CN116886631A - Routing equipment supporting detection mode - Google Patents

Routing equipment supporting detection mode Download PDF

Info

Publication number
CN116886631A
CN116886631A CN202310833791.5A CN202310833791A CN116886631A CN 116886631 A CN116886631 A CN 116886631A CN 202310833791 A CN202310833791 A CN 202310833791A CN 116886631 A CN116886631 A CN 116886631A
Authority
CN
China
Prior art keywords
data stream
mode
queue
routing device
service
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310833791.5A
Other languages
Chinese (zh)
Inventor
刘畅
王泽林
徐博华
韩博文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China United Network Communications Group Co Ltd
Original Assignee
China United Network Communications Group Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China United Network Communications Group Co Ltd filed Critical China United Network Communications Group Co Ltd
Priority to CN202310833791.5A priority Critical patent/CN116886631A/en
Publication of CN116886631A publication Critical patent/CN116886631A/en
Pending legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/24Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2425Traffic characterised by specific attributes, e.g. priority or QoS for supporting services specification, e.g. SLA
    • H04L47/2433Allocation of priorities to traffic types
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/26Route discovery packet
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/31Flow control; Congestion control by tagging of packets, e.g. using discard eligibility [DE] bits
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/32Flow control; Congestion control by discarding or delaying data units, e.g. packets or frames
    • H04L47/323Discarding or blocking control packets, e.g. ACK packets

Abstract

The application provides a routing device supporting a detection mode, which relates to the technical field of communication, and can enable a network/device to detect actual operation data routing device under the condition of high load to support the detection mode on the basis of supporting a normal mode on the premise of not influencing the transmission efficiency of actual service; the normal mode is a mode of the routing equipment in a normal working state; the detection mode is used for generating a detection data stream and differentially processing the service data stream and the detection data stream; the routing device is applicable to a data transmission method, the method comprising: determining a mode of the routing equipment; determining a service data flow and a detection data flow in received data under the condition that the routing equipment is in a detection mode; the priority of the service data stream is higher than the priority of the detection data stream; forwarding the traffic data stream to the routing device in normal mode or the routing device not supporting the probing mode, discarding the probing data stream.

Description

Routing equipment supporting detection mode
Technical Field
The present application relates to the field of communications technologies, and in particular, to a routing device supporting a probing mode.
Background
At present, the functions of the router/switch are very complex and the openness is very high, and a large number of adjustable parameter configurations are generated accordingly, so that the optimal transmission effect can be realized according to different requirement scenes. From the introduction of AI/adaptive/self-learning concepts into networks in recent years, routers/switches are also gradually trying to use adaptive/self-learning algorithms for autonomous optimization to adapt to the traffic scenario in which the device is currently located.
In actual use of the network, in order to ensure stability and transmission efficiency of the network, the network is usually not allowed to reach a high-load running condition, so that actual data under the high-load condition is difficult to generate by itself, and therefore, the actual data aiming at the high-load condition in the training set of the AI/self-adaptive/self-learning algorithm is often missing or insufficient, and the prediction accuracy of the training algorithm for the high-load condition can be seriously affected. However, in actual network operation and maintenance, one of the most often concerns is the desire to be able to predict how much congestion affects the efficiency of network operation under high load conditions. Furthermore, how to make the network/device detect the actual operation data under the high load condition without affecting the actual service transmission efficiency is a current urgent problem to be solved.
Disclosure of Invention
The application provides a routing device supporting a detection mode, which can enable a network/device to detect actual operation data under a high load condition on the premise of not influencing the actual service transmission efficiency.
In order to achieve the above purpose, the application adopts the following technical scheme:
in a first aspect, the present application provides a routing device supporting a probing mode, where the routing device further supports the probing mode on the basis of supporting a normal mode; the normal mode is a mode of the routing equipment in a normal working state; the detection mode is used for generating a detection data stream and differentially processing a service data stream and the detection data stream; the routing device is applicable to a data transmission method, the method comprising: determining a mode of the routing equipment; determining a service data flow and a detection data flow in the received data under the condition that the routing equipment is in the detection mode; the priority of the service data stream is higher than the priority of the detection data stream; forwarding the service data stream to the routing device in the normal mode or the routing device not supporting the probing mode, and discarding the probing data stream.
With reference to the first aspect, in one possible implementation manner, in a case of mode configuration with device granularity, the routing device being in the normal mode includes: all interfaces of the routing device are in the normal mode; the routing device being in the probing mode comprises: all interfaces of the routing device are in the probing mode; in the case of mode configuration with granularity of interfaces, each interface of the routing device is independently in the normal mode or the probing mode; and if the input interface is in the detection mode and the output interface is in the normal mode, the routing equipment discards the detection data stream in the data stream.
With reference to the first aspect, in one possible implementation manner, whether the operation mode of the routing device/interface is the probing mode is determined by a configuration mode or a protocol negotiation mode; the manner of configuration/the manner of protocol negotiation may specify a trigger condition; the conditional configuration includes a time period and/or a resource utilization.
With reference to the first aspect, in a possible implementation manner, the detection data stream generated by the detection mode is determined based on the service data stream and a specified multiplying power; the packet characteristics of the probe data stream are consistent with the packet characteristics of the traffic data stream.
With reference to the first aspect, in a possible implementation manner, the detecting data flow generated by the detecting mode is determined based on the service data flow and the specified multiplying power, and includes: determining a service data flow which flows through the routing equipment inlet interface or accords with the screening rule; determining a detection data stream based on the service data stream subjected to re-engraving of the re-engraving multiplying power; wherein the re-engraving function is active when the routing device/interface is in the probing mode.
With reference to the first aspect, in a possible implementation manner, after the determining the probe data flow based on the multi-engraving rate for multi-engraving the service data flow, the method further includes: marking the detected data stream; the marking mode comprises the following steps: setting a specific field in the probing data stream to a specific value or adding an optional extension field in the probing data stream.
With reference to the first aspect, in a possible implementation manner, after the determining a probe data flow, the method further includes: modifying a destination address for the probe data stream or not modifying a destination address for the probe data stream; the modified destination address of the detection data stream is a service address or a black hole address; the black hole address is used to offload the probe data stream out of the network.
With reference to the first aspect, in a possible implementation manner, a binding queue is provided in the routing device; the binding queue comprises a first queue for bearing business data flows and a second queue for bearing marked detection data flows; the priority of the first queue is higher than the priority of the second queue; the routing equipment is configured according to the parameters of the queue, and takes effect according to the binding queue; the parameter configuration of the binding queue comprises a discard rule during enqueue and a speed limit rule during dequeue.
With reference to the first aspect, in one possible implementation manner, when the first queue receives a service data stream, determining a discard rule during enqueuing by using a current occupation amount of the first queue as a reference amount; when the second queue receives the detection data stream, the sum of the current occupation amount of the first queue and the current occupation amount of the second queue is used as a reference amount, and discard rule judgment during enqueuing is carried out; the discard rule adopts the parameter configuration of the binding queue.
With reference to the first aspect, in a possible implementation manner, each of the probing data flows is configured with an enqueue delay; the enqueue delay is used for deferring the time of the probe data flow entering the second queue.
In the present application, the names of the above-mentioned routing devices do not constitute limitations on the devices or function modules themselves, and in actual implementation, these devices or function modules may appear under other names. Insofar as the function of each device or function module is similar to that of the present application, it falls within the scope of the claims of the present application and the equivalents thereof.
These and other aspects of the application will be more readily apparent from the following description.
Based on the technical scheme, the routing equipment provided by the embodiment of the application supports the normal mode and the detection mode at the same time, and actively forms the network congestion state by copying the service data flow; when the working state of the routing equipment is the detection mode, after the service data stream and the detection data stream are received, the service data stream and the detection data stream can be processed in a differentiated mode, that is, the actual transmission efficiency under the condition of high detection load can be avoided, and meanwhile, the influence of detection on the transmission efficiency of the actual service can be avoided.
Drawings
Fig. 1 is a schematic structural diagram of a data transmission system applied to a routing device according to the present application;
FIG. 2 is a schematic diagram of a data transmission according to the present application;
FIG. 3 is a schematic diagram of a queue according to the present application;
fig. 4 is a flowchart of a data transmission method applied to a routing device according to the present application;
FIG. 5 is a flowchart of another data transmission method according to the present application;
fig. 6 is a schematic diagram of a data discarding method according to the present application;
FIG. 7 is a flowchart of another data transmission method according to the present application;
FIG. 8 is a flowchart of another data transmission method according to the present application;
FIG. 9 is a flowchart of another data transmission method according to the present application;
fig. 10 is a flowchart of another data transmission method provided by the present application.
Detailed Description
The following describes in detail a routing device supporting a probe mode according to an embodiment of the present application with reference to the accompanying drawings.
The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone.
The terms "first" and "second" and the like in the description and in the drawings are used for distinguishing between different objects or between different processes of the same object and not for describing a particular order of objects.
Furthermore, references to the terms "comprising" and "having" and any variations thereof in the description of the present application are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or apparatus.
It should be noted that, in the embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "e.g." in an embodiment should not be taken as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.
As is well known, routers/switches are the core building blocks of a network, which process data flows with granularity of packets, mainly through store and forward modes. The current router/switch has very complex functions and high openness, and a large number of adjustable parameter configurations are generated accordingly to cope with different demand scenes, especially the current network technology has been developed very deeply, the demand scenes are more and more finely subdivided, and many businesses hope to find the optimal parameter combinations according to the specific business modes so as to realize the optimal transmission effect. From the introduction of AI/adaptive/self-learning concepts into networks in recent years, routers/switches are also gradually trying to use adaptive/self-learning algorithms for autonomous optimization to adapt to the traffic scenario in which the device is currently located.
At present, the AI algorithm needs to be trained based on data, and a network between routers/switches usually has multiple protection mechanisms, so that the network is not allowed to enter abnormal working states such as overload, and particularly a top-quality network for bearing large clients/high-quality private lines, and a very low utilization rate is often maintained for ensuring an optimal transmission effect. The data available by the current AI algorithm is limited to the operation data in the normal working state, so that the parameter combination trained by the AI cannot be well adapted to the never-seen high-load condition in practice. That is, AI training algorithms are often missing for high load data, while the original purpose of many parameter designs is to provide safeguards in extremely high load scenarios in order to cope with such extreme scenarios. However, AI training results which are not currently referenced to extreme data deviate from the original design of parameters and no adaptation exists in terms of values.
In the related art, one way to generate high-load data is to artificially mix some historical extreme data from outside to form a training set, but the disadvantage of the above way is that the historical extreme data is not generated by the network/device itself, and there is a high probability that there is a problem of parameter mismatch; the other is the need to configure the high load data that is missing in the extreme case of the network/device, but it is very easy to cause the actual traffic of the network/device to be affected during the data transmission process. Therefore, how to avoid the influence of the actual services of the network/device in the data transmission process is a current urgent problem to be solved.
Firstly, the application provides a routing device supporting a detection mode, wherein the routing device also supports the detection mode on the basis of supporting a normal mode; the normal mode is a mode of the routing equipment in a normal working state; the detection mode is used for generating a detection data stream and differentially processing a service data stream and the detection data stream; the routing device is applicable to a data transmission method, the method comprising: determining a mode of the routing equipment; determining a service data flow and a detection data flow in the received data under the condition that the routing equipment is in the detection mode; the priority of the service data stream is higher than the priority of the detection data stream; forwarding the service data stream to the routing device in the normal mode or the routing device not supporting the probing mode, and discarding the probing data stream.
In a possible implementation manner, the routing device provided by the embodiment of the present application generates a probe data flow in a probe mode, where the probe data flow is established in a subsequent processing process without affecting an actual traffic flow, that is, controlling congestion effects to the probe data flow.
Optionally, in the case of mode configuration with device granularity, the routing device being in the normal mode includes: all interfaces of the routing device are in a normal mode; the routing device being in the probe mode includes: all interfaces of the routing device are in a probing mode; under the condition that the interfaces are used for mode configuration at granularity, each interface of the routing equipment is independently in a normal mode or a detection mode; if the ingress interface is in the probe mode and the egress interface is in the normal mode, the routing device discards the probe data stream in the data stream.
In one possible implementation, when the mode configuration is performed with the device granularity, the working mode of the routing device may be directly configured, and then the interface enters the working mode, as shown in fig. 1, where the salesman device 101 and the first routing device 102 may be configured with the device granularity or may be configured with the interface granularity. When the mode configuration is performed with the granularity of the interface, the working mode of the interface can be directly configured, the second routing device 103 can be configured with the granularity of the interface, the output interface of the second routing device 103 is in a normal mode, and the input interface is in a detection mode.
Optionally, determining whether the working mode of the routing device/interface is a probing mode by a configuration mode or a protocol negotiation mode; the manner of configuration/manner of protocol negotiation may specify a trigger condition; the conditional configuration includes a time period and/or a resource utilization.
As a possible implementation, the condition configuration may be that 1-4 a.m. the utilization is lower than 10%, that is, in the case that 1-4 a.m. or the device routing rate is lower than 10%, the routing device/interface enters the probing mode. Or protocol negotiation between two adjacent routing devices/interfaces.
Optionally, the detection data stream generated by the detection mode is determined based on the service data stream and the designated multiplying power; the packet characteristics of the probe data stream are consistent with the packet characteristics of the traffic data stream.
Optionally, the detecting data stream generated by the detecting mode is determined based on the service data stream and the designated multiplying power, and includes: determining a business data flow which flows through an access interface of the routing equipment or accords with a screening rule; determining a detection data stream based on the re-engraving multiplying power re-engraving service data stream; wherein the re-engraving function is active when the routing device/interface is in the probing mode.
In one possible implementation, as shown in fig. 1, after the service source device 101 enters the probe mode, the service data stream is first copied according to a 1:1 magnification and maintained for 5 minutes; copying the business data stream according to the multiplying power of 1:2, and maintaining for 5 minutes; copying the business data stream according to the multiplying power of 1:4, and maintaining for 5 minutes; copying the business data stream according to the multiplying power of 1:8, and maintaining for 5 minutes; then circulating, and copying the service data flow again according to the multiplying power of 1:1; the loop is continued until the source device 101 exits the probing mode, and the traffic copy is disabled at the same time.
Optionally, after determining the probe data stream based on the multi-magnification multi-engraving service data stream, the method further includes: marking the detected data stream; the marking mode comprises the following steps: setting a specific field in the probing data stream to a specific value or adding an optional extension field in the probing data stream.
In one possible implementation, as shown in fig. 1, after the service source device 101 determines the probe data flow, it marks the probe data flow, so that the subsequent routing device can distinguish the probe data flow from the service data flow.
Optionally, after the determining the probe data stream, the method further includes: modifying a destination address for the probe data stream or not modifying a destination address for the probe data stream; the modified destination address of the detection data stream is a service address or a black hole address; the black hole address is used to offload the probe data stream out of the network.
In one possible implementation, as shown in fig. 1, the service source device 101 marks the probe data flow and may also modify the destination address of the probe data flow, so that the second routing device 103 offloads the probe data flow from the network. It will be appreciated that the source device 101 may not modify the destination address of the probe data stream, and the second routing device 103 may identify the probe data stream by a tag.
Optionally, a binding queue is arranged in the routing device; the binding queue comprises a first queue for bearing service data flows and a second queue for bearing marked detection data flows; the priority of the first queue is higher than the priority of the second queue; further ensuring that the detection data stream does not influence the service data stream; the routing equipment is configured according to the parameters of the queue and takes effect according to the binding queue; wherein the parameter configuration of the binding queue comprises a discard rule during enqueue and a speed limit rule during dequeue.
Optionally, when the first queue receives the service data stream, the current occupation amount of the first queue is used as a reference amount to judge a discarding rule during enqueuing; when the second queue receives the detection data stream, the sum of the current occupation amount of the first queue and the current occupation amount of the second queue is used as a reference amount, and the discarding rule during enqueuing is judged; the discard rule uses the parameter configuration of the binding queue.
Optionally, each probing data stream is configured with an enqueue delay; the enqueue delay is used to delay the time of probing the data stream into the second queue.
Based on the technical scheme, the routing equipment provided by the embodiment of the application supports the normal mode and the detection mode at the same time, and actively forms the network congestion state by copying the service data flow; when the working state of the routing equipment is the detection mode, after the service data stream and the detection data stream are received, the service data stream and the detection data stream can be processed in a differentiated mode, that is, the actual transmission efficiency under the condition of high detection load can be avoided, and meanwhile, the influence of detection on the transmission efficiency of the actual service can be avoided.
Referring to fig. 1, a schematic structure of a data transmission system 100 applied to a routing device is provided in an embodiment of the present application, where the system includes a plurality of routing devices; the plurality of routing devices may be: at least one source device 101, a first routing device 102, a second routing device 103, and a user device 104.
The operation states of the above devices may be divided into a normal mode/a probe mode, wherein the operation state of the user equipment 104 is always in the normal mode. In the case that the working states of at least one service source device 101, the first routing device 102 and the second routing device 103 are in the normal mode, the first queue in each device is a Qos queue, and has its independent queue parameters, including a queue depth, a WRED parameter (high/low threshold, drop probability), and a CAR speed limit parameter (cir/pir). The application can configure the condition of each device entering the detection mode, and the condition can be the condition of a preset time period and/or the device utilization rate. The probing mode may be entered when the at least one source device 101, the first routing device 102, the second routing device 103 meet the conditions (e.g., the device attempts to enter the probing mode in case 1-4 a.m. and/or the device utilization is below 10%). It should be noted that, after each device satisfies the condition and enters the probing mode, the interface of the device and the interface of the opposite device may negotiate through the protocol, and when both ends of the link agree to enter the probing mode, the dual-end interface may be simultaneously converted into the probing mode, that is, the interface satisfying the protocol may enter the probing mode. Meanwhile, each device can adjust the working state of the own interface, so that the own input interface and output interface enter a detection mode.
The service source device 101 may copy the service data stream to form a probe data stream, which may be first according to 1: copying the business data stream at 1 multiplying power, and maintaining for 5 minutes; then according to 1:2 multiplying power copying business flow, maintaining for 5 minutes; and so on to 1: after 8 times, the cycle is carried out, and the process is carried out again according to 1: the magnification of 1 replicates the traffic data stream, loops until the traffic source device 101 exits the probing mode under conditions that do not satisfy the probing mode. Notably, the traffic replication function is directly associated to the interface, and traffic replication operations are performed on all traffic flowing in by the interface; traffic replication functions may also be associated with specific screening rules (e.g., acl screening of policy policies) and traffic replication operations may be performed only on specific incoming flows that meet the screening rules. When the flow is replicated and n copies of the flow are replicated according to the appointed multiplying power and enter the second queue, delay (appointed or random delay) can be added for the replicated message, so that the replicated message enters the second queue after a specific time. Therefore, the simulation service can be generated successively, and the simulation service is not only concurrent at the same time. At least one service source device 101 sends a service data stream and a probe data stream to a first routing device 102. In the process of carrying out the re-engraving on the service data stream, the message marking is carried out on the detection data stream, the specific field in the detection data stream can be set to be a specific numerical value, and a special optional extension field can be added in the detection data stream. And the destination address of the probe data stream may be changed to a black hole address. This message marking (whether modifying a specific field or adding a special optional extension field) only allows the device supporting the present application to identify, thus further requiring the second routing device to actively discard the probe data stream when forwarding traffic to the user device.
The first routing device 102 is configured to preferentially discard the probe data flow and/or the traffic data flow when traffic is congested, and forward the traffic data flow and the probe data flow to the second routing device.
The second routing device 103 may distinguish between the traffic data stream and the probe data stream according to the two data types after receiving the traffic data stream and the probe data stream of the first routing device 102, or actively discard the probe data stream and forward only the traffic data stream to the user device 104.
It will be appreciated that the interaction between the service source device 101, the first routing device 102, and the second routing device 103 is mainly through store and forward, and uses a queue) +scheduling mechanism to control the priority processing order (Qos mechanism) of the data packets. As shown in fig. 2, a packet in a traffic flow is received by an ingress interface of a device, is processed by a packet in an ingress direction, a queue is selected according to a packet attribute, then is switched to a designated egress interface by a switching network 201, is processed in an egress direction, and finally, is sent out through the egress interface. The congestion avoidance mechanism (discarding strategy) of the data of the device in the application when the data is enqueued and the queue scheduling algorithm when the data is dequeued affect the packet loss/queuing behavior of the queue together.
At least one of the source device 101, the first routing device 102, and the second routing device 103 has a logical binding queue 301; the logical binding queue 301 includes a first queue 302 for carrying traffic data flows, a second queue 303 for carrying probe data flows, and a scheduler 304, as shown in fig. 3, described in detail below. After the working state of at least one service source device 101 enters the detection mode, the first queue and the second queue of the service source device 101 are bundled to form a logic bundling queue, the first queue carries service data flows, the second queue carries detection data flows, and the scheduler controls the priority of the service data flows in the first queue to be higher than the priority of the detection data flows in the second queue so as to achieve the effect of carrying the service data flows preferentially.
The configuration of the parameters for the queues in each device (e.g., queue depth, WRED, speed limit, etc.) is valid for the logical bundle queue 301 as a logically shared parameter for the first queue 302 and the second queue 303. The first queue 302 remains unchanged (index tag unchanged, occupancy counter unchanged) while the device switches to probe mode, but the configuration parameters cancel (queue depth, WRED parameter, CAR speed limit parameter). A second queue 303 (retrieval tag, occupancy counter) is newly configured, which has no parameters (queue depth, WRED parameter, CAR speed limit parameter). Scheduler 304, first queue 302 is mounted on this scheduler 304 with the highest priority, and second queue 303 is mounted on this scheduler 304 with a lower priority than first queue 302. The logical binding queue 301 (retrieval tag, occupancy counter) is newly configured, and this logical binding queue 301 inherits the parameters of the original first queue 302 (queue depth, WRED parameter, CAR speed limit parameter). Logical binding queue 301 is associated with a first queue 302, a second queue 303, and a scheduler 304. Then establishing a dynamic update relationship, wherein the occupation counter of the logic binding queue 301 is equal to the sum of the occupation counter of the first queue/2; the scheduler exit acts as a logical bundle queue 301 exit; detecting at the entrance of the logical binding queue 301, detecting that the probe data flow mark is False pointing to the first queue 302, and detecting that the probe data flow mark is True pointing to the second queue 303; in the first queue 301 discarding rule, the occupation amount counter adopts a counter of the first queue 302, the discrimination threshold parameter adopts a parameter (queue depth, WRED parameter) of the logic binding queue 301, and in the second queue 303 discarding rule, the occupation amount counter adopts a counter of the logic binding queue 301, and the discrimination threshold parameter adopts a parameter (queue depth, WRED parameter) of the logic binding queue 301.
As shown in fig. 4, a flowchart of a data transmission method provided by an embodiment of the present application is shown, where the data transmission method is applied to a routing device, and the data transmission method provided by the embodiment of the present application may be applied to the data transmission system shown in fig. 1, and the data transmission method provided by the embodiment of the present application may be implemented by the following steps.
S401, at least one service source device sends a service flow to a first routing device. Correspondingly, the first routing device receives the service flow sent by at least one service source device.
Wherein the service flow comprises a service data flow and a detection data flow; the detection data stream is determined by multiple business data streams according to specified multiplying power.
In one possible implementation manner, at least one service source device and the first routing device enter a detection mode, the service source device replicates the service data stream according to a specific multiplying power to form a detection data stream, and the service data stream can be understood as an actual service data stream in the device.
S402, discarding the service flow to generate a discarded service flow when the sum of the unoccupied amounts of the first queue and the second queue is smaller than the service flow and/or the unoccupied amount of the first queue is smaller than the service data flow.
The first service flow is a service data flow received by the first routing equipment and a service data flow to be sent in the first routing equipment; the first detection flow is a detection data flow received by the first routing equipment and a detection data flow to be sent in the first routing equipment.
As a possible implementation manner, the implementation process of S402 may be: after receiving the service data stream and the detection data stream, the first routing device enters a first queue for the service data stream and enters a second queue for the detection data stream; judging whether the sum of the current unoccupied amounts of the first queue and the second queue can meet the detection data flow, and discarding the detection data flow in the service flow if the sum cannot meet the detection data flow; and secondly, judging whether the current unoccupied quantity of the first queue can meet the service data flow, if not, discarding the service data flow in the service flow, and further generating the discarded service flow.
It should be noted that, when the first routing device provided in the embodiment of the present application reaches the data critical value, the probe data stream is discarded preferentially, so as to ensure normal transmission of the service data stream.
S403, the first routing device sends the discarded service flow to the second routing device. Correspondingly, the second routing device receives the discarded service flow sent by the first routing device.
S404, the second routing equipment sends the service data stream in the discarded service stream to the user equipment. Correspondingly, the user equipment receives the service data stream in the discarded service stream sent by the second routing equipment.
In one possible implementation, since the user equipment does not need to receive the probe data stream in the dropped service stream, the second routing device only needs to send the service data stream in the dropped service stream to the user equipment.
Based on the above technical solution, in the data transmission method provided by the embodiment of the present application, firstly, the data transmission device receives a service flow sent by at least one service source device, where a detected data flow is determined by multiple service data flows, and by copying the service data flows, a network congestion state is actively formed, so that the problem of low AI training accuracy caused by different device parameter configurations is avoided; under the condition that the sum of the length of the first service flow and the length of the first detection flow is larger than a preset threshold value, the data transmission device discards the service flow, so that the service data flow is not influenced by the detection data flow, and normal transmission can be realized; and finally, sending the discarded service flow to the second routing equipment.
In a possible implementation manner, as shown in fig. 5 in connection with fig. 4, in the case where the sum of the unoccupied amounts of the first queue and the second queue is smaller than the traffic flow and/or the unoccupied amount of the first queue is smaller than the traffic data flow, the discarding process is performed on the traffic flow to generate a discarded traffic flow, which may be specifically implemented in the following S501-S502. The queue discarding rule includes two threshold values, which are respectively: a first threshold value and a second threshold value; the first threshold value is greater than the second threshold value.
After the detected data flow enters the second queue, if the sum of the occupation amounts of the first queue and the second queue is greater than a first threshold value of the binding logic queue depth, executing S501; after the probe data stream enters the second queue, the sum of the occupation amounts of the first queue and the second queue is greater than the second threshold value and less than the first threshold value, and S502 is executed.
It is noted that in the discard rule determination, the probe data stream is not actually enqueued, here the data discard prior to enqueuing.
S501, the first routing device discards all data in the detection data stream and generates a discarded service stream.
The first routing device receives a service data stream and a probe data stream of the service source device, where a bearer priority of the service data stream is higher than a bearer priority of the probe data stream; first, a first routing device judges whether the sum of a current service data stream, a service data stream to be transmitted, a current detection data stream and a detection data stream to be transmitted is larger than a first threshold value. If so, discarding the current detection data flow.
S502, the first routing equipment discards partial data in the detection data stream and generates a discarded service stream.
In combination with the example in S501, if the sum of the current service data stream and the service data stream to be transmitted, and the current probe data stream and the probe data stream to be transmitted is smaller than the first threshold value but larger than the second threshold value, the random partial discarding is performed on the data in the current probe data stream, so as to ensure that the current service data stream and the service data stream to be transmitted are not affected by the probe data stream.
It will be appreciated that embodiments of the present application employ congestion avoidance (discard strategy), either by tail-drop or weighted random early detection (WRED, weighted Random Early Detection). As shown in fig. 6, WRED is mainly to randomly discard the arriving data stream when the queue occupancy (queue occupancy) starts to rise (but before congestion actually occurs), so that a small queue is always maintained. When the length of the queue is less than the low limit, the message is not discarded; when the length of the queue is between the low threshold (second threshold) and the high threshold (first threshold), the WRED starts to discard the message randomly; and discarding messages of all data streams when the length of the queue is greater than a high threshold (a first threshold). The end-of-queue discard threshold may be understood as a first threshold, and discard is performed when the length of a newly arrived message and a message to be sent in the queue exceeds the first threshold.
Based on the technical scheme, the method and the device prevent the service data flow of the user equipment from being influenced when the equipment performs the detection mode by setting the preset threshold value on the first routing equipment.
After the business data flow enters the first queue, the occupation amount of the first queue is larger than the depth of the queue.
In a possible implementation manner, as shown in fig. 7 in connection with fig. 4, in the case where the unoccupied amount of the first queue is smaller than the service data flow, the discarding process is performed on the service flow, so as to generate a discarded service flow, which may be specifically implemented in the following S701-S702. The queue discarding rule includes two threshold values, which are respectively: a third threshold value and a fourth threshold value; the third threshold value is greater than the fourth threshold value.
After the service data flow enters the first queue, if the occupation amount of the first queue is greater than a third threshold value of the binding logic queue depth, S701 is executed; after the probe data flow enters the second queue, the occupation amount of the first queue is greater than the fourth threshold value and less than the third threshold value, and S702 is executed.
S701, the first routing device discards all data in the service data stream and generates a discarded service stream.
In one possible implementation manner, the first routing device determines whether the currently received service data stream and the service data stream to be sent are greater than a third threshold value, and if so, discards all the currently received service data streams.
S702, the first routing device discards part of data in the service data stream and generates a discarded service stream.
As a possible implementation manner, the implementation process of S702 may be: if the first routing device judges that the currently received service data stream and the service data stream to be sent are smaller than the third threshold value but larger than the fourth threshold value, the first routing device randomly discards part of data in the currently received service data stream.
It will be appreciated that congestion avoidance (discard policy) is employed in the embodiments of the present application herein, consistent with the discard policy employed in S501 and S502, by tail-drop or WRED methods.
It should be noted that, in a conventional case, the first routing device in the embodiment of the present application does not discard the service data flow first, and in a case of congestion, the first routing device first selects to detect the data flow to discard, so as to reduce the probability that the service data flow is affected.
Based on the above technical scheme, the application discards the service data stream under some extreme conditions, but the first routing device preferentially discards the probe data stream when the service stream reaches a preset threshold.
In a possible implementation manner, in conjunction with fig. 4, as shown in fig. 8, before the first routing device sends the dropped traffic flow to the second routing device in S403, the first routing device needs to update the first queue and the second queue. Specifically, this can be achieved by the following S801.
S801, based on the discarded service flow, the first routing device updates a first queue and a second queue.
The first queue is used for bearing the business data flow, and the second queue is used for bearing the detection data flow.
In one possible implementation, there may be a traffic data stream to be sent in the first queue, and the first routing device updates the traffic data stream in the discarded traffic stream into the first queue. There may be a probe data flow to be sent in the second queue, and the first routing device updates the probe data flow in the discarded traffic flow into the second queue.
Based on the technical scheme, the embodiment of the application realizes that the service data stream is forwarded preferentially in the detection mode through the first queue and the second queue, so that the network performance degradation caused by data congestion only affects the detection data stream, and the performance of the actual service stream is prevented from being affected.
In a possible implementation manner, as shown in fig. 9 in connection with fig. 4, the first routing device sends the dropped traffic to the second routing device in S403. Specifically, the method can be realized by the following steps S901 to S902.
S901, a first routing device sends service data streams in the discarded service streams to a first queue in a second routing device.
As a possible implementation manner, the implementation process of S901 may be: because the service data stream and the detection data stream both have the attribute or the identifier of the data packet, the first routing device can send the service data stream in the discarded service stream to the first queue in the second routing device preferentially.
S902, the first routing device sends the detection data flow in the discarded service flow to a second queue in the second routing device.
It can be understood that a first queue and a second queue exist in each device, and the foregoing S901 and S902 update the discarded service flows into the first queue and the second queue, so that the service data flows are further ensured not to be affected by the probe data flows due to the two different data flows being divided into two different groups of queues.
Based on the technical scheme, the second routing equipment only sends the service data stream in the discarded service stream to the user equipment, so that the user equipment is prevented from receiving useless data.
In a possible implementation manner, as shown in fig. 10 in connection with fig. 4, before the second routing device sends the service data flow in the dropped service flow to the user equipment in S404, the second routing device needs to identify data, which may be specifically implemented in S1001 below.
S1001, the second routing device identifies a service data flow and a probe data flow in the dropped service flows.
As a possible implementation manner, the implementation process of S1001 may be: the second routing device may identify the traffic data stream and the probe data stream in different ways and send only the traffic data stream (actual traffic data stream) to the user device.
The following describes the identification mode in detail. The second routing equipment identifies the service data flow and the detection data flow in the discarded service flow based on the type and/or the destination address of the data in the discarded service flow; the destination address of the probe data stream is a black hole address.
In one possible implementation manner, the service source device performs packet marking on the probe data stream that is re-engraved in the process of re-engraving the service data stream, so that the second routing device can identify the service data stream and the probe data stream according to the mark (type of data) of the probe data stream, and then send only the service data stream to the user device, where the probe data stream remains in the second routing device.
In another possible implementation manner, in the process of re-etching the service data stream, the service source device not only performs message marking on the re-etched detection data stream, but also modifies the destination address of the detection data stream into a black hole address, and because the second routing device is configured with the routing manner of the black hole address, the detection data stream with the destination address being the black hole address can be sucked into the black hole, unloaded from the network, and the service data stream is guaranteed to be sent only to the user device in the following process.
Illustratively, a black hole route is configured on the second routing device in advance, and a static route is adopted to realize, and an output interface of the black hole IP may be Nu110.
Based on the above technical solution, when the second routing device identifies the service data flow and the probe data flow in the dropped service flow, the identification may be performed by means of a destination address or a data type, if the identification is performed by means of the destination address, the probe data flow in the dropped service flow may be offloaded from the network, and if the identification is performed by means of the data type, the probe data flow in the dropped service flow still exists in the second routing device, but the second routing device only forwards the dropped service data flow to the first routing device.
It is noted that, in the present application, when the service source device, the first routing device, and the second routing device do not meet the preset conditions, the first queue, the second queue, the logical binding queue, and the scheduler already exist in the normal mode in the scene that the normal mode/the probing mode has the periodic switching requirement, so that when the normal mode/the probing mode is switched, the queues, the scheduler, and the association relation do not need to be allocated or deleted separately, and the switching can be faster and smoother. That is, the second queue is disabled when the device switches to normal mode, but the second queue is not deleted (search flag, counter, association, but no dynamic refresh is maintained). When the device switches to the probe mode, the second queue is disabled, the counter, etc., re-enters the dynamically refreshable state.
In summary, in the data transmission method provided by the embodiment of the present application, under the condition of conventional networking, the service source device 101, the first routing device 102, the second routing device 103, and the user device 104 are configured, so that the network can operate normally. (in the prior art, the description is omitted). The AI self-learning mode is configured on the first routing device 102 and the second routing device 103, the AI self-learning queue depth is configured, and the learning target SLA is { packet loss rate <0.01%, congestion time delay <100us, jitter <50us }. (self-learning algorithms are not a matter of protection of the application, and similar techniques/studies are currently available to achieve this function). Further, the detection modes of the service source device 101, the first routing device 102 and the second routing device 103 are configured, when the service source device 101, the first routing device 102 and the second routing device 103 meet the conditions, each device is converted into the detection mode, the service source device 101 carries out re-etching on the service data stream to form a detection data stream, the detection data stream is transmitted to the first routing device 102, the first routing device 102 discards the service stream, the second routing device 103 sends the discarded service stream, finally the second routing device 103 sends the service data stream in the discarded service stream to the user device 104, in the process of data transmission, the first routing device 102 and the second routing device 103 record the performance data (the { time delay, the packet loss and the jitter } performance under the conditions of different { queue depth configuration and utilization ratio } according to the AI self-learning function, form a training set to carry out AI training, obtain the optimal queue depth configuration parameters, and modify the parameters. (self-learning algorithms do not protect the application).
The embodiment of the application can divide the functional modules or functional units of the data transmission device according to the method example, for example, each functional module or functional unit can be divided corresponding to each function, or two or more functions can be integrated in one processing module. The integrated modules may be implemented in hardware, or in software functional modules or functional units. The division of the modules or units in the embodiment of the present application is schematic, which is merely a logic function division, and other division manners may be implemented in practice.
The present application is not limited to the above embodiments, and any changes or substitutions within the technical scope of the present application should be covered by the scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims (10)

1. A routing device supporting a probing mode, wherein the routing device further supports the probing mode on the basis of supporting a normal mode; the normal mode is a mode of the routing equipment in a normal working state; the detection mode is used for generating a detection data stream and differentially processing a service data stream and the detection data stream; the routing device is applicable to a data transmission method, the method comprising:
Determining a mode of the routing equipment;
determining a service data flow and a detection data flow in the received data under the condition that the routing equipment is in the detection mode; the priority of the service data stream is higher than the priority of the detection data stream;
forwarding the service data stream to the routing device in the normal mode or the routing device not supporting the probing mode, and discarding the probing data stream.
2. The apparatus of claim 1, wherein the device comprises a plurality of sensors,
in the case of mode configuration with device granularity, the routing device being in the normal mode includes: all interfaces of the routing device are in the normal mode; the routing device being in the probing mode comprises: all interfaces of the routing device are in the probing mode;
in the case of mode configuration with granularity of interfaces, each interface of the routing device is independently in the normal mode or the probing mode; and if the input interface is in the detection mode and the output interface is in the normal mode, the routing equipment discards the detection data stream in the data stream.
3. The device according to claim 2, wherein determining whether the operation mode of the routing device/interface is the probing mode is performed by a configuration method or a protocol negotiation method;
The manner of configuration/the manner of protocol negotiation may specify a trigger condition; the conditional configuration includes a time period and/or a resource utilization.
4. A device according to any one of claims 1-3, characterized in that the probe data stream generated by the probe mode is determined based on the traffic data stream and a specified magnification; the packet characteristics of the probe data stream are consistent with the packet characteristics of the traffic data stream.
5. The apparatus of claim 4, wherein the probing data stream generated by the probing mode is determined based on a traffic data stream and a specified magnification, comprising:
determining a service data flow which flows through the routing equipment inlet interface or accords with the screening rule;
determining a detection data stream based on the service data stream subjected to re-engraving of the re-engraving multiplying power; wherein the re-engraving function is active when the routing device/interface is in the probing mode.
6. The apparatus of claim 5, further comprising, after the determining the probe data stream based on the multi-scale factor multi-scaling the traffic data stream:
marking the detected data stream; the marking mode comprises the following steps: setting a specific field in the probing data stream to a specific value or adding an optional extension field in the probing data stream.
7. The apparatus of claim 6, further comprising, after said determining to probe a data stream:
modifying a destination address for the probe data stream or not modifying a destination address for the probe data stream; the modified destination address of the detection data stream is a service address or a black hole address; the black hole address is used to offload the probe data stream out of the network.
8. The device of claim 1, wherein a binding queue is provided in the routing device; the binding queue comprises a first queue for bearing business data flows and a second queue for bearing marked detection data flows; the priority of the first queue is higher than the priority of the second queue;
the routing equipment is configured according to the parameters of the queue, and takes effect according to the binding queue; the parameter configuration of the binding queue comprises a discard rule during enqueue and a speed limit rule during dequeue.
9. The apparatus according to claim 7 or 8, wherein,
when the first queue receives a service data stream, the current occupation amount of the first queue is used as a reference amount to judge a discarding rule during enqueuing;
When the second queue receives the detection data stream, the sum of the current occupation amount of the first queue and the current occupation amount of the second queue is used as a reference amount, and discard rule judgment during enqueuing is carried out;
the discard rule adopts the parameter configuration of the binding queue.
10. The apparatus according to claim 7 or 9, wherein each of the probing data streams is configured with an enqueue delay; the enqueue delay is used for deferring the time of the probe data flow entering the second queue.
CN202310833791.5A 2023-07-07 2023-07-07 Routing equipment supporting detection mode Pending CN116886631A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310833791.5A CN116886631A (en) 2023-07-07 2023-07-07 Routing equipment supporting detection mode

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310833791.5A CN116886631A (en) 2023-07-07 2023-07-07 Routing equipment supporting detection mode

Publications (1)

Publication Number Publication Date
CN116886631A true CN116886631A (en) 2023-10-13

Family

ID=88267323

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310833791.5A Pending CN116886631A (en) 2023-07-07 2023-07-07 Routing equipment supporting detection mode

Country Status (1)

Country Link
CN (1) CN116886631A (en)

Similar Documents

Publication Publication Date Title
US6094435A (en) System and method for a quality of service in a multi-layer network element
US9014006B2 (en) Adaptive routing using inter-switch notifications
CN111788803B (en) Flow management in a network
US7274666B2 (en) Method and system for managing traffic within a data communication network
US7953885B1 (en) Method and apparatus to apply aggregate access control list/quality of service features using a redirect cause
US6304552B1 (en) Memory and apparatus for input based control of discards in a lossy packet network
US7916718B2 (en) Flow and congestion control in switch architectures for multi-hop, memory efficient fabrics
US8144588B1 (en) Scalable resource management in distributed environment
US7324442B1 (en) Active queue management toward fair bandwidth allocation
Aamir et al. A buffer management scheme for packet queues in MANET
US11595315B2 (en) Quality of service in virtual service networks
JP2006506845A (en) How to select a logical link for a packet in a router
EP3588880B1 (en) Method, device, and computer program for predicting packet lifetime in a computing device
Wójcik et al. Flow-aware multi-topology adaptive routing
US7072352B2 (en) Inverse multiplexing of unmanaged traffic flows over a multi-star network
CN116886631A (en) Routing equipment supporting detection mode
Cisco QC: Quality of Service Overview
JP5753918B2 (en) Network relay device
Domżał et al. Efficient congestion control mechanism for flow‐aware networks
Jain et al. An adaptive prediction based approach for congestion estimation in active queue management (APACE)
Domżał et al. Admission control in flow-aware multi-topology adaptive routing
Nakayama et al. N Rate ${\rm N}+ 1$ Color Marking: Per-Flow Fairness in Ring Aggregation Networks
Xu Adaptive flow admission control in a software-defined network
Coelho Implementation of Advanced Mechanisms for Cross-Protect Router
Ramaswamy et al. Light-weight control of non-responsive traffic with low buffer requirements

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination