Network parameter adjustment method, network device and storage medium

Abstract

The invention discloses a network parameter adjustment method, network equipment and a storage medium, and belongs to the technical field of communication. Under the condition that the first end-to-end time delay upper bound of a current service flow is greater than the end-to-end time delay constraint of the current service flow, the first end-to-end time delay upper bound of the current service flow is adjusted; the first network parameters of the N nodes are adjusted to obtain the second network parameters, meeting the first preset condition, of the N nodes, and the adjusted network parameters of the nodes do not influence the upper bound of end-to-end delay of other deployed service flows on the nodes, so that the problem that the end-to-end delay of the other deployed service flows on the nodes cannot be influenced after the adjusted network parameters are configured into the network nodes is solved. And the end-to-end time delay of the deployed service flow does not meet the time delay constraint.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to a network parameter adjustment method, a network device, and a storage medium. Background Art

[0002] In network communication, in order to avoid network congestion, packet loss, etc. caused by time delay, it is necessary to calculate the upper bound of the end-to-end time delay of the service flow. If the upper bound of the end-to-end time delay is greater than the end-to-end time delay constraint, it is necessary to adjust the parameter configuration of the shaper at the output port of the node through which the service flow passes and the configuration of other necessary parameters, and configure the adjusted parameters into the network node to avoid network congestion, packet loss, etc. caused by time delay.

[0003] However, after the adjusted parameters are configured into the network node currently, it may cause the time delay of the service flow already deployed in the network to increase after passing through the node with the reconfigured parameters, resulting in the end-to-end time delay of the deployed service flow not meeting the time delay constraint. Summary of the Invention

[0004] The main purpose of this application is to provide a network parameter adjustment method, a network device, and a storage medium, aiming to solve the technical problem that after the adjusted parameters are configured into the network node, the end-to-end time delay of the deployed service flow does not meet the time delay constraint.

[0005] To achieve the above purpose, an embodiment of this application provides a network parameter adjustment method, including:

[0006] Determine the first end-to-end time delay upper bound of the current service flow between two end nodes;

[0007] When the first end-to-end time delay upper bound is greater than the end-to-end time delay constraint of the current service flow, based on the first network parameters of the N nodes through which the current service flow passes, determine the second network parameters of the N nodes that meet the first preset condition, where at least one of the first network parameters and the second network parameters is different, and N is a positive integer;

[0008] The first preset condition includes: the second end-to-end time delay upper bound of the current service flow calculated based on the second network parameters of the N nodes is not greater than the end-to-end time delay constraint of the current service flow, and the end-to-end time delay upper bound of the deployed service flow passing through any of the N nodes calculated based on the second network parameters of the N nodes is not greater than the end-to-end time delay constraint of the deployed service flow.

[0009] An embodiment of the present application discloses a network parameter adjustment method. When the upper bound of the first end-to-end delay of the current service flow is greater than the end-to-end delay constraint of the current service flow, the first network parameters of N nodes are adjusted to obtain the second network parameters of the N nodes that meet the first preset condition. Since the upper bound of the second end-to-end delay of the current service flow calculated based on the second network parameters of the N nodes is not greater than the end-to-end delay constraint of the current service flow, and the upper bound of the end-to-end delay of the deployed service flow passing through any of the N nodes calculated based on the second network parameters of the N nodes is not greater than the end-to-end delay constraint of the deployed service flow, the network parameters of the adjusted nodes will not affect the upper bound of the end-to-end delay of other deployed service flows on the node, solving the technical problem that the end-to-end delay of the deployed service flow does not meet the delay constraint after the adjusted network parameters are configured into the network nodes.

[0010] An embodiment of the present application further provides a network device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the above method.

[0011] An embodiment of the present application further provides a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium, and when the computer program is executed by a processor, the above method is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a schematic structural diagram of an operating device of the hardware operating environment involved in the solution of the embodiment of the present application;

[0013] Figure 2 It is a schematic structural diagram of an operating device of the hardware operating environment involved in the solution of the embodiment of the present application;

[0014] Figure 3 It is a schematic flowchart of an embodiment of the method involved in the solution of the embodiment of the present application;

[0015] Figure 4 It is a schematic diagram of the service curve and the arrival curve in the method involved in the solution of the embodiment of the present application;

[0016] Figure 5 It is a schematic flowchart of an embodiment of the method involved in the solution of the embodiment of the present application;

[0017] Figure 6 It is a schematic flowchart of an embodiment of the method involved in the solution of the embodiment of the present application.

[0018] The implementation, functional features, and advantages of the purpose of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] It should be understood that the specific embodiments described herein are merely used to explain the present application and are not used to limit the present application.

[0020] For a clearer illustration, first, a network structure that can be used to implement the embodiments of the present application will be described. As Figure 1 shown, the above network includes a sending unit 102, a receiving unit 103, and one or more forwarding units 105.

[0021] The sending unit 102 is used to send traffic flows at the end. Then, the traffic flows can be forwarded in the network via the forwarding unit 105 and finally received by the receiving unit 103 at the end.

[0022] The network may also include an ingress shaper 104 and one or more per-hop shapers 106. The ingress shaper 104 is generally used to shape a single traffic flow or multiple traffic flows with the same forwarding path or the same forwarding target device flowing into the network at a certain period at the network ingress. The single traffic flow and the multiple traffic flows can be respectively referred to as single-flow and multi-flow. Before being shaped by the ingress shaper 104, the single-flow or multi-flow may have the same form or different forms.

[0023] One or more per-hop shapers 106 are generally used to shape the single-flow or multi-flow before or after the corresponding forwarding unit 105 forwards the single-flow or multi-flow for per-hop specification. The multi-flow is formed by aggregating multiple single-flows flowing through the forwarding unit 105. Through traffic flow shaping, congestion packet loss caused by traffic flow aggregation or per-hop transmission of burst traffic flows can be effectively avoided.

[0024] The above network structure is only a possible implementation form. In some possible designs, the sending unit 102 and the ingress shaper 104 can be integrated in the same device, such as the end sending node for sending traffic flows; or the ingress shaper 104 and the first forwarding unit 105 in the traffic flow transmission process can be integrated in the same device, such as the first forwarding node for forwarding traffic flows; or the ingress shaper 104 and the receiving unit 103 can be integrated in the same device, such as the end receiving node for receiving traffic flows. Similarly, the per-hop shaper 106 can also be integrated with the forwarding unit 105 in the same device.

[0025] As Figure 1As shown, the network may further include a controller 101 for managing and controlling, among others, one or more of the forwarding unit 105, the ingress shaper 104, and the per-hop shaper 106 on any network node during the end-to-end network transmission process. The management and control may include, for example, configuring the shaping parameters for the ingress shaper 104 and one or more per-hop shapers 106, and allocating the buffer size for one or more forwarding units 105 and / or per-hop specifications. In other possible designs, in addition to controlling the above-mentioned units, the controller 101 may also control the sending unit 102 and / or the receiving unit 103 to implement the control and management of the service flow sending and receiving.

[0026] The controller 101 may be deployed independently, that is, physically independent of other controlled functional units in the network (such as the sending unit 102, the forwarding unit 105, the ingress shaper 104, etc.), or may be integrated with a certain functional unit on the same device, or even split into several sub-units and arranged on multiple devices respectively, as long as the corresponding management and control functions can be jointly realized logically.

[0027] The controller 101, the sending unit 102, the receiving unit 103, the forwarding unit 105, the ingress shaper 104, or the per-hop shaper 106 may be implemented in the form of hardware, software, or a combination of both. It may be implemented as an independent device, for example, as an independent node in the network, or as a functional module or a combination of multiple functional modules on a network node, which can be selected and designed according to specific scenario requirements. One or more of the ingress shaper 104 and the per-hop shaper 106 may execute the same shaping policy or different shaping policies.

[0028] An embodiment of the present application provides a network structure as Figure 2 shown, including: a plurality of nodes connected in a network, and a controller connected to each node. Each node in the network can serve as both a sending node and a receiving node. A shaper may be integrated within the node. The controller can not only implement the control and management of the service flow sending and receiving on the node, but also perform control such as shaping parameter configuration and / or buffer allocation for the shaper of the node.

[0029] The nodes in the embodiments of the present application may be routers, switches, etc. The network parameter adjustment method in the embodiments of the present application is implemented by the controller connecting the nodes.

[0030] In network communication, to avoid network congestion, packet loss, etc. caused by time delay, it is necessary to calculate the upper bound of the end-to-end delay of the service flow. If the upper bound of the end-to-end delay is greater than the end-to-end delay constraint, it is necessary to adjust the parameter configuration of the shaper at the output port of the node through which the service flow passes and the configuration of other necessary parameters, and configure the adjusted parameters into the network node to avoid network congestion, packet loss, etc. caused by time delay.

[0031] Specifically, based on the configuration parameters of the network ingress shaper, the arrival curve of the service flow can be obtained, where the shaper parameters include the maximum burst size CBS of the data and the average output rate CIR, etc. Based on the shaper configuration parameters and other necessary parameters, the arrival curve and the service curve are calculated, so as to calculate the upper bound of the end-to-end delay DB of the current service flow. In addition, the end-to-end delay constraint of the service flow is DB_Cons. If the upper bound of the end-to-end delay DB of the current service flow is greater than the delay constraint DB_Cons, the parameters of the shaper are adjusted. The adjusted parameters can be one or more, so that the adjusted upper bound of the end-to-end delay DB1 is less than the end-to-end delay constraint DB_Cons, and the adjusted shaper parameters are configured into the network node to avoid network congestion and data packet loss caused by time delay.

[0032] However, this method for adjusting network shaper parameters only considers that the end-to-end delay of the current service flow meets the constraint conditions, and does not consider the impact of the adjustment of the current shaper parameters on the deployed service flows. After configuring the adjusted parameters into the network node, it may cause the delay of the deployed service flows in the network to increase after passing through the node with the reconfigured parameters, resulting in the end-to-end delay of the deployed service flows not meeting the delay constraint.

[0033] In view of this, this embodiment proposes a method for adjusting network parameters, which can solve the technical problem that the end-to-end delay of the deployed service flow does not meet the delay constraint after the adjusted parameters are configured into the network node. As Figure 3 shown, the method for adjusting network parameters provided in this embodiment includes the following steps:

[0034] Step S10: Determine the upper bound of the first end-to-end delay of the current service flow between two end nodes.

[0035] Specifically, the upper bound of the end-to-end delay of the service flow between two end nodes can be calculated through network calculus. Network calculus is an end-to-end delay calculation technology based on arrival curves and service curves. By introducing min-plus algebra, the network calculus theory transforms complex non-linear queuing problems into mathematical models that are easy to analyze, and then derives the system performance through the mathematical relationships between the models, and the delay and backlog boundaries of the network can be obtained. Through the arrival curve and service curve models of network nodes established by network calculus, the upper bound of the end-to-end delay of the service flow can be accurately calculated for different service flows, providing guidance for delay guarantee in service deployment.

[0036] The first upper bound of the end-to-end delay of the service flow between two end nodes can be calculated in the following two ways.

[0037] Method 1: Calculate the overall arrival curve α(t) and service curve β(t) of the current service flow passing through all N (N≥1) nodes respectively, calculate the maximum horizontal distance between α(t) and β(t), and determine the upper bound of the end-to-end delay DB of the current service flow. As Figure 3 shown, where α(t) is the overall arrival curve of the previous service flow passing through all N nodes, β(t) is the overall service curve of the service flow passing through all N nodes, and the maximum horizontal distance h between the overall arrival curve and the overall service curve is the upper bound of the end-to-end delay DB of the service flow passing through all N nodes.

[0038] Method 2: Calculate the arrival curve αn(t) and service curve βn(t) of the current service flow passing through N (N≥1) forwarding nodes respectively, and calculate the maximum horizontal distance between each corresponding αn(t) and βn(t).

[0039] As Figure 4 shown, if the arrival curve of the current service flow passing through the node is α i (t), and the service curve of the current service flow passing through the node is β i (t), calculate the maximum horizontal distance h(α,β) between the arrival curve α i (t) and the service curve β i (t), and the condition satisfied by the upper bound of the delay D(t) of the current service flow at each node can be obtained as shown in Formula 1:

[0040] D(t) ≤ h(α,β) (Formula 1)

[0041] Among them, the maximum horizontal distance h(α,β) between the arrival curve α i (t) and the service curve β i (t) can be calculated through the following Formula 2.

[0042]

[0043] After the path of the service flow is determined, the upper bound value DBk of the delay passing through all forwarding nodes can be calculated based on the arrival curve and service curve of network calculus, and the end-to-end delay upper bound value DB of the current service flow is obtained after accumulation, as shown in Formula 3 below.

[0044]

[0045] Where k = 1, 2, 3,... m, and a total of m forwarding nodes are passed through.

[0046] Step S20: When the first end-to-end delay upper bound is greater than the end-to-end delay constraint of the current service flow, determine the second network parameters of the N nodes that meet the first preset condition based on the first network parameters of the N nodes through which the current service flow passes.

[0047] The end-to-end delay constraint of the current service flow is DB_Cons. If the end-to-end delay upper bound DB of the current service flow is greater than the end-to-end delay constraint DB_Cons of the current service flow, network congestion, packet loss, etc. may occur when the current service flow is transmitted in the network due to excessive delay. At this time, it is necessary to adjust the first network parameters of the N nodes through which the current service flow passes so that the end-to-end delay upper bound DB of the current service flow is less than the end-to-end delay constraint DB_Cons under the new network parameters.

[0048] Each node has corresponding network parameters, such as: the shaper parameters of the node's outgoing port and other necessary parameters. The network parameters of the N nodes through which the service flow passes are collectively referred to as the first network parameters. The second network parameters of the N nodes can be obtained by adjusting the above network parameters of at least one of the N nodes through which the current service flow passes. As Figure 1 shown, the service flow flows from Node 1 to Node 6. Node 1 is the sending node, Node 6 is the receiving node, and Nodes 2 and 3 are forwarding nodes. The network parameters of Node 1 can be adjusted, or the network parameters of Nodes 1 and 2 can be adjusted, or the network parameters of Nodes 1, 2, and 3 can be adjusted, or the network parameters of Nodes 1, 2, 3, and 6 can be adjusted simultaneously.

[0049] At least one of the first network parameter and the second network parameter is different, that is, at least one of the first network parameters of the N nodes through which the previous service flow passes is adjusted to obtain the second network parameter. And based on the second network parameters of the N nodes obtained after adjustment, the first preset condition is satisfied. The upper bound of the second end-to-end delay of the current service flow calculated based on the second network parameters of the N nodes is not greater than the end-to-end delay constraint of the current service flow, and the upper bound of the end-to-end delay of the deployed service flow passing through any of the N nodes calculated based on the second network parameters of the N nodes is not greater than the end-to-end delay constraint of the deployed service flow. The network parameters of the adjusted node will not affect the upper bound of the end-to-end delay of other deployed service flows on the node, which solves the technical problem that the end-to-end delay of the deployed service flow does not meet the delay constraint after the adjusted network parameters are configured into the network node.

[0050] In this embodiment, the network delay constraints DB_Cons for different service flows may be different, and may be related to the service type carried by the service flow or the transmission rate requirement for the service flow during a specific period, or other possible network data transmission requirements. The delay constraint values DB_Cons corresponding to different service types can be stored in advance. For example, they can be stored in Figure 1 the controller 101 shown in the figure, or Figure 2 the controller shown in the figure, or they can also be stored in possible storage locations such as nodes, which can be specifically set according to needs. In practical applications, the network delay constraint DB_Cons can be automatically obtained by the controller based on the correspondence between the service type and the delay constraint, or manually configured by the network administrator, etc.

[0051] Exemplarily, the network parameters of the node include: average output rate, maximum burst size, weight of the priority queue in the scheduling algorithm, and the mapping relationship between the service flow and the priority of the node. Among them, the average output rate and the maximum burst size are the shaper parameters at the output port of the node, the weight of the priority queue in the scheduling algorithm is the parameter in the scheduling algorithm of the node, and the mapping relationship between the service flow and the priority of the node is the parameter related to the service flow of the node.

[0052] Exemplarily, the priority of the deployed service flow is the same as or lower than the priority of the current service flow.

[0053] Since the traffic flows with higher priorities on the nodes are sent or forwarded first, while the traffic flows with lower priorities need to wait until the traffic flows with higher priorities are sent or forwarded and then are sent or forwarded. Therefore, the deployment of the current traffic flow will have a great impact on the deployed traffic flows on the nodes with the same priority as the current traffic flow or with priorities lower than the current traffic flow. Therefore, in this embodiment, special attention should be paid to the impact of the adjusted second network parameters on the deployed traffic flows with priorities the same as or lower than the current traffic flow.

[0054] In some embodiments, as Figure 5 shown, the network parameter adjustment method provided in this embodiment includes the following steps:

[0055] Step S10: Determine the upper bound of the first end-to-end delay of the current traffic flow between two end nodes.

[0056] Step S20: When the upper bound of the first end-to-end delay is greater than the end-to-end delay constraint of the current traffic flow, based on the first network parameters of the N nodes through which the current traffic flow passes and the priority of the current traffic flow, determine the initial delay upper bounds of each of the N nodes.

[0057] Specifically, when determining the initial delay upper bounds of each of the N nodes, not only the network parameters of each node are considered but also the priority of the current traffic flow is considered. Since the traffic flows with higher priorities on the nodes are sent or forwarded first, while the traffic flows with lower priorities need to wait until the traffic flows with higher priorities are sent or forwarded and then are sent or forwarded. Therefore, the deployment of the current traffic flow will have a great impact on the deployed traffic flows on the nodes with priorities lower than the current traffic flow. Therefore, in this embodiment, when determining the initial delay upper bounds of each node, it is necessary to jointly determine according to the first network parameters of the N nodes through which the current traffic flow passes and the priority of the current traffic flow.

[0058] Exemplarily, based on the first network parameters of the N nodes through which the current traffic flow passes and the priority of the current traffic flow, determining the initial delay upper bounds of each of the N nodes includes: based on the shaper parameters of each of the N nodes and the priority of the current traffic flow, determining the arrival curve of the current traffic flow at each node; based on the scheduling algorithms of each of the N nodes and the weights of the priority queues with the same priority as the current traffic flow in the scheduling algorithms of each node, determining the service curve of the current traffic flow at each node; based on the arrival curve of each node and the service curve of each node, calculating the initial delay upper bounds of each of the N nodes.

[0059] Specifically, the upper bound of the delay of the traffic flow on the node can be calculated by determining the arrival curve and the service curve of the traffic flow on the node.

[0060] The arrival curve α(t) of the service flow at the node can be determined according to the shaper parameters of the node and the priority of the current service flow, as follows:

[0061] Different priority queues have different shaper parameter configurations. Assume that the priority of the current service flow is i, and the current service flow enters the corresponding queue according to the priority. The arrival curve α of the service flow with priority i at the node can be calculated based on the priority of the service flow and the shaper parameters of the queue to which the priority of the service flow belongs. i (t) is shown in the following formula (4):

[0062] α i (t) = cir i *t + cbs i (Formula 4)

[0063] Where, the priority i of the service flow ∈ {0, 1, 2,..., n - 1}, n is the number of priority queues, cbs i and cir i are the shaper parameters of the queue to which the priority of the service flow belongs. Among them, cbs i is the maximum burst volume of the shaper parameter of the queue with priority i, and cir i is the average output rate of the shaper parameter of the queue with priority i.

[0064] The service curve β(t) of the service flow at the node can be determined according to the scheduling algorithm of the node and the weights of the priority queues with the same priority as the current service flow in the scheduling algorithm of each node, so as to determine the service curve of the current service flow at each node.

[0065] There are various scheduling models for nodes, including the Strict Priority (SP for short) algorithm, the Weighted Fair Queue (WFQ for short) algorithm, the Weighted Round Robin (WRR for short) algorithm, etc. Based on different scheduling models, the service curves of the current service flow calculated at each node are also different. Due to different priorities of services, high-priority service flows will be served first, which affects the service curve of the current-priority service flow. In addition, the impact of the current service flow on low-priority service flows also needs to be considered. Therefore, when calculating the service curve, high-priority service flows and low-priority service flows need to be considered.

[0066] (1) When the scheduling algorithm of the current node is the strict priority scheduling algorithm, based on the general service curve of the current node, the arrival curve of the deployed service flows with priorities higher than the current service flow, and the maximum packet length in the deployed service flows with priorities lower than the current service flow, determine the service curve of the current service flow at the current node.

[0067] Specifically, the service curve of the current service flow at the current node is the curve obtained by subtracting the arrival curve of the deployed service flows with priorities higher than the current service flow from the general service curve of the current node, and then subtracting the maximum packet length in the deployed service flows with priorities lower than the current service flow.

[0068] Assume that the general service curve of the current node is β(t), and the maximum packet length in the deployed service flows with priorities lower than the current service flow is Then, based on the strict priority scheduling algorithm, the service curve β i (t) provided to the current service flow with queue priority i is as shown in Formula 5.

[0069]

[0070] (2) When the scheduling algorithm of the current node is the weighted fair queue scheduling algorithm, based on the ratio of the weight of the priority queue to which the priority of the current service flow belongs to the sum of the weights of all priority queues, the general service curve of the current node, and the upper limit value of the packet length in each priority queue, determine the service curve of the current service flow at the current node.

[0071] Specifically, the service curve of the current service flow at the current node is the curve obtained by multiplying the ratio of the weight of the priority queue to which the priority of the current service flow belongs to the sum of the weights of all priority queues by the general service curve of the current node, and then subtracting the upper limit value of the packet length in each priority queue, and taking the maximum value compared with 0.

[0072] Assume that the weight of the priority queue to which the priority of the current service flow belongs is represented by w i The general service curve of the current node is represented by β(t), and the upper limit value of the packet length in each priority queue is represented by l u Then, based on the weighted fair queue scheduling algorithm, the service curve β i (t) provided to the current service flow with queue priority i is as shown in Formula 6.

[0073]

[0074] Among them, [x] + = max{x, 0} is to take the maximum value.

[0075] (3) When the scheduling algorithm of the current node is the weighted round-robin scheduling algorithm, based on the general service curve of the current node, the minimum number of data volume bits obtained by the current traffic flow in one round-robin from the priority queue to which the priority belongs, and the maximum number of data volume bits obtained by other queues except the priority queue to which the priority of the current traffic flow belongs in one round-robin, determine the service curve of the current traffic flow at the current node.

[0076] Specifically, the service curve of the current traffic flow at the node is the curve obtained by taking the maximum value of the value of the curve obtained by subtracting the sum of the maximum number of data volume bits obtained by other queues except the priority queue to which the priority of the current traffic flow belongs in one round-robin from the general service curve of the current node compared with 0, and then multiplying it by the ratio of the minimum number of data volume bits obtained by the priority queue to which the priority of the current traffic flow belongs in one round-robin to the sum of the minimum number of data volume bits obtained by the priority queue to which the priority of the current traffic flow belongs in one round-robin and the maximum number of data volume bits obtained by other queues except the priority queue to which the priority of the current traffic flow belongs in one round-robin.

[0077] Assume that the general service curve of the current node is represented by β(t), and the minimum number of data volume bits obtained by the priority queue to which the priority of the current traffic flow belongs in one round-robin is represented by q i denotes, where, w i is the weighted priority of the priority queue to which the priority of the current traffic flow belongs, is the minimum value of the packet length of the queue with priority i; the sum of the maximum number of data volume bits obtained by other queues except the priority queue to which the priority of the current traffic flow belongs in one round-robin is represented by Q i denotes, where, w j is the weight of other priority queues except the priority queue to which the priority of the current traffic flow belongs, is the maximum value of the packet length of other priority queues except the priority queue to which the priority of the current traffic flow belongs. Then, based on the weighted round-robin scheduling algorithm, the service curve β i (t) provided to the current traffic flow with queue priority i is as shown in Formula 7:

[0078]

[0079] where, [x] + = max{x, 0} is to take the maximum value.

[0080] In this embodiment, according to the above formula, the arrival curve and service curve of each of the N nodes through which the current service flow passes can be calculated, and the initial delay upper bound of each node can be calculated based on the arrival curve and service curve of each node.

[0081] Step S30: Sequentially adjust the network parameters of the node with the largest initial delay upper bound among the N nodes and whose network parameters have not been adjusted until the second network parameters of the N nodes that meet the first preset condition are obtained.

[0082] After obtaining the initial delay upper bounds of each node among the N nodes through which the current service flow passes, before adjusting the network parameters of the nodes, first sort the N nodes in descending order of the initial delay upper bound. When adjusting the network parameters of the nodes, first adjust the network parameters of the node with the largest initial delay upper bound according to the sorted order. If the second network parameters of the N nodes that meet the first preset condition are not obtained after the adjustment, then adjust the network parameters of the unadjusted node with the largest initial delay upper bound until the second network parameters of the N nodes that meet the first preset condition are obtained. In this embodiment, the node with a larger initial delay upper bound has a larger adjustable range, and preferentially adjusting the node with the largest and unadjusted initial delay upper bound can speed up the adjustment of network parameters.

[0083] In this embodiment, the end-to-end delay upper bound of the current service flow has been calculated as DB, the delay upper bound value passing through the k-th forwarding node is DBk, and the delay upper bound values passing through m nodes are arranged in descending order as DBmaxj, where j = 1, 2, 3,... m. Select the node with the largest node delay upper bound value, DBmax_j (j = 1), as the adjustment node. The adjusted parameters include: the maximum burst size cbs of the current priority queue shaper i and the average output rate cir i and the scheduling weight value w of each queue i etc.

[0084] It is recommended to preferentially adjust the average output rate cir i , followed by adjusting the maximum burst size cbs i . If it is the WFQ or WRR scheduling algorithm, adjust the weight parameter w i . Finally, the internal priority mapping relationship of the SP can also be adjusted. The adjusted parameter can be one of the parameters or a set of multiple parameters.

[0085] The node after adjusting the parameters needs to meet the following conditions:

[0086] (1) After the adjustment, the delay upper bound value of the current priority service flow at this node becomes smaller, and the recalculated value is BDmax_j`<DBmax_j.

[0087] (2) After adjusting the parameters, the upper bound of the delay of the low-priority traffic flow at this node will change. Recalculate the upper bound of the delay of the low-priority traffic flow at this node, and check and calculate that the end-to-end delay upper bound of the deployed traffic flow passing through this node after adjusting the parameters cannot exceed the delay constraint value of the traffic flow.

[0088] In this embodiment, the adjusted network parameters of the node satisfy the second preset condition, and the second preset condition includes: the adjusted upper bound of the delay of the current traffic flow passing through the node calculated based on the adjusted network parameters of the node is not greater than the initial upper bound of the delay of the node, and the end-to-end delay upper bound of the deployed traffic flow passing through the node calculated based on the adjusted network parameters of the node is not greater than the end-to-end delay constraint of the deployed traffic flow. That is to say, after the network parameters of the node are adjusted, the upper bound of the delay of the current traffic flow on this node cannot become larger, and the end-to-end delay upper bound of the deployed traffic flow passing through this node can become larger or smaller, but cannot be greater than the end-to-end delay constraint of the deployed traffic flow.

[0089] Since different scheduling strategies and different traffic flow priorities will have a great impact on the end-to-end delay of the deployed traffic flow, it is necessary to find a globally coordinated parameter adjustment method, that is, to make the delay of the current traffic flow meet the constraint requirements, and at the same time minimize the impact on the delay of other traffic flows. This embodiment proposes a method for adjusting network shaping parameters for the whole network traffic, so that the delay constraint of the current traffic flow can be satisfied, and at the same time it does not affect the delay constraints of other deployed traffic flows.

[0090] In some embodiments, in the process of sequentially adjusting the network parameters of the node with the largest initial upper bound of the delay and whose network parameters have not been adjusted among the N nodes, it includes: after adjusting the network parameters of one node with the largest initial upper bound of the delay and whose network parameters have not been adjusted among the N nodes, determine the sum of the reduced delay differences of the current traffic flow passing through the adjusted node; determine the total difference between the first end-to-end delay upper bound and the end-to-end delay constraint of the current traffic flow; based on the total difference and the sum of the reduced delay differences, determine whether the end-to-end delay upper bound of the adjusted current traffic flow is not greater than the end-to-end delay constraint of the current traffic flow; if it is not greater, adjust the network parameters of the next node with the largest initial upper bound of the delay and whose network parameters have not been adjusted among the N nodes until the end-to-end delay upper bound of the adjusted current traffic flow is not greater than the end-to-end delay constraint of the current traffic flow.

[0091] In this embodiment, in the process of sequentially adjusting the network parameters of the node with the largest initial delay upper bound among the N nodes and whose network parameters have not been adjusted, if the initial delay upper bound of the current traffic flow passing through the node before adjusting the network parameters of a node is DBmax_j, and the initial delay upper bound of the current traffic flow passing through the node after adjusting the network parameters of a node is DBmax_j', then the reduced delay difference of the current traffic flow passing through the adjusted node after adjustment is DBmax_j - DBmax_j'.

[0092] Assume that the first end-to-end delay upper bound of the current traffic flow between two end nodes is DB, and the delay constraint of the current traffic flow between two end nodes is DB_Cons. Then, after adjusting the node, the difference DIFF can be calculated by the following formula 8 to determine whether the end-to-end delay upper bound of the current traffic flow after adjustment is not greater than the end-to-end delay constraint of the current traffic flow.

[0093]

[0094] If Diff > 0, then after adjustment, the end-to-end delay upper bound DB of the current traffic flow is less than the end-to-end delay constraint DB_Cons of the current traffic flow, meeting the requirement; if Diff < 0, then after adjustment, the end-to-end delay upper bound DB of the current traffic flow is still greater than the end-to-end delay constraint DB_Cons of the current traffic flow, not meeting the requirement, and select the node DBmaxj (j = 2) with the largest initial delay upper bound and not yet adjusted as the adjustment node. Repeat the above adjustment process until Diff > 0.

[0095] An embodiment of the present application proposes a network parameter adjustment method based on network calculus. Calculate the arrival curve and service curve of the current traffic flow at each node based on the network shaper configuration parameters and the scheduling model algorithm, and at the same time calculate the delay upper bound value. Calculate the end-to-end delay value of the current traffic flow by accumulating the delay upper bound values of the nodes along the service path. When the end-to-end delay upper bound value of the current traffic flow is greater than the end-to-end delay constraint value of the current traffic flow, perform network parameter adjustment on the forwarding node. The adjusted parameters include the maximum burst volume of the shaper parameters of the priority queue, the average output rate of the shaper parameters of the priority queue, the queue scheduling weight, etc. First, adjust the node with a larger node delay upper bound. After adjusting the parameters, it is necessary to ensure that the node delay upper bound value of the current traffic flow decreases, and at the same time, the end-to-end delay of other deployed traffic flows passing through this node must also meet the constraint conditions.

[0096] The current solution is to adjust the satisfaction of a single service flow, adjust parameters such as the burst volume and average transmission rate of the shaper. After the adjustment, the impact on other flows is not considered. The shaper parameters of queues with different priorities are different. Adjusting the shaper parameters of one queue reduces the delay of the current service flow, but the impact on the delays of service flows with other priorities may increase. In the embodiments of the present application, global parameter adjustment is performed, that is, the delay of the current service flow meets the constraint requirements, and at the same time, the delay constraints of other service flows are not affected. The impact on other service flows after the network parameters are adjusted is quantified to guide the parameter adjustment.

[0097] The shaper parameters and scheduling weights configured in current network devices both have default configuration values and will not change once configured because adjusting one of them will affect other service flows and there is no quantitative index reference for the degree of impact. Network calculus can control the impact degree of the entire network, provide an adjustable method, quantify the impact degree after adjustment, and guide the parameter adjustment.

[0098] The following gives the specific process of network parameter adjustment, as Figure 6 shown, including:

[0099] Step 1: Based on the network parameters and service flow parameters, calculate the arrival curve and service curve of the service flow through the above formula 4-7.

[0100] Step 2: Calculate the upper bound of the delay DBk of the service flow passing through each node based on formula 8, and at the same time calculate the end-to-end upper bound of the delay DB;

[0101] Step 3: If the end-to-end upper bound of the delay DB is greater than the end-to-end delay constraint value DB_Cons, sort the upper bounds of the delays of the service flow passing through all nodes in descending order as DBmax_j.

[0102] Step 4: Select the DBmax_j with the largest remaining sorted node delay upper bound value as the adjustment node. The adjusted parameters include: the maximum burst volume cbsi of the shaper of the current priority queue, the average output rate ciri, and the scheduling weight values of each queue, etc. After adjusting the parameters, recalculate the upper bound of the delay BDmax_j` of the current service flow at this node to satisfy BDmax_j`<DBmax_j, and at the same time, the end-to-end delay of the deployed service flow passing through this node is less than the corresponding delay constraint value;

[0103] Step 5: Calculate. If Diff<0, the end-to-end upper bound of the delay DB of the current service flow after adjustment is still greater than the delay constraint DB_Cons value and does not meet the requirements. Go to Step 4; if Diff>0, after adjustment, the end-to-end upper bound of the delay DB of the current service flow is less than the delay constraint DB_Cons value and meets the requirements. End the process.

[0104] The following is a specific description of the Figure 6 process shown below.

[0105] Example 1: The forwarding node is implemented based on the strict priority scheduling algorithm. Assume that the network topology diagram is as Figure 2 shown, with a total of 6 nodes. There is a controller in the upper layer that performs end-to-end delay calculation and parameter adjustment through network calculus. All link delays are ignored, and only the upper bound of node delay is accumulated when calculating the end-to-end delay. The shaper parameters of the 6 forwarding nodes with default configurations are the same. The priority of traffic flow 1 is 5, the end-to-end delay constraint DB_Cons = 40 us, the source node is node1, the destination node is node6, and the nodes passed through on the path are node1, node2, node3, and node6 respectively. Assume that the shaper parameters cir5 = 512 kbit / s and cbs5 = 32000 byte are configured for the queue with priority 5 at all nodes. Calculate the arrival curve and service curve of traffic flow 1 at nodes node1, node2, node3, and node6 respectively, and the upper bound values of the corresponding delays at each node are DB1 = 9 us, DB2 = 8 us, DB3 = 13 us, and DB6 = 11 us. After accumulation, calculate the end-to-end delay upper bound DB = DB1 + DB2 + DB3 + DB6 = 9 + 8 + 13 + 11 = 41 us, which is greater than the end-to-end delay constraint of 40 us. Then parameter adjustment is required. Sort DB1 to DB6 from largest to smallest to get DBmax_1 = 13 us, DBmax_2 = 11 us, DBmax_3 = 9 us, and DBmax_4 = 8 us, corresponding to nodes node3, node6, node1, and node2 respectively. Select the node node3 corresponding to the largest delay DBmax_1 = 13 us for parameter adjustment. Assume that after adjustment, cbs5 = 22000 byte, and recalculate the delay upper bound DBmax_1` = 11 us of traffic flow 1 at node3. The new delay upper bound value is less than DBmax_1, meeting condition 1 after parameter adjustment. At the same time, check the deployed traffic flow passing through node3, assume it is traffic flow 2 with a priority of 4, recalculate the delay upper bound value of traffic flow 2 at node3 according to the adjusted parameters, and refresh the end-to-end delay value. If the value is less than the end-to-end delay constraint value of traffic flow 2, then condition 2 after parameter adjustment is met.

[0106] Calculate the delay difference diff after adjusting node3 = (DBmax_1 - DBmax_1`) - (DB - DB_Cons) = (13 - 11) - (41 - 40) = 1. Since the difference is greater than 0, the end-to-end delay upper bound DB of traffic flow 1 after adjusting the parameter cbs5 is 39 us, which is less than the delay constraint DB_Cons value of 40 us. The end-to-end delay constraint condition is met.

[0107] Embodiment 2: The forwarding node is implemented based on weighted fair queue scheduling. Assume that the network topology diagram is as Figure 2 shown, with a total of 6 nodes. There is a controller in the upper layer that performs end-to-end delay calculation and parameter adjustment through network calculus. All link delays are ignored, and only the upper bound of node delay is accumulated when calculating the end-to-end delay. The shaper parameters of the 6 forwarding nodes are the same by default, and the default queue weight parameters of the scheduling algorithm are the same. The WFQ scheduling weight parameters of priority queue 1 - priority queue 4 are w1:w2:w3:w4 = 1:1:1:1. The priority of traffic flow 1 is 4, the end-to-end delay constraint DB_Cons = 40 us, the source node is node1, the destination node is node6, and the nodes passed through on the path are node1, node2, node3, and node6 respectively. Assume that the shaper parameters cir5 = 512 kbit / s and cbs5 = 32000 byte are configured for the queue with priority 4 of all nodes. Calculate the arrival curve and service curve of traffic flow 1 at nodes node1, node2, node3, and node6 respectively, and the upper bound values of the corresponding delays at each node are DB1 = 9 us, DB2 = 8 us, DB3 = 15 us, and DB6 = 12 us. After accumulation, calculate the end-to-end delay upper bound DB = DB1 + DB2 + DB3 + DB6 = 9 + 8 + 15 + 12 = 44 us, which is greater than the end-to-end delay constraint of 40 us. Then parameter adjustment is required. Sort DB1 - DB6 from largest to smallest to get DBmax_1 = 15 us, DBmax_2 = 12 us, DBmax_3 = 9 us, and DBmax_4 = 8 us, corresponding to nodes node3, node6, node1, and node2 respectively. Select the node node3 corresponding to the largest delay DBmax_1 = 15 us for parameter adjustment. Assume that after adjustment, cir4 = 256 kbit / s and cbs3 = 42000 byte, and the weight parameters are w1:w2:w3:w4 = 1:1:2:2. Recalculate the upper bound of the delay DBmax_1` = 13 us of traffic flow 1 at node3. The new upper bound value of the delay is less than DBmax_1, satisfying condition 1 after parameter adjustment. At the same time, check the deployed traffic flow passing through node3, assume it is traffic flow 2 with a priority of 3, recalculate the upper bound value of the delay of traffic flow 2 at node3 according to the adjusted parameters, and refresh the end-to-end delay value. Its value is less than the end-to-end delay constraint value of traffic flow 2, then condition 2 after parameter adjustment is satisfied.

[0108] Calculate the delay difference diff after adjusting node node3: diff = (DBmax_1 - DBmax_1`) - (DB - DB_Cons) = (15 - 13) - (44 - 40) = -2. Since the difference is less than 0, the upper bound of the end-to-end delay DB of traffic flow 1 after adjusting the parameters cbs3, cir4, and the weight parameter is 42 us, which is greater than the delay constraint value DB_Cons of 40 us, not meeting the end-to-end delay constraint condition.

[0109] Continue to select the node node6 corresponding to the maximum delay DBmax_2 = 12 us among the remaining nodes for parameter adjustment. Assume that the adjusted weight parameters are w1:w2:w3:w4 = 1:1:1:2. Recalculate the upper bound of the delay DBmax_2` = 9 us of traffic flow 1 at node6. The new upper bound value of the delay is less than DBmax_2, meeting condition 1 after adjusting the parameters. At the same time, check the deployed traffic flows passing through node6. Assume it is traffic flow 3 with a priority of 6. Recalculate the upper bound of the delay value of traffic flow 3 at node6 according to the adjusted parameters and refresh the end-to-end delay value. If the value is less than the end-to-end delay constraint value of traffic flow 3, then condition 2 after adjusting the parameters is met.

[0110] Continue to calculate the delay difference diff after adjusting node node6: diff = (DBmax_1 - DBmax_1`) + (DBmax_2 - DBmax_2`) - (DB - DB_Cons) = (15 - 13) + (12 - 9) - (44 - 40) = 1. Since the difference is greater than 0, the upper bound of the end-to-end delay DB of traffic flow 1 after adjusting the parameters is 39 us, which is less than the delay constraint DB_Cons value of 40 us. The end-to-end delay constraint condition is met. The adjusted parameters include cbs3, cir4, and the weight parameter of node3, and the weight parameter of node6.

[0111] In addition, an embodiment of the present application further provides a network device, where the network device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. The computer program is configured to implement the method described above or the steps of the network parameter adjustment method described above.

[0112] In addition, an embodiment of the present application further provides a storage medium. The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the method described above or the network parameter adjustment method described above are implemented.

[0113] It should be noted that in this text, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or system comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or system. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or system comprising that element.

[0114] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described example methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium as described above (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions for causing a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the present application.

[0115] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A method for adjusting network parameters, characterized in that, Including: Determine an upper bound of the first end-to-end delay of the current service flow between two end nodes; When the upper bound of the first end-to-end delay is greater than the end-to-end delay constraint of the current service flow, based on the first network parameters of N nodes through which the current service flow passes, determine the second network parameters of the N nodes that meet the first preset condition, where at least one of the first network parameters and the second network parameters is different, and N is a positive integer; The first preset condition includes: the upper bound of the second end-to-end delay of the current service flow calculated based on the second network parameters of the N nodes is not greater than the end-to-end delay constraint of the current service flow, and the upper bound of the end-to-end delay of the deployed service flow passing through any of the N nodes calculated based on the second network parameters of the N nodes is not greater than the end-to-end delay constraint of the deployed service flow.

2. The method for adjusting network parameters according to claim 1, characterized in that, The determining the second network parameters of the N nodes that meet the first preset condition based on the first network parameters of the N nodes through which the current service flow passes includes: Based on the first network parameters of the N nodes through which the current service flow passes and the priority of the current service flow, determine the initial delay upper bound of each node in the N nodes; Adjust the network parameters of the node with the largest initial delay upper bound and whose network parameters have not been adjusted among the N nodes in sequence until the second network parameters of the N nodes that meet the first preset condition are obtained, where the network parameters of the adjusted node meet the second preset condition; The second preset condition includes: the adjusted delay upper bound of the current service flow passing through the node calculated based on the adjusted network parameters of the node is not greater than the initial delay upper bound of the node, and the upper bound of the end-to-end delay of the deployed service flow passing through the node calculated based on the adjusted network parameters of the node is not greater than the end-to-end delay constraint of the deployed service flow.

3. The method for adjusting network parameters according to claim 2, characterized in that, The determining the initial delay upper bound of each node in the N nodes based on the first network parameters of the N nodes through which the current service flow passes and the priority of the current service flow includes: Based on the shaper parameters of each node in the N nodes and the priority of the current service flow, determine the arrival curve of the current service flow at each node; Based on the scheduling algorithm of each node in the N nodes and the weight of the priority queue with the same priority as the current service flow in each node's scheduling algorithm, determine the service curve of the current service flow at each node; Based on the arrival curve of each node and the service curve of each node, calculate the initial delay upper bound of each node in the N nodes.

4. The method for adjusting network parameters according to claim 3, characterized in that, The determining the service curve of the current service flow at each node based on the scheduling algorithm of each node in the N nodes and the weight of the priority queue with the same priority as the current service flow in each node's scheduling algorithm includes: When the scheduling algorithm of the current node is the strict priority scheduling algorithm, based on the general service curve of the current node, the arrival curve of the deployed service flows with priorities higher than that of the current service flow, and the maximum packet length in the deployed service flows with priorities lower than that of the current service flow, determine the service curve of the current service flow at the current node.

5. The method for adjusting network parameters according to claim 3, characterized in that, The determining the service curve of the current service flow at each node based on the scheduling algorithms of the N nodes and the weights of the priority queues with the same priorities as the current service flow in the scheduling algorithms of the nodes includes: When the scheduling algorithm of the current node is the weighted fair queue scheduling algorithm, based on the ratio of the weight of the priority queue to which the priority of the current service flow belongs to the sum of the weights of all priority queues, the general service curve of the current node, and the upper limit value of the packet length in each priority queue, determine the service curve of the current service flow at the current node.

6. The method for adjusting network parameters according to claim 3, characterized in that, The determining the service curve of the current service flow at each node based on the scheduling algorithms of the N nodes and the weights of the priority queues with the same priorities as the current service flow in the scheduling algorithms of the nodes includes: When the scheduling algorithm of the current node is the weighted round-robin scheduling algorithm, based on the general service curve of the current node, the minimum number of data bits served by the priority queue to which the priority of the current service flow belongs in one round-robin, and the maximum number of data bits served by other queues except the priority queue to which the priority of the current service flow belongs in one round-robin, determine the service curve of the current service flow at the current node.

7. The method for adjusting network parameters according to claim 2, characterized in that, In the process of successively adjusting the network parameters of the node among the N nodes with the largest initial delay upper bound and whose network parameters have not been adjusted, it includes: After adjusting the network parameters of one node among the N nodes with the largest initial delay upper bound and whose network parameters have not been adjusted, determine the sum of the reduced delay differences of the current service flow passing through the adjusted node; Determine the total difference between the first end-to-end delay upper bound and the end-to-end delay constraint of the current service flow; Based on the total difference and the sum of the reduced delay differences, determine whether the end-to-end delay upper bound of the adjusted current service flow is not greater than the end-to-end delay constraint of the current service flow; In the case of not being greater than, adjust the network parameters of the next node among the N nodes with the largest initial delay upper bound and whose network parameters have not been adjusted until the end-to-end delay upper bound of the adjusted current service flow is not greater than the end-to-end delay constraint of the current service flow.

8. The method for adjusting network parameters according to any one of claims 1 to 7, characterized in that, The network parameters of the node include: average output rate, maximum burst volume, weights of the priority queues in the scheduling algorithm, and the mapping relationship between the service flows and priorities of the node.

9. The method for adjusting network parameters according to any one of claims 1 to 7, characterized in that, The priorities of the deployed service flows are the same as or lower than the priority of the current service flow.

10. A network device, characterized in that, The network device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the method according to any one of claims 1 to 9.

11. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the method according to any one of claims 1 to 9.

Images