Network control device, network control method, and network control program
By reducing WRED parameters to four degenerate values (Qs, Qd, α, β) based on queue length and priority, the complexity of setting WRED parameters in edge routers is simplified, enhancing QoS control efficiency.
Patent Information
- Application Number
- JP2024566970
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-12-26
- Publication Date
- 2025-12-16
- Estimated Expiration
- 2042-12-26
AI Technical Summary
The heavy load on network operators in setting Weighted Random Early Detection (WRED) parameters for edge routers due to the need to set 3 x 3 = 9 parameters for packets with priority set by 2Rate-3Color, complicating QoS control.
Implementing QoS control using WRED based on the length of a queue and priority, with a parameter input unit, calculation unit, and setting unit to simplify parameter setting by reducing parameters to four degenerate values (Qs, Qd, α, β) for edge routers.
Facilitates easier and more efficient setting of WRED parameters, reducing the complexity and load on network operators while maintaining effective QoS control.
Smart Images

Figure 0007786619000003 
Figure 0007786619000004 
Figure 0007786619000005
Abstract
Description
[Technical Field]
[0001] The present invention relates to a network control device, a network control method, and a network control program for setting parameters for WRED (Weighted Random Early Detection) in an edge router. [Background technology]
[0002] Conventionally, in order to achieve fairness in communication (provision of equal communication capacity) for each user in best-effort traffic at the physical IF (interface) of an edge router, there is a technology that sets priority (G / Y / R) for traffic packets of each user (each VLAN (Virtual LAN)) using 2Rate-3Color and performs QoS control using WRED. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] Japanese Patent Publication No. 2020-136925 [Non-patent literature]
[0004] [Non-Patent Document 1] J.Lee et al., “WRED PERFORMANCE ANALYSIS AND DESIGN FOR TACTICAL SATELLITE COMMUNICATION,” MILCOM 2012 - 2012 IEEE Military Communications Conference, Oct.29 Nov.1, 2012. https: / / ieeexplore.ieee.org / stamp / stamp.jsp?tp=&arnumber=6415577 [Non-patent document 2] LBLim et al., “RED and WRED Performance Analysis Based on Superposition of N MMBP Arrival Process,” 2010 24th IEEE International Conference on Advanced Information Networking and Applications, Apl.20-23, 2010 Summary of the Invention [Problem to be solved by the invention]
[0005] In order for an edge router to control QoS using WRED, a network operator must set three WRED parameters (Min Threshold, Max Threshold, and Mark Probability Denominator (MPD)) in the edge router. Therefore, for example, to apply WRED to packets with priority set by 2Rate-3Color, the network operator must set 3 x 3 = 9 parameters in the edge router. This has led to the problem of a heavy load on the network operator when setting the WRED parameters.
[0006] Therefore, an object of the present invention is to solve the above-mentioned problems and to make it easier to set parameters for WRED. [Means for solving the problem]
[0007] In order to solve the above-mentioned problems, the present invention provides QoS control for discarding packets by WRED (Weighted Random Early Detection) based on the length of a queue that aggregates user traffic packets and the priority set for each user traffic packet, the QoS control comprising: a parameter input unit that accepts input of a first value indicating the ratio of the interval between the queue length thresholds at which random discard of packets, which are set for each priority, and the queue length threshold at which tail discard of packets, which are set for each priority, and a second value indicating the ratio of the maximum discard rate for each priority when randomly discarding packets; a calculation unit that uses the input first and second values to calculate the queue length threshold at which random discard of packets is started, the maximum discard rate for the random discard, and the queue length threshold at which tail discard is started, which are used for WRED; and a setting unit that sets the calculated values in an edge router that performs QoS control by WRED. [Effects of the Invention]
[0008] According to the present invention, it is possible to easily set parameters for WRED. [Brief explanation of the drawings]
[0009] [Figure 1] FIG. 1 is a diagram showing an example of a process for applying WRED to packets colored by 2Rate-3Color to perform QoS control. [Figure 2] FIG. 2 is a diagram for explaining the three parameters of WRED. [Figure 3] FIG. 3 is a diagram for explaining constraints imposed on WRED parameters. [Figure 4] FIG. 4 is a diagram for explaining constraints imposed on WRED parameters. [Figure 5] FIG. 5 is a diagram for explaining constraints imposed on WRED parameters. [Figure 6]FIG. 6 is a diagram for explaining constraints imposed on WRED parameters. [Figure 7] FIG. 7 is a diagram illustrating an example of WRED parameters set in an edge router. [Figure 8] FIG. 8 is a diagram illustrating an example of the configuration of a system including the NW control device of the first embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of the configuration of the NW control device according to each embodiment. [Figure 10] FIG. 10 is a flowchart illustrating an example of a processing procedure executed by the NW control device according to the first embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of the configuration of a system including a NW control device according to the second embodiment. [Figure 12] FIG. 12 is a diagram illustrating an example of changing parameters of WRED. [Figure 13] FIG. 13 is a sequence diagram illustrating an example of a processing procedure executed by the system of the second embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of the configuration of a computer that executes a network control program. DETAILED DESCRIPTION OF THE INVENTION
[0010] Hereinafter, with reference to the drawings, modes for carrying out the present invention (embodiments) will be described in two parts, a first embodiment and a second embodiment, but the present invention is not limited to each embodiment.
[0011] [Prerequisites] First, the prerequisites for the network control device of each embodiment will be explained using Figure 1. In a network, a transfer device that directly accommodates user terminals and servers is called an edge router. An edge router has multiple physical interfaces (IFs). Each physical IF accommodates traffic from multiple users, separated by VLANs. A network monitoring device records information about traffic passing through each physical IF of the edge router and stores it in a database (DB).
[0012] Focusing on the BE (Best Effort) traffic of one physical IF of an edge router, traffic packets are colored, for example, by VLAN using 2Rate-3Color: Red (R), Yellow (Y), and Green (G). Also, one queue for BE traffic is prepared per physical IF, and QoS control is performed using WRED (Weighted Random Early Detection). For this reason, WRED parameters are set in the edge router.
[0013] [2Rate-3Color] Network operators want to provide equal communication capacity to each user for BE traffic. Therefore, in each VLAN, to perform QoS control in the queue for BE traffic at the subsequent stage, traffic packets are colored as follows based on the bandwidth used by the traffic of each VLAN.
[0014] · Green when the bandwidth used is less than the specified bandwidth When the bandwidth usage exceeds the specified bandwidth, the Yellow If the bandwidth usage exceeds the specified bandwidth, the Red
[0015] [WRED] WRED requires the following three parameters to be set: Min Threshold: The queue length threshold at which random packet dropping begins. Max Threshold: The queue length threshold at which packets are tail dropped (ending random packet discard). Mark Probability Denominator (MPD): The denominator of the probability of randomly discarding a packet. Packets are discarded with a probability of 1 / MPD.
[0016] For example, consider a case where for a certain physical interface of an edge router, Min Threshold = 0.3, Max Threshold = 0.4, and MPD = 2 are set as shown in FIG. 2. In this case, when buffers accumulate in the queue and the queue length exceeds 30% of the maximum queue length, the edge router starts random packet drops.
[0017] Also, when the queue length is up to 40% of the maximum queue length, the edge router increases the packet discard rate up to a maximum of 50% in proportion to the incoming traffic (= queue length). And when the queue length exceeds 40% of the maximum queue length, the edge router discards 100% of the traffic (performs tail drop).
[0018] When applying WRED to packets colored with 2rate-3Color by color G / Y / R, 3 × 3 = 9 parameters are required, and the degree of freedom of each parameter is also high. Therefore, the NW operator has a problem that the load for determining the value of each parameter is large. Thus, by imposing constraints on the values of each parameter set for WRED, the degree of freedom of the values of each parameter is reduced.
[0019] Hereinafter, referring to FIGS. 3 to 5, the concept of restricting the values of parameters in each embodiment will be described.
[0020] First, the priority of packets processed by WRED is R < Y < G. Therefore, as shown in FIG. 3, after 100% of the R packets are discarded, the discarding of the Y packets should start. Also, after 100% of the Y packets are discarded, the discarding of the G packets should start.
[0021] Also, as shown in FIG. 4, the discard of packets should start gradually. Also, if it seems that buffers will still accumulate in the queue even when 100% of the R packets are discarded, the Y packets may also start to be discarded immediately. That is, the Max Threshold of the R packets and the Min Threshold of the Y packets may be the same value. This is because if the buffer accumulation situation does not improve even when 100% of the R packets are discarded, there is no need to maintain that state and it is necessary to start the next measure.
[0022] Also, if it seems that buffers will still accumulate in the queue even when 100% of the Y packets are discarded, the G packets may also start to be discarded immediately. That is, the Max Threshold of the Y packets and the Min Threshold of the G packets may be the same value. This is because, similarly to the above, if the buffer accumulation situation does not improve even when 100% of the Y packets are discarded, there is no need to maintain that state and it is necessary to start the next measure.
[0023] In addition, when the priority of the G packets is significantly higher than that of the Y packets or the R packets, a margin may be provided between the Max Threshold of the Y packets and the Min Threshold of the G packets, but it is preferable to make it a fixed length or a value that can be calculated from other values so that the parameters do not increase.
[0024] Also, the fact that the packet priorities are R < Y < G may also be represented by the difference in the slopes of the discard rates of the R, Y, and G packets respectively. For example, as shown in FIG. 5, the slope of the discard rate of the R packets is made the gentlest, the slope of the discard rate of the Y packets is made the next gentlest, and the slope of the discard rate of the G packets is made the steepest.
[0025] By doing so, when buffers start to accumulate in the queue, the discarding of R packets will start earlier. As a result, it becomes difficult for buffers to accumulate in the queue, so the possibility of G packets and Y packets being discarded can be reduced. Also, the discarding of Y packets will start at a much earlier timing than that of G packets. As a result, the possibility of G packets being discarded can be reduced.
[0026] As a result, compared with FIG. 4, in FIG. 5, since there is a difference in the slope of the packet discard rate among each packet, the interval from the Min Threshold of R packets to the Max Threshold of G packets widens. Also, the magnitude of the interval from the Min Threshold to the Max Threshold of each of the R, Y, and G packets is R < Y < G.
[0027] Also, the fact that the packet priorities are R < Y < G may appear in the differences in the MPDs of the R, Y, and G packets respectively. For example, as shown in FIG. 6, for G packets and Y packets, by reducing 1 / MPD, the slope of the packet discard rate with respect to the queue length can be suppressed. That is, although it is inevitable that packets are discarded 100% when the Max Threshold is reached, it is considered that the packet discard rate should be moderated according to the priority until then. For example, the interval from the Min Threshold to the Max Threshold of the G packets shown in FIG. 6 and the interval from the Min Threshold to the Max Threshold of the Y packets should be intervals for observing the situation, expecting that the buffers accumulated in the queue will gradually decrease.
[0028] When restricting the values of the parameters in WRED according to the above concept, the nine parameters of WRED can be degenerated into, for example, four parameters (see FIG. 7).
[0029] For example, the ratio of the interval between the thresholds R, Y, and G for the queue length is R:Y:G=2α:α:1. Also, the ratio of the interval between 1 and 1 / MPD for the maximum drop rate is R:Y:G=0:β:2β. Also, the queue length threshold at which the lowest priority packets (packets of R) start to be dropped is Q s Also, Q d is the queue length threshold at which all packets (R, Y, G) start to be discarded. d Let's say.
[0030] By using the above constraints, all parameters of WRED (R min :R Min Threshold, R max :R Max Threshold, R mpd :R's MPD, Y min :Y Min Threshold, Y max :Y Max Threshold, Y mpd :Y's MPD, G min :G Min Threshold, G max :G Max Threshold, G mpd :G's MPD) is Q s , Q d , α, and β.
[0031] For example, as shown in Figure 7, min =Q s Also, R max is expressed as the following equation (1): s , Q d , α. Also, R max =Y min So, Y min MoQ s , Q d , α.
[0032]
number
[0033] Also, Y maxAs shown in the following equation (2), s , Q d , α. Also, Y max =G min So, G min MoQ s , Q d , α. Also, G max =Q d is.
[0034]
number
[0035] Furthermore, as shown in the following equation (3), R mpd , Y mpd , G mpd can also be expressed in terms of β.
[0036] R mpd =1, Y mpd =1 / (1-β), G mpd =1 / (1-2β)…Equation (3)
[0037] In the following, Q s , Q d , α, and β are referred to as “degenerate parameters” as appropriate. Note that the parameter degenerate method is not limited to the above method, and the number of degenerate parameters is not limited to four.
[0038] [First embodiment] [System configuration example] Next, the NW control device 10 of the first embodiment will be described. First, an example of a system configuration including the NW control device 10 will be described with reference to Fig. 8. The system includes, for example, the NW control device 10 and an edge router 20, as shown in Fig. 8.
[0039] The NW control device 10 sets parameters for WRED in the edge router 20 based on instructions input by the NW operator or the like. For example, the NW control device 10 sets four degeneration parameters (Q s , Q d, α, β), the nine parameters of WRED (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) is calculated. Then, the NW control device 10 sets the calculated nine parameters of WRED in the edge router 20.
[0040] The edge router 20 accommodates traffic from multiple users (e.g., users A, B, C, etc.). Based on the WRED parameters set by the NW control device 10, the edge router 20 performs QoS control using WRED, based on the length of the queue that aggregates the traffic packets of each user and the priority (R, Y, G) set to the traffic packets of each user using 2Rate-3Color.
[0041] In the following description, the priority of packets in each user's traffic is R, Y, and G set by 2Rate-3Color, but other priorities may be set.
[0042] [Example of network control device configuration] Next, a configuration example of the NW control device 10 will be described with reference to Fig. 9. The NW control device 10 includes, for example, an input / output unit 11, a communication unit 12, a storage unit 13, and a control unit 14.
[0043] The input / output unit 11 is an interface that controls the input and output of various data. For example, the input / output unit 11 receives four degeneration parameters (Q s , Q d , α, β). The communication unit 12 is an interface that controls communication with external devices. For example, the communication unit 12 transmits a command to set the nine parameters of WRED output by the control unit 14 via a network to the edge router 20.
[0044] The storage unit 13 stores data, programs, etc. that are referenced when the control unit 14 executes various processes. The storage unit 13 is realized by a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. For example, the storage unit 13 stores four degeneracy parameters (Q s , Q d , α, β) etc.
[0045] The control unit 14 is responsible for overall control of the NW control device 10. The functions of the control unit 14 are realized by, for example, a CPU (Central Processing Unit) executing a program stored in the storage unit 13.
[0046] The control unit 14 includes, for example, a parameter input unit 141, a calculation unit 142, and a setting unit 143. Note that the disposal information acquisition unit 144, the determination unit 145, and the update unit 146 shown by dashed lines may or may not be provided, and the cases in which they are provided will be described later.
[0047] The parameter input unit 141 receives input of degeneration parameters. The degeneration parameters include a first value, a second value, a third value, and a fourth value.
[0048] The first value is a value that indicates the ratio of the interval between the queue length thresholds for each priority level, which are set for each packet priority level and for which random packet discarding starts and tail discarding starts. For example, if the ratio of the intervals between the R, Y, and G thresholds for queue length is R:Y:G=2α:α:1, and α=2, then the ratio of the intervals between the R, Y, and G thresholds for queue length is 4:2:1.
[0049] The second value is a value indicating the ratio of the maximum discard rate for each priority level when randomly discarding packets, which is set for each priority level of packets. For example, if the ratio of the interval between 1 and 1 / MPD for R, Y, and G is R:Y:G=0:β:2β, and β=0.1, then the ratio of the interval between 1 and 1 / MPD for R, Y, and G is 0:0.1:0.2.
[0050] The third value is a queue length threshold (Q) that indicates the threshold at which random discarding of packets with the lowest priority (e.g., R) begins. s The fourth value is the queue length threshold (Q) at which the tail of the highest priority (e.g., G) packets starts to be discarded. d )
[0051] The calculation unit 142 calculates the nine parameters of WRED (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) is calculated (see Figure 7).
[0052] For example, the calculation unit 142 calculates R min =Q s Also, G max =Q d Furthermore, the calculation unit 142 calculates Q s Based on α and R:Y:G=2α:α:1, R max and Y min Furthermore, the calculation unit 142 calculates Q s Based on α and R:Y:G=2α:α:1, Y max and Y min Furthermore, the calculation unit 142 calculates R mpd , Y mpd , G mpd Calculate.
[0053] In addition, in WRED, Q s is the initial value of the queue (e.g., 0.05-0.1), and Q d Therefore, these values are stored in the storage unit 13 in advance, and the calculation unit 142 calculates the values stored in the storage unit 13 as Q s , Q d In this case, the parameter input unit 141 receives a parameter from a network operator or the like, such as Q s , Q d There is no need to accept input from
[0054] The setting unit 143 sets the nine WRED parameters (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) in the edge router 20. For example, the setting unit 143 converts the calculated WRED parameters into commands for the edge router, thereby inputting the WRED parameter settings into the edge router 20.
[0055] [Example of processing procedure] Next, an example of a processing procedure executed by the NW control device 10 will be described with reference to FIG.
[0056] For example, the parameter input unit 141 receives four degeneration parameters (α, β, Q s , Q d ) is input (S1). Then, the parameter input unit 141 determines whether the inputted degeneration parameters are normal or not (S2). For example, the parameter input unit 141 may receive a negative value of the input degeneration parameters or a value of Q s >Q dIf a value such as this is input, the parameter input unit 141 determines that the degeneration parameters are not normal (No in S2), outputs a message to the NW administrator or the like to prompt them to re-input the NW degeneration parameters, and then returns to S1. On the other hand, if the parameter input unit 141 determines that the input degeneration parameters are normal (Yes in S2), the process proceeds to S3.
[0057] The calculation unit 142 calculates the parameters for WRED (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) (S3). Then, the setting unit 143 converts the WRED parameters calculated in S3 into an edge router command and sets and inputs it to the edge router 20 (S4).
[0058] In this way, the NW control device 10 uses, for example, four degeneration parameters (α, β, Q s , Q d ) to calculate the WRED parameters (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) can be calculated and set.
[0059] [Second embodiment] When the packet discard rate for each priority level in the edge router 20 changes, the NW control device 10 adjusts the WRED parameter (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd) may be changed. The NW control device 10 in this case will be described as the NW control device 10a of the second embodiment. The same components as those in the first embodiment are denoted by the same reference numerals and will not be described again.
[0060] [System configuration example] Next, a NW control device 10a according to a second embodiment will be described. An example of the configuration of a system including the NW control device 10a will be described with reference to Fig. 11. The system includes, for example, the NW control device 10a, an edge router 20, a NW monitoring device 30, and a DB 40, as shown in Fig. 11.
[0061] The NW monitoring device 30 monitors, for example, the packet discard rate for each priority (R / Y / G) in the edge router 20. The NW monitoring device 30 then stores time-series data or statistical data of the monitoring results of the packet discard rate for each priority in the DB 40. The DB 40 accumulates information (discard information) indicating the packet discard rate for each priority in the edge router 20.
[0062] The NW control device 10a refers to the discard information accumulated in the DB 40 at predetermined time intervals, and when it determines that a high-priority packet (packet of G) is being discarded (=the amount of data accumulated in the queue is increasing), it adjusts the WRED parameter (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) and set it in the edge router 20.
[0063] The above parameters for WRED (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd An example of changing the parameter ) will be described with reference to FIG.
[0064] As mentioned above, Q s is set to the initial value of the queue (e.g., 0.05-0.1), and Q d is often set to a value at the end of the queue (e.g., 0.9-0.95). s and Q d It is thought that the value of Q does not need to be changed significantly depending on the queue situation. s and Q d The value of remains unchanged from the value initially given.
[0065] For example, consider the situation in the upper diagram of Figure 12 where the buffer has accumulated up to the queue length indicated by the dashed line. In this situation, even though all packets of R and Y have been discarded, there is still buffer space in the queue. Therefore, the only way to observe this is to see that the buffer is released as packets of G are discarded.
[0066] When this state occurs, the NW control device 10a updates the value of α to a value smaller than the original value (for example, α=2) (for example, to α=0.5 as shown in the lower diagram of FIG. 12). This widens the period in which packets of G are discarded. Also, when the buffer has accumulated up to the queue length indicated by the dashed line in the upper diagram of FIG. 12, there is no longer any room to hesitate to discard packets. Therefore, for example, β is set to 0 as shown in the lower diagram of FIG. 12.
[0067] This allows us to set the parameters for WRED (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) will have the values shown in the bottom diagram of Figure 12. As a result, packets from G will be discarded earlier than before, which is expected to reduce the amount of buffer accumulation in the queue. The value of α to be applied at this time can be determined in advance or calculated in some way.
[0068] [Configuration example] An example of the configuration of the NW control device 10a will be described with reference to Fig. 9. The NW control device 10a includes a discarding information acquisition unit 144, a determination unit 145, and an update unit 146.
[0069] The discard information acquisition unit 144 acquires discard information of the edge router 20 from the DB 40. Based on the acquired discard information of the edge router 20, the determination unit 145 determines whether or not a state in which the discard rate of packets with a priority equal to or higher than a predetermined value (for example, G) in the edge router 20 exceeds a predetermined value continues for a predetermined time.
[0070] When it is determined that the discard rate of packets having a priority equal to or higher than a predetermined value (for example, G) has continued for a predetermined period of time in the edge router 20, the update unit 146 updates the queue length threshold (for example, G) at which random discarding of packets having a priority equal to or higher than a predetermined value (for example, G min ) becomes smaller than the original threshold. The update unit 146 also updates the first value (for example, α) so that the maximum value of the packet discard rate (R mpd , Y mpd , G mpd ) is updated to 1.
[0071] Then, the calculation unit 142 calculates the WRED parameter (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd The setting unit 143 recalculates the recalculated WRED parameters (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) is set in the edge router 20.
[0072] The update unit 146 may update only the first value without updating the second value. In this case, the calculation unit 142 calculates the WRED parameter (R min , R max , Y min , Y max , G min , G max ) is recalculated. The setting unit 143 then recalculates the recalculated WRED parameters (R min , R max , Y min , Y max , G min , G max ) is set in the edge router 20.
[0073] [Example of processing procedure] Next, an example of a processing procedure executed by a system including the NW control device 10a will be described with reference to FIG.
[0074] The NW monitoring device 30 monitors the packet discard rate for each priority in the edge router 20 at predetermined intervals (S11: status monitoring (periodic)), and collects information (discard information) indicating the packet discard rate for each priority (S12: information collection (periodic)). The NW monitoring device 30 then stores the information collected in S12 (S13: information storage (periodic)). The NW monitoring device 30 repeats the processes of S11 to S13 to store, for example, discard information indicated by reference numeral 1301 in the DB 40. The discard information indicated by reference numeral 1301 is information indicating the packet discard rate in the edge router 20 for each time and for each packet priority (G, Y, G).
[0075] Furthermore, the disposal information acquisition unit 144 of the NW control device 10a refers to the DB 40 at predetermined intervals (S14: information reference (periodically)) and collects disposal information (S15: information collection (periodically)).
[0076] Then, based on the discard information collected in S15, the determination unit 145 determines whether the state in which the discard rate of packets having a priority equal to or greater than a predetermined value (for example, G) exceeds a predetermined value has continued for a predetermined time (S16). If the determination unit 145 determines that the state in which the discard rate of packets having a priority equal to or greater than a predetermined value exceeds the predetermined value has continued for a predetermined time (Yes in S16), the process proceeds to S17. On the other hand, if the determination unit 145 determines that the state in which the discard rate of packets having a priority equal to or greater than a predetermined value exceeds the predetermined value has not continued for a predetermined time (No in S16), the process returns to S14.
[0077] For example, consider a case where the determination unit 145 determines whether the time during which the packet drop rate of G is 10% or more has continued for 5 minutes or more. In this case, in the drop information indicated by reference numeral 1301, the packet drop rate of G packets at 09:05 is 11% and the packet drop rate of G packets at 09:10 is 25%, so the determination unit 145 determines that the time during which the packet drop rate of G packets exceeds 10% has continued for 5 minutes or more (Yes in S16). Then, the process proceeds to S17.
[0078] In S17, the NW control device 10a recalculates the parameters for WRED. For example, the update unit 146 min Update α so that is smaller than the original threshold, and R mpd , Y mpd , G mpd Then, the calculation unit 142 updates β so that α becomes 1. Then, based on the updated α and β, the calculation unit 142 calculates the parameter for WRED (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , G mpd ) is recalculated. After that, the setting unit 143 recalculates the recalculated WRED parameters (R min , R max , R mpd , Y min , Y max , Y mpd , G min , G max , Gmpd ) is set in the edge router 20 (S18: resetting).
[0079] By executing the above process, the NW control device 10a can change the WRED parameters in the edge router 20 to appropriate values according to the packet loss rate of each priority level in the edge router 20.
[0080] If the amount of data stored in the queue is improved as a result of changing the WRED parameters in the edge router 20, the NW control device 10a may return the values of the WRED parameters to their original values.
[0081] For example, if the NW control device 10a determines that the edge router 20 has not discarded packets with a priority equal to or lower than a predetermined value (e.g., G, Y) for a predetermined period of time, the NW control device 10a may return the value of the parameter for WRED to its original value.
[0082] For example, after the recalculated WRED parameters are set in the edge router 20, the judgment unit 145 judges, based on the discard information, whether the discard rate of packets with a priority equal to or lower than a predetermined value (e.g., G, Y) in the edge router 20 has remained equal to or lower than a predetermined value for a predetermined period of time.
[0083] Then, when the determining unit 145 determines that the discard rate of packets whose priority is equal to or lower than a predetermined value (for example, G, Y) in the edge router continues to be equal to or lower than a predetermined value for a predetermined time, the setting unit 143 restores the values of the parameters for WRED set in the edge router 20 to their original values. For example, the setting unit 143 resets the parameters for WRED in the edge router 20, calculated based on α and β input by the NW operator.
[0084] As a result, the NW control device 10a may change the WRED parameters of the edge router 20, and for example, 100% of packets of R and Y may be discarded, and G min From G maxThis can prevent the situation where packets of G are discarded at a discard rate proportional to the queue length from continuing until
[0085] [System configuration, etc.] Furthermore, the components of each unit shown in the figure are conceptual functional units and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. Furthermore, all or any part of the processing functions performed by each device can be realized by a CPU and a program executed by the CPU, or can be realized as hardware using wired logic.
[0086] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method.In addition, the information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified.
[0087] [program] The network control device 10 can be implemented by installing a program (network control program) as package software or online software on a desired computer. For example, by executing the program on an information processing device, the information processing device can function as the network control device 10. The information processing device referred to here includes mobile communication terminals such as smartphones, mobile phones, and PHS (Personal Handyphone Systems), as well as terminals such as PDAs (Personal Digital Assistants).
[0088] 14 is a diagram showing an example of a computer that executes a network control program. The computer 1000 includes, for example, a memory 1010 and a CPU 1020. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.
[0089] The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012. The ROM 1011 stores a boot program such as a BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to a hard disk drive 1090. The disk drive interface 1040 is connected to a disk drive 1100. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to a mouse 1110 and a keyboard 1120, for example. The video adapter 1060 is connected to a display 1130, for example.
[0090] The hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, the programs that define the processes executed by the network control device 10 are implemented as program modules 1093 in which computer-executable code is written. The program modules 1093 are stored, for example, in the hard disk drive 1090. For example, a program module 1093 for executing processes similar to those of the functional configuration of the network control device 10 is stored in the hard disk drive 1090. The hard disk drive 1090 may be replaced by an SSD (Solid State Drive).
[0091] Data used in the processing of the above-described embodiment is stored as program data 1094 in, for example, the memory 1010 or the hard disk drive 1090. The CPU 1020 then reads the program module 1093 or the program data 1094 stored in the memory 1010 or the hard disk drive 1090 into the RAM 1012 as needed and executes them.
[0092] The program module 1093 and program data 1094 are not limited to being stored in the hard disk drive 1090, but may also be stored in, for example, a removable storage medium and read by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and program data 1094 may be stored in another computer connected via a network (such as a LAN (Local Area Network) or WAN (Wide Area Network)). The program module 1093 and program data 1094 may then be read by the CPU 1020 from the other computer via the network interface 1070. [Explanation of symbols]
[0093] 10 Network control device 11 Input / output section 12 Communications Department 13 Storage section 14 Control Unit 141 Parameter input section 142 Calculation section 143 Settings 144 Disposal Information Acquisition Department 145 Judgment section 146 Update Department
Claims
1. a parameter input unit for receiving input of a first value indicating a ratio of an interval between a queue length threshold for starting random discard of packets set for each priority and a queue length threshold for starting tail discard of packets set for each priority, and a second value indicating a ratio of a maximum discard rate for each priority when randomly discarding packets set for each priority, in QoS control for discarding packets by WRED (Weighted Random Early Detection) based on a length of a queue that aggregates packets of traffic of each user and a priority set for the packets of traffic of each user; a calculation unit that calculates, for each of the priorities, a queue length threshold at which random packet discarding is initiated, a maximum discard rate in the random discarding, and a queue length threshold at which tail discarding is initiated, the queue length threshold being used for the WRED, using the input first value and the second value; a setting unit that sets the calculated value in the edge router that performs QoS control using the WRED; A network control device comprising:
2. The parameter input unit further receiving an input of a third value indicating a queue length threshold at which random discarding of the lowest priority packets is initiated and a fourth value indicating a queue length threshold at which tail discarding of the highest priority packets is initiated; The calculation unit further Using the input third and fourth values, a queue length threshold at which random packet discarding begins, a maximum discard rate in the random discarding, and a queue length threshold at which tail discarding begins are calculated for each priority.
2. The network control device according to claim 1.
3. The calculation unit further The queue length threshold at which the tail discard of the packet is initiated is set to the same value as the queue length threshold at which the random discard of the packet with the next highest priority is initiated.
2. The network control device according to claim 1.
4. a determination unit that determines whether a state in which a packet discard rate of packets having a priority equal to or higher than a predetermined value in the edge router exceeds a predetermined value for a predetermined period of time based on discard information indicating a packet discard rate for each priority in the edge router; an update unit that updates the first value so that a queue length threshold at which random discarding of packets having a priority equal to or greater than a predetermined value starts becomes smaller than an original threshold when it is determined that the discard rate of packets having a priority equal to or greater than a predetermined value has continued for a predetermined period of time in the edge router, The calculation unit recalculating, for each of the priorities, a queue length threshold at which random packet discarding begins and a queue length threshold at which tail discarding begins, using the updated first value; The setting unit The recalculated value is set in the edge router.
2. The network control device according to claim 1.
5. The update unit when it is determined that a state in which a discard rate of packets having a priority equal to or greater than a predetermined value continues for a predetermined time period in the edge router, the second value is updated so that a maximum value of a packet discard rate when randomly discarding packets for each priority becomes 1; The calculation unit Using the updated first value and the second value, a queue length threshold at which random packet discarding is initiated, a maximum packet discard rate in the random discarding, and a queue length threshold at which tail discarding of packets is initiated are recalculated for each of the priorities.
5. The network control device according to claim 4.
6. The determination unit further After the recalculated value is set in the edge router, it is determined based on the discard information whether or not a state in which a discard rate of packets having a priority equal to or less than the predetermined value continues to be equal to or less than a predetermined value for a predetermined period of time in the edge router; The setting unit When it is determined that the discard rate of packets having the priority equal to or less than the predetermined value continues to be equal to or less than the predetermined value for a predetermined period of time in the edge router, the queue length threshold at which random discard of packets is started, the maximum discard rate in the random discard, and the queue length threshold at which tail discard is started are reset in the edge router, for each priority, calculated using the input first value and second value.
5. The network control device according to claim 4.
7. A network control method executed by a network control device, comprising: In QoS control for discarding packets by WRED (Weighted Random Early Detection) based on the length of a queue aggregating packets of traffic of each user and the priority set for the packets of traffic of each user, a process of receiving input of a first value indicating a ratio of an interval between a queue length threshold for starting random discard of packets set for each priority and a queue length threshold for starting tail discard of packets set for each priority, and a second value indicating a ratio of a maximum discard rate for each priority when randomly discarding packets set for each priority; a step of calculating, for each of the priorities, a queue length threshold at which random packet discarding is initiated, a maximum discard rate in the random discarding, and a queue length threshold at which tail discarding is initiated, the queue length threshold being used for the WRED, using the input first value and the second value; setting the calculated value in the edge router that performs QoS control using the WRED; A network control method comprising:
8. In QoS control for discarding packets by WRED (Weighted Random Early Detection) based on the length of a queue aggregating packets of traffic of each user and the priority set for the packets of traffic of each user, a process of receiving input of a first value indicating a ratio of an interval between a queue length threshold for starting random discard of packets set for each priority and a queue length threshold for starting tail discard of packets set for each priority, and a second value indicating a ratio of a maximum discard rate for each priority when randomly discarding packets set for each priority; a step of calculating, for each of the priorities, a queue length threshold at which random packet discarding is initiated, a maximum discard rate in the random discarding, and a queue length threshold at which tail discarding is initiated, the queue length threshold being used for the WRED, using the input first value and the second value; setting the calculated value in the edge router that performs QoS control using the WRED; A network control program that causes a computer to execute the above.
Citation Information
Patent Citations
Packet relay device and packet relay method
JP2011151601A
Service chaining system, priority transfer controller, service function transport apparatus and communication optimization method
JP2018129759A
Method and network node for traffic management at a network node in a packet-switched network
JP2018500851A
Network controller device, network control system, communication network control method, and program
JP2020136925A
Packet buffer device and packet discarding method
WO2010089886A1