal delay, downlink internal delay, network side delay, user side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network side downlink packet loss quantity, network side uplink packet loss quantity, user side uplink packet loss quantity, and user side downlink packet loss quantity of the application to be analyzed exceed the quality degradation threshold of the application to be analyzed, it is determined as a single - time application quality degradation;

[0045] Based on a preset period, counting the number of times of single - time application quality degradation of the application to be analyzed within the preset period;

[0046] When within the preset period, the number of times of single - time application quality degradation of the uplink internal delay, and / or downlink internal delay, and / or uplink internal packet loss quantity, and / or downlink internal packet loss quantity of the application to be analyzed exceeds the first preset number of times, it is determined as an internal fault, where the first preset number of times includes the preset numbers corresponding to the uplink internal delay, downlink internal delay, uplink internal packet loss quantity, and downlink internal packet loss quantity of the application to be analyzed;

[0047] When within the preset period, the number of times of single - time application quality degradation of the network side delay, and / or network side downlink packet loss quantity, and / or network side downlink packet loss quantity of the application to be analyzed exceeds the second preset number of times, it is determined as a network side fault, where the second preset number of times includes the preset numbers corresponding to the network side delay, network side downlink packet loss quantity, and network side downlink packet loss quantity of the application to be analyzed;

[0048] When the user - side delay of the application to be analyzed, and / or the number of uplink packet losses on the user side, and / or the number of times of single - application quality degradation caused by the number of uplink packet losses on the user side exceeds the third preset number within the preset period, it is determined as a user - side fault. The third preset number includes the preset numbers corresponding to the user - side delay of the application to be analyzed, the number of uplink packet losses on the user side, and the number of uplink packet losses on the user side.

[0049] In a second aspect, the present invention further provides an application quality - degradation analysis device, and the application quality - degradation analysis device includes:

[0050] A first uplink module, configured to receive the original uplink data stream by a line card, add a first label to the data frames in the original uplink data stream to obtain a first uplink data stream, and the line card forwards the first uplink data stream to the main control board;

[0051] A second uplink module, configured to receive the first uplink data stream by the main control board, add a second label to the data frames in the first uplink data stream to obtain a second uplink data stream, where the second label includes an application ID, and the main control board sends the second uplink data stream to the edge computing board;

[0052] A first downlink module, configured to receive the original downlink data stream by the main control board, add a third label including the application ID to the data frames in the original downlink data stream to obtain a first downlink data stream, and the main control board forwards the first downlink data stream to the line card;

[0053] A second downlink module, configured to receive the first downlink data stream by the line card, add a fourth label to the data frames in the first downlink data stream to obtain a second downlink data stream, and the line card sends the second downlink data stream to the edge computing board;

[0054] A receiving module, configured to receive the second uplink data stream from the main control board by the edge computing board, determine the priority of the second uplink data stream according to the application ID, receive the second downlink data stream from the line card, and determine the priority of the second downlink data stream according to the application ID;

[0055] A quality - degradation analysis module, configured to perform application quality - degradation analysis and quality - degradation location by the edge computing board according to the priority, based on the second uplink data stream and the first label and the second label in the second uplink data stream, and the second downlink data stream and the third label and the fourth label in the second downlink data stream.

[0056] In a third aspect, the present invention further provides an application quality - degradation analysis device, and the application quality - degradation analysis device includes a processor, a memory, and an application quality - degradation analysis program stored on the memory and executable by the processor. When the application quality - degradation analysis program is executed by the processor, the steps of the application quality - degradation analysis method as described above are implemented.

[0057] Fourthly, the present invention further provides a readable storage medium, on which an application quality difference analysis program is stored. When the application quality difference analysis program is executed by a processor, the steps of the application quality difference analysis method as described above are implemented.

[0058] In the present invention, a line card receives an original upstream data stream, adds a first label to the data frames in the original upstream data stream to obtain a first upstream data stream, and the line card forwards the first upstream data stream to a main control board; the main control board receives the first upstream data stream, adds a second label to the data frames in the first upstream data stream to obtain a second upstream data stream, and the second label includes an application ID. The main control board sends the second upstream data stream to an edge computing power board; the main control board receives an original downstream data stream, adds a third label to the data frames in the original downstream data stream, and the third label includes an application ID to obtain a first downstream data stream. The main control board forwards the first downstream data stream to the line card; the line card receives the first downstream data stream, adds a fourth label to the data frames in the first downstream data stream to obtain a second downstream data stream, and the line card sends the second downstream data stream to the edge computing power board; the edge computing power board receives the second upstream data stream from the main control board, determines the priority of the second upstream data stream according to the application ID, receives the second downstream data stream from the line card, and determines the priority of the second downstream data stream according to the application ID; the edge computing power board performs application quality difference analysis and quality difference positioning according to the priority, based on the second upstream data stream and the first label and the second label in the second upstream data stream, and the second downstream data stream and the third label and the fourth label in the second downstream data stream. In the present invention, by adding label fields for quality difference analysis to the data streams when the upstream data stream and the downstream data stream flow through the line card and the main control board, and adding an application ID field to the upstream data stream and the downstream data stream can save a large amount of computing power required for the edge computing power board to identify applications, and by dividing the upstream data stream and the downstream data stream into priority queues for processing, it can enable high-priority applications to be preferentially analyzed for application quality difference even when the network is congested, so as to better cope with and solve application quality differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic hardware structure diagram of an embodiment of the application quality difference analysis device of the present invention;

[0060] Figure 2 It is a schematic flowchart of an embodiment of the application quality difference analysis method of the present invention;

[0061] Figure 3 is Figure 2 a detailed flowchart of step S20 in;

[0062] Figure 4 is Figure 2 a detailed flowchart of step S30 in;

[0063] Figure 5 Schematic diagrams of the uplink data stream and the downlink data stream of an embodiment of the quality difference analysis method applied in the present invention;

[0064] Figure 6 Schematic diagram of the functional modules of an embodiment of the quality difference analysis device applied in the present invention;

[0065] Figure 7 Schematic diagram of the system architecture of an embodiment of the quality difference analysis system applied in the present invention;

[0066] Figure 8 Schematic diagram of the data processing flow of an embodiment of the quality difference analysis system applied in the present invention.

[0067] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0068] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0069] In a first aspect, an embodiment of the present invention provides a quality difference analysis device.

[0070] Referring to Figure 1 , Figure 1 , which is a schematic diagram of the hardware structure of an embodiment of the quality difference analysis device applied in the present invention. In the embodiment of the present invention, the quality difference analysis device may include a processor 1001 (such as a Central Processing Unit, CPU), a communication bus 1002, a network interface 1003, and a memory 1004. Among them, the communication bus 1002 is used to realize the connection and communication between these components; the network interface 1003 may optionally include a standard wired interface and a wireless interface (such as a Wireless-Fidelity, WI-FI interface); the memory 1004 may be a high-speed random access memory (random access memory, RAM), or a stable memory (non-volatile memory), such as a disk memory, and the memory 1004 may optionally be a storage device independent of the aforementioned processor 1001. Those skilled in the art can understand that Figure 1 the hardware structure shown in

[0071] Continuing to refer to Figure 1 , Figure 1In the memory 1004 as a computer storage medium, an operating system, a network communication module, and an application quality difference analysis program may be included. Among them, the processor 1001 may call the application quality difference analysis program stored in the memory 1004 and execute the application quality difference analysis method provided by the embodiments of the present invention.

[0072] In a second aspect, an embodiment of the present invention provides an application quality difference analysis method.

[0073] To more clearly show the application quality difference analysis method provided by the embodiments of the present application, first, the application scenario of the application quality difference analysis method provided by the embodiments of the present application will be introduced.

[0074] The application quality difference analysis method provided by the embodiments of the present application is applied in a PON network architecture. With more and more multimedia services (such as applications like images, videos, and games), end-users have higher and higher requirements for the quality of network transmission. It is necessary to accurately analyze whether there is an application quality difference in network devices related to services. Currently, the efficiency of application identification and quality difference analysis achieved by the edge computing board in the OLT device is relatively low, bringing a great load pressure to the OLT device.

[0075] In one embodiment, refer to Figure 2 , Figure 2 which is a schematic flowchart of an embodiment of the application quality difference analysis method of the present invention. As shown in Figure 2 , the application quality difference analysis method includes:

[0076] Step S10: The line card receives the original upstream data stream and adds a first tag to the data frames in the original upstream data stream to obtain a first upstream data stream, and the line card forwards the first upstream data stream to the main control board.

[0077] In this embodiment, the line card receives the original upstream data stream sent by the end-user to the upper-layer network from the PON port, adds a first tag (i.e., ONU-TAG) to the data frames in the original upstream data stream to obtain a first upstream data stream, and then the line card forwards the first upstream data stream to the main control board. Among them, the tag field may include, for example, the inflow and outflow time, packet count, inflow and outflow port numbers, etc., which can enable the edge computing board to perform application quality difference analysis more accurately and efficiently, and locate the quality difference position more quickly, such as locating whether the application quality difference belongs to the user side, the network side, or the OLT internal application quality difference.

[0078] Step S20: The main control board receives the first upstream data stream, adds a second tag to the data frames in the first upstream data stream to obtain a second upstream data stream, where the second tag includes an application ID, and the main control board sends the second upstream data stream to the edge computing board.

[0079] In this embodiment, after receiving the first upstream data stream from the line card, the main control board adds a second tag (i.e., UPLINK-TAG) to the data frames in the first upstream data stream to obtain a second upstream data stream. That is, the tag field of the second upstream data stream is the first tag (ONU-TAG) + the second tag (UPLINK-TAG). The second tag includes an application ID, which is used to identify the application to which the data stream belongs, such as images, videos, and games. Based on the application ID field carried in the data stream, the edge computing board can directly identify the application of the data stream for application quality difference analysis, thereby saving a large amount of computing power consumed by the edge computing board for application identification of the data stream. The main control board sends the second upstream data stream to the edge computing board for application quality difference analysis. At the same time, the main control board strips the content of the first tag field in the first upstream data stream, that is, the first upstream data stream is restored to the original upstream data stream and forwarded to the upper-layer network through the uplink port. In another implementation, the main control board sends the second upstream data stream to the edge computing board for application quality difference analysis, and at the same time strips the content of the first tag and the second tag field in the second upstream data stream, which can also restore the second upstream data stream to the original upstream data stream.

[0080] Step S30: The main control board receives the original downstream data stream and adds a third tag to the data frames in the original downstream data stream. The third tag includes an application ID to obtain a first downstream data stream. The main control board forwards the first downstream data stream to the line card.

[0081] In this embodiment, the main control board receives the original downstream data stream sent by the upper-layer network to the end user from the uplink port, and adds a third tag (i.e., UPLINK-TAG) to the data frames in the original downstream data stream to obtain a first downstream data stream. The field of the third tag includes an application ID field, which is also used to identify the application to which the data stream belongs. Then, the main control board forwards the first downstream data stream to the line card.

[0082] Step S40: The line card receives the first downstream data stream and adds a fourth tag to the data frames in the first downstream data stream to obtain a second downstream data stream. The line card sends the second downstream data stream to the edge computing board.

[0083] In this embodiment, after receiving the first downlink data stream from the main control board, the line card adds a fourth tag (i.e., ONU-TAG) to the data frames in the first downlink data stream to obtain a second downlink data stream. That is, the tag field in the second downlink data stream is the third tag (UPLINK-TAG) + the fourth tag (ONU-TAG). The line card forwards the second downlink data stream to the edge computing board through the main control board for application quality difference analysis. At the same time, the line card strips the content of the third tag field in the first downlink data stream, that is, it realizes restoring the first downlink data stream to the original downlink data stream and forwarding it to the end user through the PON port. In another implementation, the line card forwards the second downlink data to the edge computing board through the main control board for application quality difference analysis. At the same time, the line card strips the content of the third tag and the fourth tag field in the second downlink data stream, which can also realize restoring the second downlink data stream to the original downlink data stream.

[0084] Step S50: The edge computing board receives the second uplink data stream from the main control board, determines the priority of the second uplink data stream according to the application ID, receives the second downlink data stream from the line card, and determines the priority of the second downlink data stream according to the application ID.

[0085] In this embodiment, the edge computing board receives the second uplink data stream from the main control board and the second downlink data stream from the line card, and determines the priorities of the second uplink data stream and the second downlink data stream respectively according to the application ID. The second uplink data stream and the second downlink data stream enter the queues with corresponding priorities.

[0086] Step S60: The edge computing board performs application quality difference analysis and quality difference positioning based on the second uplink data stream, the first tag and the second tag in the second uplink data stream, the second downlink data stream, and the third tag and the fourth tag in the second downlink data stream according to the priority.

[0087] In this embodiment, the edge computing board performs application quality difference analysis and quality difference positioning according to the second uplink data stream, the second downlink data stream, the first tag, the second tag, the third tag, and the fourth tag in the order of the priority of the queue.

[0088] In this embodiment, by adding tag fields for quality difference analysis to the data stream when the upstream data stream and the downstream data stream flow through the line card and the main control board respectively, where the application ID is used to identify the application to which the data stream belongs, such as images, videos, and games, etc. Based on the application ID field carried in the data stream, the edge computing power board can directly identify the application of the data stream for application quality difference analysis, thereby saving a large amount of computing power required for the edge computing power board to identify the application. The added tag fields can include the inflow and outflow time, packet count, inflow and outflow port numbers, etc. After the edge computing power board obtains the second upstream data stream from the main control board and the second downstream data stream from the line card, it can quickly and efficiently perform application quality difference analysis and quickly locate the quality difference position, such as locating whether the application quality difference belongs to the user side, the network side, or the OLT internal application quality difference, and dividing the upstream data stream and the downstream data stream into priority queues for processing, which can enable high-priority applications to be preferentially analyzed for application quality difference even when the network is congested, so as to better cope with and solve application quality differences.

[0089] Further, in one embodiment, referring to Figure 3 , Figure 3 is Figure 2 a detailed flowchart of step S20 in Figure 3 as shown in

[0090] Step S201, the main control board receives the first upstream data stream;

[0091] Step S202, the main control board determines the application ID based on the application identification rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the first upstream data stream;

[0092] Step S203, the main control board adds a second tag to the data frame in the first upstream data stream to obtain the second upstream data stream, and the second tag includes the application ID.

[0093] In this embodiment, the application identification rule identifies the corresponding relationship between the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the data stream. For example, if a data stream is represented by "{source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number}", then the identification rule can be expressed as "{source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number}, application ID". Therefore, the main control board can determine the application ID of the upstream data stream based on the application identification rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the first upstream data stream.

[0094] Further, in one embodiment, referring to Figure 4 , Figure 4 is Figure 2 a detailed flowchart of step S30 in Figure 4 As shown, step S30 includes:

[0095] Step S301, the main control board receives the original downstream data stream;

[0096] Step S302, the main control board determines the application ID based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the original downstream data stream;

[0097] Step S303, the main control board adds a third tag to the data frame in the original downstream data stream, and the third tag includes the application ID.

[0098] In this embodiment, using the same method as in step S202, the main control board can determine the application ID of the downstream data stream based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the original downstream data stream.

[0099] Further, in one embodiment, before the main control board is based on the application recognition rule, it includes:

[0100] The edge computing board performs application recognition and calibrates the corresponding relationship between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID;

[0101] The edge computing board sends the calibrated corresponding relationship between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID as the application recognition rule to the main control board.

[0102] In this embodiment, the edge computing power board uses DPI (Deep Packet Inspection) to perform application identification on the upstream data stream and the downstream data stream. The so-called "deep" is in comparison with the ordinary packet analysis level. "Ordinary packet detection" only analyzes the content below the 4th layer of the IP packet, including the source address, destination address, source port, destination port, and protocol type. In addition to analyzing the previous levels, DPI also adds application layer analysis to identify various applications and their content. Therefore, performing application identification consumes a large amount of computing power resources. Exactly for this reason, the present invention proposes to calibrate the identification rules after the edge computing power board performs application identification, and send the calibrated identification rules to the main control board. After receiving the upstream data stream or the downstream data stream, the main control board can

[0103] determine the application ID based on the identification rules and the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the data stream. Thus, by carrying the application ID field in the data stream, the edge computing power board can directly identify the application, which can save a large amount of computing power required for the edge computing power board to perform application identification. The above

[0104] is the application identification of the TCP data stream. Using the same method, based on the identification rules and the source IP address, destination IP address, IP protocol type, source UDP port number, and destination UDP port number in the data stream, the application identification of the UDP data stream can be realized.

[0105] Further, in one embodiment, step S60 includes:

[0106] Step S601, the first tag includes the upstream inflow port, upstream inflow time, and upstream packet

[0107] count. The second tag further includes the upstream outflow port and upstream outflow time. The third tag further includes the downstream inflow port, downstream inflow time, and downstream packet count. The fourth tag includes the downstream outflow port and downstream outflow time;

[0108] Step S602, the edge computing power board identifies the second upstream data stream and the second downstream data stream based on the positions of the tag fields in the packet, and identifies the user side and the network side based on the content of the port field in the packet. The

[0109] second upstream data stream and the second downstream data stream are data streams that communicate using the TCP protocol;

[0110] Step S603, the edge computing power board marks the upstream inflow time of the packet with sequence number X and length Y in the second upstream data stream as T1, and the upstream outflow time as T2, and for the second downstream data stream

[0111] The downlink inflow time of the packet with ACK being X+Y is marked as T3, and the downlink outflow time is marked as T4; in step S604, the edge computing board marks the downlink inflow time of the packet with sequence number U and length V in the second downlink data stream as T5, the downlink outflow time as T6, marks the uplink inflow time of the packet with ACK being U+V in the second uplink data stream as T7, and the uplink outflow time as T8;

[0112] Step S605, based on T1, T2, T3, T4, T5, T6, T7, and T8, calculate the uplink

[0113] internal delay, downlink internal delay, network-side delay, and user-side delay;

[0114] Step S606, based on the uplink packet count, calculate the uplink internal packet loss quantity;

[0115] Step S607, based on the downlink packet count, calculate the downlink internal packet loss quantity;

[0116] Step S608, based on the packet sequence numbers in the downlink data stream, calculate the network-side downlink packet loss quantity;

[0117] Step S609, based on the packet retransmission times in the downlink data stream, calculate the network-side uplink packet loss quantity;

[0118] Step S6010, based on the packet sequence numbers in the uplink data stream, calculate the user-side uplink packet loss quantity;

[0119] Step S6011, based on the packet retransmission times in the uplink data stream, calculate the user-side downlink packet loss quantity;

[0120] Step S6012, based on a preset period, according to the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity, perform application quality degradation analysis and quality degradation location.

[0121] In this embodiment, it is used to perform application quality difference analysis and quality difference location on the data stream communicated using the TCP protocol. The tag field of the second upstream data stream is the first tag (ONU-TAG) + the second tag (UPLINK-TAG), and the tag field in the second downstream data stream is the third tag (UPLINK-TAG) + the fourth tag (ONU-TAG). Among them, the fields in the first tag (ONU-TAG) include the upstream inflow port (Object, the content is slot / pon / onu, used to identify the line card slot / PON port / ONU corresponding to the data stream), the upstream inflow time (Timestamp), and the upstream packet count (Sequence). The fields in the second tag (UPLINK-TAG) include the upstream outflow port (Object, the content is slot / port, used to identify the upstream link slot / port corresponding to the data stream), the upstream outflow time (Timestamp), and the application ID (Application-ID). The fields in the third tag (UPLINK-TAG) include the downstream inflow port (Object, the content is slot / port, used to identify the upstream link slot / port corresponding to the data stream), the downstream inflow time (Timestamp), the downstream packet count (Sequence), and the application ID (Application-ID). The fields in the fourth tag (ONU-TAG) include the downstream outflow port (Object, the content is slot / pon / onu, used to identify the line card slot / PON port / ONU corresponding to the data stream) and the downstream outflow time (Timestamp). The above is the introduction of the field structures of the first tag, the second tag, the third tag, and the fourth tag. The content carried by the fields in each data stream is determined according to the specific data stream. The edge computing board card identifies the second upstream data stream and the second downstream data stream based on the front and back positions of the tag fields added to the packet. For example, if the application ID field is located outside, that is, added later, in the packet, it can be determined as an upstream data stream. If the application ID field is located inside, that is, added first, it can be determined as a downstream data stream. Based on the difference in the content of the port fields in the packet, the occurrence location of the data stream can be identified as being on the user side or the network side. For example, if the content of Object in the packet of a certain data stream is slot / pon / onu, it can be identified that the occurrence location of this data stream belongs to the user side. If the content of Object is slot / port, it can be identified that the occurrence location of this data stream belongs to the network side.

[0122] In this embodiment, refer to Figure 5 , Figure 5 is the schematic diagram of the upstream data stream and the downstream data stream of an embodiment of the application quality difference analysis method of the present invention. As Figure 5As shown, in the TCP communication protocol, if the receiver successfully receives data, it will reply with an ACK (Acknowledge character) data. The packet with the acknowledgenumber = X + Y sent by the receiver is the reply packet for the packet with the sequencenumber (sequence number) of X and the length of Y sent by the sender. The packet with the acknowledgenumber = U + V sent by the receiver is the reply packet for the packet with the sequencenumber of U and the length of V sent by the sender, indicating that the received data has been confirmed to be received correctly. Therefore, after obtaining the information about ACK and sequence number from the TCP data stream packet, the round-trip delay can be calculated through the above confirmation mechanism, and then the inflow and outflow times of the corresponding data stream can be marked.

[0123] Specifically, in the upstream direction, the upstream inflow time T1 is the time when the upstream data stream flows into the OLT, and the upstream outflow time T2 is the time when the upstream data stream flows out of the OLT. Then, the downstream data stream replies to the upstream data stream. The downstream inflow time T3 is the time when the replied downstream data stream flows into the OLT, and the downstream outflow time T4 is the time when the replied downstream data stream flows out of the OLT. Similarly, in the downstream direction, the downstream inflow time T5 is the time when the downstream data stream flows into the OLT, and the downstream outflow time T6 is the time when the downstream data stream flows out of the OLT. Then, the upstream data stream replies to the downstream data stream. The upstream inflow time T7 is the time when the replied upstream data stream flows into the OLT, and the upstream outflow time T8 is the time when the replied upstream data stream flows out of the OLT. Therefore, by marking the inflow and outflow times of the corresponding upstream data stream and downstream data stream, the upstream internal delay, downstream internal delay, network-side delay, and user-side delay can be calculated. The upstream packet count and downstream packet count are self-incrementing fields, and the value of this field automatically increases when each data stream passes through. The internal packet loss numbers of the upstream and downstream during the transmission process can be calculated through the upstream packet count and downstream packet count. Then, based on the upstream internal delay, downstream internal delay, network-side delay, user-side delay, upstream internal packet loss number, downstream internal packet loss number, network-side downstream packet loss number, network-side upstream packet loss number, user-side upstream packet loss number, and user-side downstream packet loss number, application quality difference analysis and quality difference positioning are carried out.

[0124] Specifically, based on a preset period, the number of uplink internal packet losses = the uplink packet count of the last uplink TCP packet within the period - the uplink packet count of the first uplink TCP packet within the period - the total number of actually received uplink TCP packets within the period; the number of downlink internal packet losses = the downlink packet count of the last downlink TCP packet within the period - the downlink packet count of the first downlink TCP packet within the period - the total number of actually received downlink TCP packets within the period; the number of network - side downlink packet losses = the sequence number of the last downlink TCP packet within the period - the sequence number of the first downlink TCP packet within the period+the number of re - transmitted packets of the downlink TCP packet sequence number - the total number of actually received downlink TCP packets within the period; the number of network - side uplink packet losses = the number of re - transmitted packets of the downlink TCP packet sequence number; the number of user - side uplink packet losses = the sequence number of the last uplink TCP packet within the period - the sequence number of the first uplink TCP packet within the period+the number of re - transmitted packets of the uplink TCP packet sequence number - the total number of actually received uplink TCP packets within the period; the number of user - side downlink packet losses = the number of re - transmitted packets of the uplink TCP packet sequence number.

[0125] Further, in one embodiment, step S605 includes:

[0126] According to T1, T2, T3, T4, T5, T6, T7 and T8, the uplink internal delay, downlink internal delay, network - side delay and user - side delay are calculated through formulas, where the calculation formulas are:

[0127] Uplink internal delay = T2 - T1 or T8 - T7;

[0128] Downlink internal delay = T4 - T3 or T6 - T5;

[0129] Network - side delay = T3 - T2;

[0130] User - side delay = T7 - T6.

[0131] In this embodiment, since T1, T2, T3 and T4 respectively represent the uplink inflow time, uplink outflow time, downlink inflow time and downlink outflow time of an uplink data stream and a corresponding replied downlink data stream, and T5, T6, T7 and T8 respectively represent the downlink inflow time, downlink outflow time, uplink inflow time and uplink outflow time of a downlink data stream and a corresponding replied uplink data stream, therefore, the uplink internal delay, downlink internal delay, network - side delay and user - side delay can be calculated through the above formulas.

[0132] Further, in one embodiment, step S6012 includes:

[0133] Determine the application to be analyzed and the quality difference threshold of the application to be analyzed according to the application ID. The quality difference threshold of the application to be analyzed includes the quality difference thresholds corresponding to the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity;

[0134] When the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity of the application to be analyzed exceed the quality difference threshold of the application to be analyzed, it is determined as a single application quality difference;

[0135] Based on a preset period, count the number of times of single application quality difference of the application to be analyzed within the preset period;

[0136] When within the preset period, the number of times of single application quality difference of the uplink internal delay, and / or downlink internal delay, and / or uplink internal packet loss quantity, and / or downlink internal packet loss quantity of the application to be analyzed exceeds the first preset number of times, it is determined as an internal failure. The first preset number of times includes the preset numbers corresponding to the uplink internal delay, downlink internal delay, uplink internal packet loss quantity, and downlink internal packet loss quantity of the application to be analyzed;

[0137] When within the preset period, the number of times of single application quality difference of the network-side delay, and / or network-side downlink packet loss quantity, and / or network-side downlink packet loss quantity of the application to be analyzed exceeds the second preset number of times, it is determined as a network-side failure. The second preset number of times includes the preset numbers corresponding to the network-side delay, network-side downlink packet loss quantity, and network-side downlink packet loss quantity of the application to be analyzed;

[0138] When within the preset period, the number of times of single application quality difference of the user-side delay, and / or user-side uplink packet loss quantity, and / or user-side uplink packet loss quantity of the application to be analyzed exceeds the third preset number of times, it is determined as a user-side failure. The third preset number of times includes the preset numbers corresponding to the user-side delay, user-side uplink packet loss quantity, and user-side uplink packet loss quantity of the application to be analyzed.

[0139] In this embodiment, after calculating the relevant metrics of latency and packet loss for quality difference analysis in step S60, quality difference analysis is performed according to specific applications. First, the thresholds of the relevant metrics for quality difference analysis are determined. When the quality difference threshold of the application to be analyzed is exceeded, it is determined that a single application quality difference has occurred. Based on a preset period, the number of times a single application quality difference occurs for the application to be analyzed within the preset period is statistically analyzed. For example, the number of times a single application quality difference occurs within the period is statistically analyzed based on days, weeks, and months, and can be output as a statistical analysis report. Further, location analysis of the occurrence location of the application quality difference can be performed. The occurrence location of the application quality difference includes user-side faults, network-side faults, and internal faults of the OLT device, that is, it is determined whether the number of times a single application quality difference occurs for the relevant metrics of the corresponding user-side faults, network-side faults, and internal faults of the OLT device within the preset period exceeds the corresponding preset number of times.

[0140] In a third aspect, an embodiment of the present invention further provides an application quality difference analysis device.

[0141] Referring to Figure 6 , Figure 6 is a schematic diagram of the functional modules of an embodiment of the application quality difference analysis device of the present invention.

[0142] In this embodiment, the application quality difference analysis device includes:

[0143] A first upstream module 10, configured to receive an original upstream data stream by a line card, add a first tag to the data frames in the original upstream data stream to obtain a first upstream data stream, and the line card forwards the first upstream data stream to the main control board;

[0144] A second upstream module 20, configured to receive the first upstream data stream by the main control board, add a second tag to the data frames in the first upstream data stream to obtain a second upstream data stream, where the second tag includes an application ID, and the main control board sends the second upstream data stream to the edge computing board;

[0145] A first downstream module 30, configured to receive an original downstream data stream by the main control board, add a third tag including an application ID to the data frames in the original downstream data stream to obtain a first downstream data stream, and the main control board forwards the first downstream data stream to the line card;

[0146] A second downstream module 40, configured to receive the first downstream data stream by the line card, add a fourth tag to the data frames in the first downstream data stream to obtain a second downstream data stream, and the line card sends the second downstream data stream to the edge computing board;

[0147] A receiving module 50, configured to receive the second upstream data stream from the main control board by the edge computing board, determine the priority of the second upstream data stream according to the application ID, receive the second downstream data stream from the line card, and determine the priority of the second downstream data stream according to the application ID;

[0148] The quality difference analysis module 60 is used for the edge computing board to perform application quality difference analysis and quality difference positioning based on the second upstream data stream, the first label and the second label in the second upstream data stream, the second downstream data stream, and the third label and the fourth label in the second downstream data stream according to the priority.

[0149] Further, in an embodiment, the second upstream module 20 is used for:

[0150] The main control board receives the first upstream data stream;

[0151] The main control board determines the application ID based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the first upstream data stream;

[0152] The main control board adds a second label to the data frame in the first upstream data stream to obtain the second upstream data stream, and the second label includes the application ID;

[0153] The first downstream module 30 is used for:

[0154] The main control board receives the original downstream data stream;

[0155] The main control board determines the application ID based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the original downstream data stream;

[0156] The main control board adds a third label to the data frame in the original downstream data stream, and the third label includes the application ID.

[0157] Further, in an embodiment, the application quality difference analysis device further includes a calibration module, which is used for:

[0158] The edge computing board performs application recognition and calibrates the corresponding relationship between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID;

[0159] The edge computing board sends the calibrated corresponding relationship between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID as the application recognition rule to the main control board.

[0160] Further, in an embodiment, the quality difference analysis module 60 includes:

[0161] A defining unit, for the first tag including an upstream inflow port, an upstream inflow time, and an upstream packet count, the second tag further including an upstream outflow port and an upstream outflow time, the third tag further including a downstream inflow port, a downstream inflow time, and a downstream packet count, and the fourth tag including a downstream outflow port and a downstream outflow time;

[0162] An identifying unit, for the edge computing board to identify the second upstream data stream and the second downstream data stream based on the positions of the tag fields in the packet, and identify the user side and the network side based on the contents of the port fields in the packet, where the second upstream data stream and the second downstream data stream are data streams communicating using the TCP protocol;

[0163] A first marking unit, for the edge computing board to mark the upstream inflow time of the packet with sequence number X and length Y in the second upstream data stream as T1, the upstream outflow time as T2, mark the downstream inflow time of the packet with ACK as X+Y in the second downstream data stream as T3, and the downstream outflow time as T4;

[0164] A second marking unit, for the edge computing board to mark the downstream inflow time of the packet with sequence number U and length V in the second downstream data stream as T5, the downstream outflow time as T6, mark the upstream inflow time of the packet with ACK as U+V in the second upstream data stream as T7, and the upstream outflow time as T8;

[0165] A first computing unit, for calculating the upstream internal delay, the downstream internal delay, the network side delay, and the user side delay according to T1, T2, T3, T4, T5, T6, T7, and T8;

[0166] A second computing unit, for calculating the upstream internal packet loss quantity according to the upstream packet count;

[0167] A third computing unit, for calculating the downstream internal packet loss quantity according to the downstream packet count;

[0168] A fourth computing unit, for calculating the network side downstream packet loss quantity according to the packet sequence numbers in the downstream data stream;

[0169] A fifth computing unit, for calculating the network side upstream packet loss quantity according to the packet retransmission times in the downstream data stream;

[0170] A sixth computing unit, for calculating the user side upstream packet loss quantity according to the packet sequence numbers in the upstream data stream;

[0171] A seventh computing unit, for calculating the user side downstream packet loss quantity according to the packet retransmission times in the upstream data stream;

[0172] The quality degradation analysis unit is used to perform application quality degradation analysis and quality degradation location based on a preset period according to the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity.

[0173] Further, in one embodiment, the first calculation unit is used to:

[0174] Calculate the uplink internal delay, downlink internal delay, network-side delay, and user-side delay according to T1, T2, T3, T4, T5, T6, T7, and T8 through a formula, where the calculation formula is:

[0175] Uplink internal delay = T2 - T1 or T8 - T7;

[0176] Downlink internal delay = T4 - T3 or T6 - T5;

[0177] Network-side delay = T3 - T2;

[0178] User-side delay = T7 - T6.

[0179] Further, in one embodiment, the quality degradation analysis unit is used to:

[0180] Determine the application to be analyzed and the quality degradation threshold of the application to be analyzed according to the application ID. The quality degradation threshold of the application to be analyzed includes the quality degradation thresholds corresponding to the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity;

[0181] When the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity of the application to be analyzed exceed the quality degradation threshold of the application to be analyzed, it is determined as a single application quality degradation;

[0182] Based on a preset period, count the number of times of single application quality degradation of the application to be analyzed within the preset period;

[0183] When within the preset period, the number of times of single application quality degradation of the uplink internal delay, and / or downlink internal delay, and / or uplink internal packet loss quantity, and / or downlink internal packet loss quantity of the application to be analyzed exceeds the first preset number, it is determined as an internal failure. The first preset number includes the preset numbers corresponding to the uplink internal delay, downlink internal delay, uplink internal packet loss quantity, and downlink internal packet loss quantity of the application to be analyzed;

[0184] When, within a preset period, the number of times of single-application quality degradation caused by the network-side latency, and / or the network-side downlink packet loss quantity, and / or the network-side downlink packet loss quantity of the application to be analyzed exceeds a second preset number, it is determined as a network-side fault, and the second preset number includes the preset numbers corresponding to the network-side latency, the network-side downlink packet loss quantity, and the network-side downlink packet loss quantity of the application to be analyzed;

[0185] When, within a preset period, the number of times of single-application quality degradation caused by the user-side latency, and / or the user-side uplink packet loss quantity, and / or the user-side uplink packet loss quantity of the application to be analyzed exceeds a third preset number, it is determined as a user-side fault, and the third preset number includes the preset numbers corresponding to the user-side latency, the user-side uplink packet loss quantity, and the user-side uplink packet loss quantity of the application to be analyzed.

[0186] Among them, the function implementation of each module in the above application quality degradation analysis device corresponds to each step in the above application quality degradation analysis method embodiment, and its function and implementation process will not be elaborated here one by one.

[0187] In a fourth aspect, an embodiment of the present invention further provides a readable storage medium.

[0188] An application quality degradation analysis program is stored on the readable storage medium of the present invention. When the application quality degradation analysis program is executed by a processor, the steps of the application quality degradation analysis method as described above are implemented.

[0189] Among them, the method implemented when the application quality degradation analysis program is executed can refer to each embodiment of the application quality degradation analysis method of the present invention, and will not be elaborated here.

[0190] In a fifth aspect, an embodiment of the present invention further provides an application quality degradation analysis system.

[0191] Refer to Figure 7 , Figure 7 which is a schematic diagram of the system architecture of an embodiment of the application quality degradation analysis system of the present invention.

[0192] In this embodiment, the application quality degradation analysis system includes: a line card first switching unit, a main control second switching unit, a main control application identification unit, an edge computing board application mapping engine, an edge computing board application identification engine, and an edge computing board application quality degradation analysis engine. The data processing flow of the application quality degradation analysis system refers to Figure 8 , Figure 8 which is a schematic diagram of the data processing flow of an embodiment of the application quality degradation analysis system of the present invention.

[0193] It should be noted that in this document, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or system comprising a series of elements not only includes those elements but also other elements not expressly listed, or elements inherent to such process, method, article or system. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or system comprising that element.

[0194] The serial numbers of the above embodiments of the present invention are only for description and do not represent the superiority or inferiority of the embodiments.

[0195] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc) as described above and includes several instructions to enable a terminal device to execute the methods described in various embodiments of the present invention.

[0196] The above are only the preferred embodiments of the present invention and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the description and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. An application quality difference analysis method, characterized in that, it includes: The line card receives the original upstream data stream, adds a first label to the data frames in the original upstream data stream to obtain a first upstream data stream, and the line card forwards the first upstream data stream to the main control board; The main control board receives the first upstream data stream, adds a second label to the data frames in the first upstream data stream to obtain a second upstream data stream, the second label includes an application ID, and the main control board sends the second upstream data stream to the edge computing board; The main control board receives the original downstream data stream, adds a third label to the data frames in the original downstream data stream, the third label includes an application ID, to obtain a first downstream data stream, and the main control board forwards the first downstream data stream to the line card; The line card receives the first downstream data stream, adds a fourth label to the data frames in the first downstream data stream to obtain a second downstream data stream, and the line card sends the second downstream data stream to the edge computing board; The edge computing board receives the second upstream data stream from the main control board, determines the priority of the second upstream data stream according to the application ID, receives the second downstream data stream from the line card, and determines the priority of the second downstream data stream according to the application ID; The edge computing board performs application quality difference analysis and quality difference positioning according to the priority, based on the second upstream data stream and the first label and the second label in the second upstream data stream, the second downstream data stream and the third label and the fourth label in the second downstream data stream; When within a preset period, the number of times of single application quality difference of the upstream internal delay, and / or the downstream internal delay, and / or the upstream internal packet loss number, and / or the downstream internal packet loss number of the application to be analyzed exceeds a first preset number, it is determined as an internal fault; When within a preset period, the number of times of single application quality difference of the network-side delay, and / or the network-side downstream packet loss number, and / or the network-side downstream packet loss number of the application to be analyzed exceeds a second preset number, it is determined as a network-side fault; When within a preset period, the number of times of single application quality difference of the user-side delay, and / or the user-side upstream packet loss number, and / or the user-side upstream packet loss number of the application to be analyzed exceeds a third preset number, it is determined as a user-side fault.

2. The application quality difference analysis method according to claim 1, characterized in that, The main control board receives the first upstream data stream, adds a second label to the data frames in the first upstream data stream to obtain a second upstream data stream, and the second label includes an application ID, including: The main control board receives the first upstream data stream; The main control board determines the application ID based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number and destination TCP port number in the first upstream data stream; The main control board adds a second label to the data frames in the first upstream data stream to obtain a second upstream data stream, and the second label includes an application ID.

3. The application quality difference analysis method according to claim 1, characterized in that, The main control board receives the original downstream data stream, adds a third label to the data frames in the original downstream data stream, and the third label includes an application ID, including: The main control board receives the original downstream data stream; The master control board determines the application ID based on the application recognition rule according to the source IP address, destination IP address, IP protocol type, source TCP port number, and destination TCP port number in the original downstream data stream; The master control board adds a third label to the data frame in the original downstream data stream, and the third label includes the application ID.

4. The application quality degradation analysis method according to claim 2 or 3, characterized in that, before the master control board is based on the application recognition rule, it includes: The edge computing power board performs application recognition and calibrates the corresponding relationship between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID; The edge computing power board sends the calibrated corresponding relationship between the source IP address, destination IP address, IP protocol type, source TCP port number, destination TCP port number, and application ID to the master control board as the application recognition rule.

5. The application quality degradation analysis method according to claim 1, characterized in that, The application quality degradation analysis and quality degradation positioning based on the second upstream data stream and the first label and the second label in the second upstream data stream, the second downstream data stream and the third label and the fourth label in the second downstream data stream include: The first label includes the upstream inflow port, upstream inflow time, and upstream packet count. The second label further includes the upstream outflow port and upstream outflow time. The third label further includes the downstream inflow port, downstream inflow time, and downstream packet count. The fourth label includes the downstream outflow port and downstream outflow time; The edge computing power board identifies the second upstream data stream and the second downstream data stream based on the position of the label field in the packet, and identifies the user side and the network side based on the content of the port field in the packet. The second upstream data stream and the second downstream data stream are data streams for communication using the TCP protocol; The edge computing power board marks the upstream inflow time of the packet with sequence number X and length Y in the second upstream data stream as T1, and the upstream outflow time as T2. It marks the downstream inflow time of the packet with ACK as X+Y in the second downstream data stream as T3, and the downstream outflow time as T4; The edge computing power board marks the downstream inflow time of the packet with sequence number U and length V in the second downstream data stream as T5, and the downstream outflow time as T6. It marks the upstream inflow time of the packet with ACK as U+V in the second upstream data stream as T7, and the upstream outflow time as T8; According to T1, T2, T3, T4, T5, T6, T7, and T8, calculate the upstream internal delay, downstream internal delay, network side delay, and user side delay; According to the upstream packet count, calculate the upstream internal packet loss quantity; According to the downstream packet count, calculate the downstream internal packet loss quantity; According to the packet sequence number in the downstream data stream, calculate the network side downstream packet loss quantity; According to the packet retransmission times in the downstream data stream, calculate the network side upstream packet loss quantity; According to the packet sequence number in the upstream data stream, calculate the user side upstream packet loss quantity; According to the packet retransmission times in the upstream data stream, calculate the user side downstream packet loss quantity; Based on a preset period, perform application quality degradation analysis and quality degradation location according to the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity.

6. The application quality degradation analysis method according to claim 5, wherein, the calculating the uplink internal delay, downlink internal delay, network-side delay, and user-side delay according to T1, T2, T3, T4, T5, T6, T7, and T8 includes: According to T1, T2, T3, T4, T5, T6, T7, and T8, calculate the uplink internal delay, downlink internal delay, network-side delay, and user-side delay through a formula, where the calculation formula is: Uplink internal delay = T2 - T1 or T8 - T7; Downlink internal delay = T4 - T3 or T6 - T5; Network-side delay = T3 - T2; User-side delay = T7 - T6.

7. The application quality degradation analysis method according to claim 5, wherein, the performing application quality degradation analysis and quality degradation location based on a preset period according to the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity includes: Determine the application to be analyzed and the quality degradation threshold of the application to be analyzed according to the application ID, and the quality degradation threshold of the application to be analyzed includes the quality degradation thresholds corresponding to the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity; When the uplink internal delay, downlink internal delay, network-side delay, user-side delay, uplink internal packet loss quantity, downlink internal packet loss quantity, network-side downlink packet loss quantity, network-side uplink packet loss quantity, user-side uplink packet loss quantity, and user-side downlink packet loss quantity of the application to be analyzed exceed the quality degradation threshold of the application to be analyzed, it is determined as a single application quality degradation; Based on a preset period, count the number of times of single application quality degradation of the application to be analyzed within the preset period; When within the preset period, the number of times of single application quality degradation of the uplink internal delay, and / or downlink internal delay, and / or uplink internal packet loss quantity, and / or downlink internal packet loss quantity of the application to be analyzed exceeds the first preset number of times, it is determined as an internal failure, and the first preset number of times includes the preset numbers corresponding to the uplink internal delay, downlink internal delay, uplink internal packet loss quantity, and downlink internal packet loss quantity of the application to be analyzed; When within the preset period, the number of times of single application quality degradation of the network-side delay, and / or network-side downlink packet loss quantity, and / or network-side downlink packet loss quantity of the application to be analyzed exceeds the second preset number of times, it is determined as a network-side failure, and the second preset number of times includes the preset numbers corresponding to the network-side delay, network-side downlink packet loss quantity, and network-side downlink packet loss quantity of the application to be analyzed; When, within a preset period, the number of times of single-application quality degradation of the user-side latency, and / or the number of uplink packet losses on the user side, and / or the number of uplink packet losses on the user side of the application to be analyzed exceeds a third preset number, it is determined as a user-side fault. The third preset number includes the preset numbers corresponding to the user-side latency, the number of uplink packet losses on the user side, and the number of uplink packet losses on the user side of the application to be analyzed.

8. An application quality degradation analysis device, characterized in that, the application quality degradation analysis device includes: A first uplink module, configured to receive the original uplink data stream by a line card, add a first tag to the data frames in the original uplink data stream to obtain a first uplink data stream, and the line card forwards the first uplink data stream to the main control board; A second uplink module, configured to receive the first uplink data stream by the main control board, add a second tag to the data frames in the first uplink data stream to obtain a second uplink data stream, where the second tag includes an application ID, and the main control board sends the second uplink data stream to the edge computing board; A first downlink module, configured to receive the original downlink data stream by the main control board, add a third tag including an application ID to the data frames in the original downlink data stream to obtain a first downlink data stream, and the main control board forwards the first downlink data stream to the line card; A second downlink module, configured to receive the first downlink data stream by the line card, add a fourth tag to the data frames in the first downlink data stream to obtain a second downlink data stream, and the line card sends the second downlink data stream to the edge computing board; A receiving module, configured to receive the second uplink data stream from the main control board by the edge computing board, determine the priority of the second uplink data stream according to the application ID, receive the second downlink data stream from the line card, and determine the priority of the second downlink data stream according to the application ID; A quality degradation analysis module, configured to perform application quality degradation analysis and quality degradation positioning by the edge computing board according to the priority, based on the second uplink data stream and the first and second tags in the second uplink data stream, and the second downlink data stream and the third and fourth tags in the second downlink data stream; A quality degradation analysis unit, configured to: When, within a preset period, the number of times of single-application quality degradation of the uplink internal latency, and / or the downlink internal latency, and / or the number of uplink internal packet losses, and / or the number of downlink internal packet losses of the application to be analyzed exceeds a first preset number, it is determined as an internal fault; When, within a preset period, the number of times of single-application quality degradation of the network-side latency, and / or the number of network-side downlink packet losses, and / or the number of network-side downlink packet losses of the application to be analyzed exceeds a second preset number, it is determined as a network-side fault; When, within a preset period, the number of times of single-application quality degradation of the user-side latency, and / or the number of user-side uplink packet losses, and / or the number of user-side uplink packet losses of the application to be analyzed exceeds a third preset number, it is determined as a user-side fault.

9. An application quality degradation analysis device, characterized in that, the application quality degradation analysis device includes a processor, a memory, and an application quality degradation analysis program stored on the memory and executable by the processor. When the application quality degradation analysis program is executed by the processor, it implements the steps of the application quality degradation analysis method according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that, an application quality difference analysis program is stored on the readable storage medium, and when the application quality difference analysis program is executed by a processor, the steps of the application quality difference analysis method according to any one of claims 1 to 7 are implemented.

Images