End-to-end available bandwidth monitoring method and system
Through the end-to-end available bandwidth monitoring method, combined with the packetizer and the response agent, dynamically adjust the detection traffic rate, and combined with historical data to optimize measurement, the problem of difficult to take into account the real-time and accuracy of bandwidth monitoring in the existing technology is solved, and high accuracy is achieved Low invasive bandwidth monitoring improves detection efficiency.
Patent Information
- Application Number
- CN202210451742.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-26
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-04-26
AI Technical Summary
Existing bandwidth monitoring tools are difficult to balance between real-time and accuracy, and intrusion is related to detection traffic overhead, resulting in long measurement time and low accuracy.
Using the end-to-end available bandwidth monitoring method, by deploying the packetizer and the destination end, deploying the response agent on the probe end, using the controller to dynamically adjust the detection traffic rate, calculate the real-time available bandwidth based on historical measurement results, and through a multi-stage detection program (initialization, Fast attenuation, slow growth, dual detection, and result storage) optimized measurement process.
While ensuring accuracy, the detection overhead is greatly reduced, the balance between high accuracy and low intrusion is achieved, the real-time detection is improved, and the dynamic changes of available bandwidth can be better monitored.
Smart Images

Figure CN114884846B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of flow detection, and in particular to a method and system for monitoring end-to-end available bandwidth. Background Art
[0002] In computer networks, available bandwidth refers to the maximum unused bandwidth on a link or network path. It can reflect the current transmission capacity and load conditions on the network path (when the total capacity of the path is known). For high-traffic applications (such as video, VR, AR, etc.), the present invention always hopes to use the bandwidth resources on the path as much as possible without causing congestion, which requires the present invention to perceive the available bandwidth on the path in real time and accurately. In addition, available bandwidth measurement also plays an important role in scenarios such as dynamic server and proxy selection, adaptive path planning of overlay networks, etc. Summary of the invention
[0003] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.
[0004] Generally speaking, bandwidth monitoring needs to meet the characteristics of high real-time performance, high accuracy, and low intrusion. In terms of real-time performance, some existing tools have a single measurement duration that is too long, and it takes tens of seconds or even minutes of measurement time to obtain the result (such as Pathload). The intrusion is related to the detection traffic overhead. It is generally believed that the larger the traffic overhead, the stronger the intrusion. But in fact, the intrusion is not only related to the absolute value of the overhead, but also to the relative value of the overhead to the total path capacity. For example, the 1MB overhead is obviously more affected by the 10Mbps link and the 1Gbps link. In addition, low intrusion and high accuracy are often not taken into account at the same time. If you want to reduce intrusion, you have to reduce the number of detection packets. The fewer the detection packets, the greater the impact of noise in the measurement process, and the less likely it is to guarantee the measurement accuracy. To this end, the purpose of the present invention is to greatly shorten this time, reduce the measurement duration to the second level, and reduce the detection overhead while ensuring accuracy, so as to strike a balance between high accuracy and low intrusion, and propose an end-to-end available bandwidth monitoring method.
[0005] Another object of the present invention is to provide an end-to-end available bandwidth monitoring system.
[0006] To achieve the above object, the present invention provides an end-to-end available bandwidth monitoring method, comprising:
[0007] A packet transmitter for sending detection traffic at a preset rate is deployed at the detection end, and a response agent for receiving packets, calculating and feeding back the receiving rate is deployed at the destination end; a controller is used to dynamically change the detection rate of the detection traffic according to a preset measurement program, instruct the packet transmitter to send packets for detection, and calculate the real-time available bandwidth measurement result according to the receiving rate fed back by the response agent; the controller is used to store the real-time available bandwidth measurement result in a local database; the local database is used to store the historical measurement result of the available bandwidth, and the real-time available bandwidth measurement result is compared with the historical measurement result of the available bandwidth to visually monitor the dynamic change trend of the available bandwidth.
[0008] In addition, the end-to-end available bandwidth monitoring method according to the above embodiment of the present invention may also have the following additional technical features:
[0009] Furthermore, in one embodiment of the present invention, the available bandwidth is an average value within a preset time τ:
[0010]
[0011] Where AB is the available bandwidth, u(t) = 1 or 0 indicates whether the link is being used at time t, and equation (a) is used to calculate the proportion of the usage time during the time interval τ.
[0012] Further, in one embodiment of the present invention, two situations occur according to whether the detection flow is greater than the available bandwidth, including:
[0013] First case: R snd ≤C tight -R cross , no congestion is caused, then R snd ≤R rcv ;
[0014] Second case: R snd >C tight -R cross , congestion occurs, then R snd >R rcv ;
[0015] Among them, R snd To detect the traffic sending rate, R rcv To detect the traffic receiving rate, R cross is the cross traffic rate on the link.
[0016] Further, in one embodiment of the present invention, based on the second situation that the traffic follows the FIFO principle, the first packet receiving rate is:
[0017]
[0018] AB cur =C tight -R cross , and get AB cur The expression is:
[0019]
[0020] According to the second detection rate R snd2 Performing a second probe yields the equation:
[0021]
[0022] Combining equations (2), (3) and (4) to calculate the available bandwidth on the current path, we obtain the measurement result:
[0023]
[0024] Among them, AB cur is the current measured value of the available bandwidth of the path, R snd1 is the first packet sending rate, R rcv2 is the second packet receiving rate.
[0025] Furthermore, in one embodiment of the present invention, the measurement procedure includes: initialization, fast decay, slow growth, double detection and result storage stages.
[0026] Furthermore, in one embodiment of the present invention, the initialization phase includes: obtaining a startup rate for measuring available bandwidth, and the selection of the startup rate includes: if the user specifies the startup rate, using the user-specified value; otherwise, if the local database contains historical measurement results, calculating the startup measurement rate based on the historical measurement results; if the user does not specify and the local database has no historical measurement results, obtaining the initial detection rate through ADR measurement.
[0027] Furthermore, in one embodiment of the present invention, the fast decay stage includes: reducing the detection traffic sending rate based on the detection traffic receiving rate of the previous round of detection.
[0028] Furthermore, in one embodiment of the present invention, the slow growth stage includes: increasing the detection rate by multiplying it by a factor greater than 1 each time, so as to obtain a detection rate greater than the available bandwidth.
[0029] Furthermore, in one embodiment of the present invention, the dual detection stage includes: performing a first detection using the detection rate greater than the available bandwidth to obtain a first receiving rate; performing a second detection using the first receiving rate to obtain a second receiving rate; and combining the first receiving rate and the second receiving rate to form a set of equations, and solving them to obtain the current available bandwidth.
[0030] The end-to-end available bandwidth monitoring method of the embodiment of the present invention can reduce the detection overhead while ensuring the accuracy, and achieve a balance between high accuracy and low intrusion.
[0031] To achieve the above object, the present invention proposes an end-to-end available bandwidth monitoring system, comprising: a detection end and a destination end, wherein the detection end comprises: a controller, a packet sender and a local database, and the destination end comprises: a response agent, wherein:
[0032] The controller is used to parse the parameters of the monitoring task, trigger the measurement task according to the preset time period of the parameters of the monitoring task to call the packet transmitter, and determine the current detection rate;
[0033] The packet sender is used to initiate the measurement task, establish a TCP connection with the response proxy on the destination end to synchronize the measurement parameters of the detection rate, and send the constructed detection packet to the UDP port opened by the response proxy on the destination end and receive a reply;
[0034] The local database is used to provide the controller with historical measurement results of the available bandwidth when measuring the available bandwidth, so as to serve as a reference for the detection rate;
[0035] The response agent is used to open a preset TCP port. If the measurement parameter meets the preset conditions, the UDP port is opened and a notification signal is sent to the packet sender of the detection end to prepare for the sending and receiving of the detection packet.
[0036] The end-to-end available bandwidth monitoring system of the embodiment of the present invention can reduce detection overhead while ensuring accuracy, thereby achieving a balance between high accuracy and low intrusion.
[0037] The beneficial effects of the present invention are:
[0038] 1) The present invention is a new technology for real-time monitoring of available bandwidth. Compared with existing tools, the real-time performance of detection is improved. The present invention provides a new option for measuring and monitoring available bandwidth.
[0039] 2) The historical bandwidth measurement value is used as a reference to estimate the current bandwidth measurement value, and then the bandwidth detection can be performed at a detection rate slightly larger than the current available bandwidth, which can greatly reduce the intrusiveness of the available bandwidth measurement while ensuring accuracy;
[0040] 3) Monitoring of available bandwidth can help network administrators better optimize network performance, diagnose network anomalies, and detect network attack incidents.
[0041] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0043] Figure 1 is a flow chart of an end-to-end available bandwidth monitoring method according to an embodiment of the present invention;
[0044] Figure 2 A schematic diagram of the working principle of the end-to-end available bandwidth monitoring method according to an embodiment of the present invention;
[0045] Figure 3 A diagram illustrating an available bandwidth measurement procedure according to an embodiment of the present invention;
[0046] Figure 4 FIG. 4 is a schematic diagram of the structure of an end-to-end available bandwidth monitoring system according to an embodiment of the present invention. DETAILED DESCRIPTION
[0047] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0048] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0049] The following describes an end-to-end available bandwidth monitoring method and system according to an embodiment of the present invention with reference to the accompanying drawings.
[0050] Figure 1 The figure is a flow chart of an end-to-end available bandwidth monitoring method according to an embodiment of the present invention.
[0051] like Figure 1 As shown, the method includes but is not limited to the following steps:
[0052] S1, a packet sender is deployed at the detection end to send detection traffic at a preset rate, and a response agent is deployed at the destination end to receive packets, calculate and feedback the receiving rate;
[0053] S2, using the controller to dynamically change the detection rate of the detection traffic according to the preset measurement program, instructing the packet sender to send packets for detection, and calculating the real-time available bandwidth measurement result according to the receiving rate fed back by the response agent;
[0054] S3, using the controller to store the real-time available bandwidth measurement results into a local database;
[0055] S4, using the local database to store the historical measurement results of the available bandwidth, compares the real-time available bandwidth measurement results with the historical measurement results of the available bandwidth to visually monitor the dynamic change trend of the available bandwidth.
[0056] The embodiments of the present invention are further described below in conjunction with the accompanying drawings.
[0057] Table 1 shows the terms and definitions related to available bandwidth.
[0058] Table 1
[0059] the term definition link A physical line between two directly connected network nodes; path It consists of a series of links connected in sequence; capacity The maximum rate at which a link or path can transmit packets; Available bandwidth Unused capacity in a link or path; Narrow Link The link with the smallest capacity on a path; Tight Link The link with the smallest available bandwidth on a path.
[0060] It is understandable that for a link or path, the available bandwidth refers to the unused portion of its capacity. The present invention focuses on how much data a path can transmit within a period of time, and the concept of instantaneous bandwidth is not very meaningful. Therefore, the available bandwidth AB in the present invention refers to the average value within a period of time τ:
[0061]
[0062] Where u(t) = 1 or 0 indicates whether the link is being used at time t. Equation (1.a) calculates the proportion of the time used during the time interval τ.
[0063] Specifically, when measuring available bandwidth, it is generally assumed that:
[0064] 1) The router queue in the path follows the first-in-first-out principle;
[0065] 2) The cross traffic is stable during the measurement process. Most of the currently available bandwidth measurement tools are based on the above assumptions. The new bandwidth measurement method proposed in the present invention is also based on this assumption.
[0066] This paper summarizes the existing tools based on the detection flow pattern, measurement model, principle and other aspects:
[0067] Packet pairs and packet trains: Measurement tools can be divided into two categories based on whether they send two packets (packet pairs) or a series of more than two packets per probe. Generally speaking, tools that send packet pairs are more affected by noise, but the advantage is that they are less invasive to the network because they send fewer packets. Tools that send packet trains tend to be more tolerant to noise, but introduce more load. How to balance robustness and invasiveness is an important issue.
[0068] Probe Gap Model and Probe Rate Model: The Probe Gap Model infers the available bandwidth from the interval change information between two consecutive packets. The Packet Pair / Train Dispersion (PPTD) will be changed by the cross traffic in the path. The available bandwidth information can be indirectly obtained by sensing the size of the cross traffic. The Probe Rate Model uses a packet sequence of a specified rate for detection and estimates the bandwidth by observing whether it causes path congestion (self-induced congestion). Table 2 shows the existing available bandwidth measurement tools and related information, as shown in Table 2.
[0069] Table 2
[0070]
[0071] As an example, Figure 2 The schematic diagram of the working principle of the end-to-end available bandwidth monitoring method is as follows: Figure 2 As shown:
[0072] Figure 2 The module composition of the present invention is shown, which mainly includes four parts: controller, packet sender, local database and response agent.
[0073] Controller: Located at the detection end, it is responsible for parsing the parameters of the monitoring task, triggering the measurement task (calling the packet sender) according to the task parameters, and determining the current detection rate based on the historical measurement results in the local database.
[0074] Packet sender: Located at the detection end, the actual initiator of the measurement task. It establishes a TCP connection with the response agent on the target end and synchronizes the measurement parameters. After receiving confirmation, it sends the constructed detection packet to the UDP port opened by the response agent on the target end and receives the reply.
[0075] Local database: Located at the detection end, it stores the measurement results. When measuring the available bandwidth, it provides the controller with the historical measurement results of the available bandwidth as a reference value for the current detection rate.
[0076] Response agent: Located at the destination, it opens a fixed TCP port to provide measurement services. If the measurement parameters meet the requirements and resources allow, it opens the UDP port and informs the packet sender at the detection end to prepare for the sending and receiving of detection packets.
[0077] Further, Table 3 shows the variables used in the present invention and their meanings, as shown in Table 3:
[0078]
[0079]
[0080] In principle, the present invention determines whether the detection traffic is greater than or less than the available bandwidth by observing whether the detection traffic rate causes congestion.
[0081] Specifically, two situations may occur depending on whether the detection traffic is greater than the available bandwidth:
[0082] Case 1: R snd ≤C tight -R cross :
[0083] No congestion is caused at this time, so there should be: R snd ≤R rcv .
[0084] Case 2: R snd >C tight -R cross :
[0085] At this time, congestion occurs, resulting in a decrease in the transmission rate of the received traffic, and R snd >R rcv .
[0086] The present invention focuses on analyzing situation 2. Under the assumption that traffic follows the FIFO principle, the detection traffic and cross traffic compete fairly for bandwidth. So the packet receiving rate at this time is (the packet sending rate R snd1 ):
[0087]
[0088] Because AB cur =C tight -R cross We can get AB cur The expression is:
[0089]
[0090] There are two unknown variables AB in the above formula cur and C tight To finally solve AB cur , we need to use a different detection rate R that is larger than the available bandwidth snd2 Performing a second probe yields a similar equation:
[0091]
[0092] Combining the above two equations, we can calculate the available bandwidth on the current path, that is, obtain the measurement result:
[0093]
[0094] Although the detection rate in the above formula only needs to be greater than the available bandwidth, an excessively large detection rate may lead to an increase in the error of the measurement result. Using a detection rate that is only slightly greater than the available bandwidth can obtain a more accurate measurement result. In addition, the smaller the detection rate, the lower the detection overhead and the less invasiveness. In order to obtain a detection rate that is slightly greater than the available bandwidth, the present invention designs a measurement procedure comprising five stages.
[0095] As an example, Figure 3 1 is a diagram showing the available bandwidth measurement procedure according to an embodiment of the present invention. Figure 3 As shown:
[0096] Initialization: The purpose of this stage is to obtain the startup rate of bandwidth measurement. The selection of the startup rate is divided into three cases: if the user specifies the startup rate, the user-specified value is used; otherwise, if the local database contains historical measurement results, the startup measurement rate is calculated based on the historical measurement results; if the user does not specify and the local database has no historical measurement results, the present invention obtains the initial detection rate of the present invention through ADR measurement (sending back-to-back packet trains).
[0097] Fast decay: This stage aims to prevent the initial rate from being too high, which may cause serious congestion and packet loss on the path. snd In the previous round of detection R rcv This ensures that the detection rate quickly drops below the available bandwidth (congestion avoidance).
[0098] Slow growth: The detection rate is slowly increased by multiplying it by a factor slightly greater than 1 each time, thereby obtaining a rate slightly greater than the available bandwidth. This stage is similar to the slow start process in TCP. In addition, unlike the fast decay stage, the fast decay is based on R rcv As the base number, the slow growth increases by the R of the previous round each time. snd Is the base number.
[0099] Double detection: The first detection is performed using the detection rate slightly larger than the available bandwidth obtained in the first three stages, and the second detection is performed using the receiving rate of the first detection as the detection rate. The two detections constitute a set of equations, which can be solved to obtain the current available bandwidth (Formula 5).
[0100] Save results: Save the available bandwidth to the local database. In subsequent measurements, these historical results can be used as a reference for selecting the initial value of the detection rate.
[0101] The above five stages seem to require multiple iterations to complete a measurement, especially the second and third stages require self-looping. In fact, as long as the initial value of the detection rate is properly selected, the present invention can directly skip the loop of the second and third stages (green path). Actual measurement experiments show that most measurements only require 3 to 4 detections. In addition, since the detection rates in the second and third stages are monotonically decreasing and monotonically increasing, respectively, the entire detection process will not fall into an infinite loop.
[0102] The end-to-end available bandwidth monitoring method according to the embodiment of the present invention provides a new option for measuring and monitoring available bandwidth. The current bandwidth measurement value is estimated by taking the historical bandwidth measurement value as a reference, and then the bandwidth detection can be performed at a detection rate slightly larger than the current available bandwidth, which can greatly reduce the intrusiveness of the available bandwidth measurement while ensuring accuracy. The monitoring of available bandwidth can help network management personnel better optimize network performance, diagnose network anomalies, and detect network attack events.
[0103] In order to implement the above embodiment, Figure 4 As shown, this embodiment also provides an end-to-end available bandwidth monitoring system 10, which includes: a detection end 100 and a destination end 200, wherein the detection end 100 includes: a controller 101, a packet sender 102 and a local database 103, and the destination end 200 includes: a response agent 201, wherein,
[0104] The controller 101 is used to analyze the parameters of the monitoring task, trigger the measurement task according to the preset period of the parameters of the monitoring task to call the packet sender 102, and determine the current detection rate;
[0105] The packet sender 102 is used to initiate the measurement task, establish a TCP connection with the response agent 201 on the destination end 200 to synchronize the detection rate measurement parameters, and send the constructed detection packet to the UDP port opened by the response agent 201 on the destination end 200 and receive a reply;
[0106] The local database 103 is used to provide the controller 101 with historical measurement results of the available bandwidth when measuring the available bandwidth, so as to serve as a reference for the detection rate;
[0107] The response agent 201 is used to open a preset TCP port. If the measurement parameters meet the preset conditions, the UDP port is opened and a notification signal is sent to the packet sender 102 of the detection end 100 to prepare for the sending and receiving of the detection packet.
[0108] The end-to-end available bandwidth monitoring system according to the embodiment of the present invention provides a new option for measuring and monitoring available bandwidth. The current bandwidth measurement value is estimated by taking the historical bandwidth measurement value as a reference, and then the bandwidth detection can be performed at a detection rate slightly larger than the current available bandwidth, which can greatly reduce the intrusiveness of the available bandwidth measurement while ensuring accuracy. The monitoring of available bandwidth can help network management personnel better optimize network performance, diagnose network anomalies, and detect network attack events.
[0109] It should be noted that the above explanations of the embodiment of the end-to-end available bandwidth monitoring method are also applicable to the end-to-end available bandwidth monitoring system of this embodiment, and will not be described in detail here.
[0110] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0111] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0112] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.
Claims
1. A method for monitoring end-to-end available bandwidth, characterized in that: The following steps are involved: A packet sender is deployed at the detection end to send detection traffic at a preset rate, and a response agent is deployed at the destination end to receive packets, calculate and feedback the receiving rate; wherein the packet sender establishes a TCP connection with the response agent on the destination end and synchronously measures parameters, and after confirmation, sends the constructed detection packet to the UDP port opened by the response agent on the destination end and receives a reply; Using a controller to dynamically change the detection rate of the detection traffic according to a preset measurement program, instructing the packet sender to send packets for detection, and calculating a real-time available bandwidth measurement result according to the receiving rate fed back by the response agent; Using the controller to store the real-time available bandwidth measurement result in a local database; Utilizing the local database to store the historical measurement results of the available bandwidth, and comparing the real-time available bandwidth measurement results with the historical measurement results of the available bandwidth, so as to visually monitor the dynamic change trend of the available bandwidth; Depending on whether the detection flow is greater than the available bandwidth, two situations may occur, including: First case: R snd ≤C tight -R cross , no congestion is caused, then R snd ≤R rcv ; Second case: R snd >C tight -R cross , congestion occurs, then R snd >R rcv ; Among them, R snd is the flow detection rate, R rcv To detect the traffic receiving rate, R cross is the cross traffic rate on the link; Based on the second case, under the assumption that the traffic follows the FIFO principle, the first receiving rate is: AB cur =C tight -R cross , and get AB cur The expression is: According to the second detection rate R snd2 Performing a second probe yields the equation: Combining equations (2), (3) and (4) to calculate the available bandwidth on the current path, we obtain the measurement result: Among them, AB cur is the current measured value of the available bandwidth of the path, R snd1 is the first detection rate, R rcv2 is the second receiving rate; Among them, C tight is the total capacity of the tight link.
2. The method according to claim 1, characterized in that The available bandwidth is the average value within a preset time τ: Where AB is the available bandwidth, u(t) = 1 or 0 indicates whether the link is being used at time t, and equation (a) calculates the proportion of the usage time during the time interval τ. is the measured average value of the available bandwidth AB over a period of time, and C represents the total bandwidth capacity of the link.
3. The method according to claim 1, characterized in that The measurement procedure includes: Initialization, fast decay, slow growth, double detection and storage results phase; among them, The fast decay stage includes: reducing the detection rate of the detection flow based on the detection flow receiving rate of the previous round of detection; The slow growth phase includes: increasing the detection rate by multiplying the detection rate by a factor greater than 1 each time to obtain a detection rate greater than the available bandwidth; The dual detection stage includes: Performing a first detection using the detection rate greater than the available bandwidth to obtain a first receiving rate; Performing a second detection using the first receiving rate to obtain a second receiving rate; The first receiving rate and the second receiving rate are combined to form a set of equations, and the current available bandwidth is obtained by solving the equations.
4. The method according to claim 3, characterized in that The initialization phase includes: obtaining an initial detection rate for measuring available bandwidth, and the selection of the initial detection rate includes: If the user specifies the initial detection rate, the user-specified value is used; otherwise, If the local database contains historical measurement results, calculating the initial detection rate according to the historical measurement results; If the user does not specify and the local database has no historical measurement results, the initial detection rate is obtained through ADR measurement, where the ADR is the asymptotic dispersion rate. The bandwidth is calculated by the arrival time and message length of the first and last messages in a group of end-to-end packet queues to estimate the available bandwidth.
5. An end-to-end available bandwidth monitoring system using the method of claim 1, characterized in that: The system comprises: a detection end and a destination end, wherein the detection end comprises: a controller, a packet sender and a local database, and the destination end comprises: a response agent, wherein, The controller is used to parse the parameters of the monitoring task, trigger the measurement task according to the preset time period of the parameters of the monitoring task to call the packet transmitter, and determine the current detection rate; The packet sender is used to initiate the measurement task, establish a TCP connection with the response proxy on the destination end to synchronize the measurement parameters of the detection rate, and send the constructed detection packet to the UDP port opened by the response proxy on the destination end and receive a reply; The local database is used to provide the controller with historical measurement results of the available bandwidth when measuring the available bandwidth, so as to serve as a reference for the detection rate; The response agent is used to open a preset TCP port. If the measurement parameter meets the preset conditions, the UDP port is opened and a notification signal is sent to the packet sender of the detection end to prepare for the sending and receiving of the detection packet.
Citation Information
Patent Citations
End-to-end low available bandwidth measuring method
CN101026509A
Method and system for measuring end-to-end available bandwidth
CN101958811A
Method for measuring end-to-end available bandwidth based on network compact link position
CN1658583A
Available bandwidth estimating device
US20120253761A1
Method for controlling network congestion and related apparatus
WO2022001291A1
