Network resource redistribution method

Abstract

The invention relates to a network resource redistribution method, and belongs to the technical field of network switching. The method comprises the following steps: initializing system parameters; when entering a path maintenance period, setting a current path serial number, and entering a path of the current serial number; traversing all current paths generated based on the bisecting moth optimization method, and performing path maintenance on the current paths; the state of each node in the current path is checked, and if a failure node exists, a network resource scheduling method based on a bipartite moth optimization method is adopted to carry out failure node resource redistribution; checking the state of each link of the current path, if a failed link or a congested link exists, regarding the failed link or the congested link as the failed link, and performing failed link resource redistribution by adopting a network resource scheduling method based on a bipartite moth optimization method; and entering the circulation again until the system finishes working. The problem that local correction cannot be performed when the resource scheduling strategy fails is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of network switching, in particular to a network resource reallocation method. BACKGROUND

[0002] In the process of network operation, resource scheduling strategies are often calculated under certain network topology, node state and link conditions. However, when the network environment changes dynamically, for example, the network topology changes frequently, some nodes are damaged or offline, the link is disturbed by external interference, the channel is congested or even interrupted, the original resource scheduling strategy often loses its applicability and cannot continue to guarantee the stable operation and efficient utilization of the network. In this case, network resources must be reallocated.

[0003] In the prior art, especially in the resource scheduling scene based on intelligent algorithms, the global recalculation process often involves complex optimization operations and model reasoning, which occupies a large amount of processor computing power, thereby significantly increasing the energy consumption of network devices, and even causing the overall performance of the system to decline, which is difficult to bear in practical applications. Whether the prior art adopts traditional non-machine learning scheduling methods or relies on advanced scheduling methods such as deep learning and reinforcement learning, most resource reallocation methods usually adopt the way of discarding the original scheduling strategy and generating a new scheduling scheme by global calculation. However, this method has obvious disadvantages: when the resource scheduling strategy fails, the network topology changes rapidly, and if a complete recalculation process is triggered every time, it is impossible to make fine-grained dynamic adjustments to resource scheduling in a local range, and it lacks flexibility. Therefore, in the application scenarios where resources are scarce, energy is limited, and computing power is limited, it often shows problems such as slow response, low resource utilization, and high energy consumption, which is difficult to meet the actual application requirements of high efficiency, low energy consumption and real-time performance. SUMMARY

[0004] The present application provides a network resource reallocation method, which can solve the problem of poor flexibility caused by the inability to make local adjustments when the resource scheduling strategy fails, and the consumption of a large amount of computing resources and energy.

[0005] In order to achieve the above-mentioned purpose, according to the first aspect of the present application, a network resource reallocation method is provided, which comprises the following steps: S1, entering a path maintenance period, setting a current path serial number, and entering the path of the current serial number; S2, generating a scheduled path for the path of the current serial number; S3, checking the state of each node in the scheduled path, and if there is a failed node, performing resource reallocation for the failed node; S4, checking the status of each link of the scheduled path, if there is a failed link or a congested link, regarding the failed link or the congested link as a failed link, and performing resource redistribution of the failed link; S5, re-entering S1 until no resource redistribution of failed nodes and resource redistribution of failed links is needed.

[0006] Optionally, before entering S2, the method further comprises system initialization, and parameters of the system initialization comprise a current path , a failed node number threshold , a failed link number threshold , and a sub-path adjustment number .

[0007] Further, S3 comprises the following steps: S301, detecting a path failed node number ; S302, if , entering S303, otherwise entering S304; S303, judging a current service path as a failed path, re-generating the service path by using a network resource scheduling method based on a bisection fly optimization method, and ending the resource redistribution; S304, reading a previous node and a next node of all failed nodes in a path , respectively, to obtain an adjustment node set ; S305, obtaining a local path flag matrix of the failed nodes based on a service fitness weight multi-sampling method according to the adjustment node set , wherein ; ; S306, initializing a local path serial number ; S307, setting a sub-source node and a sub-destination node , to obtain a local path , wherein is a first row and a first column element of a matrix , and the like; q ; S308, replacing a sub-path between same nodes in a path with each local path , to obtain an adjustment path ; S309, if the adjustment path has two or more same nodes , enter S310, otherwise enter S311; S310, delete the path between the frontmost node and the rearmost node, only keep one node , enter S309; S311, if , then , enter S307, otherwise enter S312; S312, complete the calculation of all adjusted paths, select the adjusted path with the optimal service fitness replacement path .

[0008] Further, the local path generation step is: specify the sub-source node and the sub-destination node, and take the remaining TTL packet survival time as the maximum length of the path , a network resource scheduling method based on the bisection fly optimization algorithm is used to generate the local path, and the remaining TTL packet survival time is calculated by the following formula:

[0009] wherein is the length of the remaining path of the path between node and node , node and node are the sub-source node and the sub-destination node specified for the local path, and path is the corresponding path to which the local path belongs.

[0010] Further, the service fitness of S312 is calculated according to the following formula:

[0011] wherein , , and are the delay coefficient, the bandwidth coefficient, the packet loss coefficient and the jitter coefficient respectively, , , and are the delay function, the bandwidth function, the packet loss function and the jitter function of the service path j respectively.

[0012] Further, the delay function , the bandwidth function , the packet loss function and the jitter function of S312 are defined by the following formulas respectively:

[0013]

[0014]

[0015]

[0016] wherein , , and are coefficients greater than 0, , , and are the delay, bandwidth, packet loss rate and jitter of the th node in the path , is the total number of nodes in the path .

[0017] Further, the S305 comprises the following steps: S3051, calculating the average service fitness of the local path:

[0018]

[0019]

[0020]

[0021] wherein, is the average service fitness of the path , is the hop number of the path , is the path between the th node and the th node in the path , and the th node and the th node cannot be simultaneously the source node and the destination node of the path S3052, selecting the first local path with the maximum service fitness , wherein is used to mark the ranking number of the service fitness; S3053, obtaining the adjusted local path flag matrix according to the following formula: .

[0022] Further, the S4 comprises the following steps: S401, detecting the number of failed paths ; S402, if , entering S403, otherwise entering S404; S403, regarding the current service path j as a failed path, using a network resource scheduling method based on the bisection fly optimization method to regenerate the service path, and ending the resource reallocation this time; S404, reading the front-end node , the back-end node , the front-end pre-node and the back-end post-node of all failed links in the path respectively to obtain an adjustment node set ; S405, based on the adjustment node set , obtaining a local path flag matrix based on the service adaptability weight multi-sampling method; S406, initializing the local path serial number ; S407, setting the sub-source node and the sub-destination node to obtain a local path ; S408, replacing the sub-path between the same nodes in the path with each local path to obtain an adjusted path ; S409, if the adjusted path has two or more same nodes , entering S410, otherwise entering S411; S410, deleting the path between the frontmost node and the rearmost node, and retaining only one node , entering S409; S411, if , then , entering S407, otherwise entering S412; S412, completing the calculation of all adjusted paths, and selecting the adjusted path with the optimal service adaptability to replace the path .

[0023] Further, the step S405 comprises the following steps: S4051, calculating the average service adaptability of the local path:

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030] wherein the above local path is not the path ; S4052, selecting the first local path with the maximum service adaptability , wherein the superscript ; S4053, obtaining the adjusted local path flag matrix according to the following formula : .

[0031] To achieve the above object, according to the second aspect of the present application, a computer device is further provided, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to realize the steps of the network resource re-allocation method according to the first aspect.

[0032] In the present application, a network resource re-allocation method based on a bisection firefly optimization algorithm is proposed. When the resource scheduling strategy is invalid, the invalid node resource re-allocation and invalid link resource re-allocation based on the bisection firefly optimization method are used to solve the problems of link disconnection, link congestion and node failure occurred in the generated path, and then the network resources are re-allocated, the local path is generated, and the new path is formed by integrating the original path. The method described in the present application can correct the resource scheduling strategy according to the changing network state without re-scheduling the resources. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the description of the embodiments of the present application will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0034] Figure 1 ​A flow chart of a network resource re-allocation method based on the bisection moth optimization algorithm according to the present application; Figure 2 A flow chart of a failed node resource re-allocation work process according to the present application; Figure 3 A flow chart of a local path flag matrix of a failed node obtained based on a service fitness weight multi-sampling method according to the present application; Figure 4 An example schematic diagram according to the present application. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0036] As shown in Figure 1 , a bisection moth optimization method for network resource re-allocation includes the following steps: S1, initializing system parameters, including a current path , a failed node number threshold , a failed link number threshold , and a sub-path adjustment number ; S2, when entering a path maintenance period, setting a current path serial number and entering the path of the current serial number; S3, traversing all current paths generated based on the bisection moth optimization method, performing path maintenance on the current paths, and updating the current path serial number ; S4, checking the state of each node in the current path , and if there is a failed node, performing failed node resource re-allocation using a network resource scheduling method based on the bisection moth optimization method; S5, checking the state of each link in the current path , and if there is a failed link or a congested link, regarding the failed link or the congested link as a failed link, and performing failed link resource re-allocation using a network resource scheduling method based on the bisection moth optimization method; S6, re-entering S2 until the system ends work.

[0037] As shown in Figure 2 , in S4, the specific work flow of failed node resource re-allocation is as follows: S401, detecting the number of path failed nodes ; S402, if If the condition is met, proceed to S403; otherwise, proceed to S404. S403. If the current service path is determined to be a failed path, the service path is regenerated using a network resource scheduling method based on the binary moth optimization method, and the current resource reallocation ends. S404, Read all failed nodes in the path respectively The previous node in and the next node Obtain the adjusted node set:

[0038] S405, Adjust the node set as described above. The local path marker matrix of the failed node is obtained based on the business fitness weight multi-sampling method. ,in ; S406. Initialize local path sequence number ; S407, Configure Sub-Source Node and sub-target nodes To obtain the local path ,in For matrix Line 1, page q Column elements, and so on; S408, using local paths Replace path Sub-paths between the same nodes are used to obtain the adjusted path. ; S409, if the adjustment path There are two or more identical nodes If the condition is met, proceed to S410; otherwise, proceed to S411. S410. Delete the path between the first and last nodes, keeping only one node. Enter S409; S411, if ,but If the condition is met, proceed to S407; otherwise, proceed to S412. S412. Complete the calculation of all adjustment paths and select the adjustment path with the best business adaptability. Replace path .

[0039] Furthermore, the steps for generating the local path are as follows: Specify the source and destination nodes, and use the remaining TTL (Time To Live) of the packets as the maximum path length. A local path is generated using a network resource scheduling method based on the binary moth optimization algorithm. The remaining TTL (Time To Live) of packets is calculated by the following formula:

[0040] in For path Remove nodes and nodes The length of the remaining path between the nodes and nodes These are the source and destination nodes specified for this local path, respectively. This refers to the corresponding path to which this local path belongs.

[0041] Furthermore, the service adaptability of S412 is calculated according to the following formula:

[0042] in , , and These are the latency coefficient, bandwidth coefficient, packet loss coefficient, and jitter coefficient, respectively. , , and These are the business paths. j The delay function, bandwidth function, packet loss function, and jitter function are defined by the following formulas:

[0043]

[0044]

[0045]

[0046] in , , and All are coefficients greater than 0. , , and Paths The Middle The latency, bandwidth, packet loss rate, and jitter of each node. For path The total number of nodes. The latency function measures the real-time performance of the service path, providing a basis for path selection; the bandwidth function serves as an optimization target during resource allocation, avoiding path overload and ensuring maximum utilization of link resources; the packet loss function measures the reliability of the service path and is a key indicator in resource scheduling; the jitter function measures the smoothness of transmission, ensuring transmission continuity. Through these four functions, this application can more precisely evaluate and select service paths.

[0047] like Figure 3 As shown in S405, the local path flag matrix of the failed node is obtained based on the business fitness weight multi-sampling method. Includes the following steps: S4051. Calculate the average service fitness of local paths:

[0048]

[0049]

[0050]

[0051] in, For path Average business adaptability For path The number of jumps, For path Middle node With nodes The path between, and the nodes With nodes Cannot be used as a path at the same time The source node and the destination node; S4052. Select the option with the highest business adaptability. Local paths ,in Used to label the business adaptability ranking sequence number; S4053. The local path marker matrix is ​​obtained according to the following formula. : .

[0052] Furthermore, S5 includes the following steps: S501, Number of failed paths detected ; S502, if If the condition is met, proceed to S503; otherwise, proceed to S504. S503, Change the current business path jIf the path is determined to be invalid, a network resource scheduling method based on the bisection method is used to re-generate the service path, and the resource re-allocation is ended. S504, read all invalid links in the path , front-end node , back-end node , front-end pre-node and back-end post-node , get the adjustment node set:

[0053] S505, according to the adjustment node set , based on the service adaptability weight multi-sampling method, obtain the local path flag matrix ; S506, initialize the local path serial number ; S507, set the sub-source node and the sub-destination node , get the local path ; S508, replace the sub-path between the same nodes in the path with each local path , get the adjustment path ; S509, if the adjustment path has two or more same nodes , enter S510, otherwise enter S511; S510, delete the path between the frontmost node and the rearmost node, and only keep one node , enter S509; S511, if , then , enter S507, otherwise enter S512; S512, complete the calculation of all adjustment paths, and select the adjustment path with the optimal service adaptability to replace the path .

[0054] Further, the step S505 includes the following steps: S5051, calculate the average service adaptability of the local path:

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061] wherein the above local path is not the path ; S5052, selecting the first local path with the maximum service adaptability , wherein the superscript ; ; S5053, obtaining the adjusted local path flag matrix according to the following formula : .

[0062] The application is described below by examples.

[0063] In combination with the examples shown in the drawings, a network system with 8 nodes is provided, wherein the TTL packet survival time is 7, and the path Figure 4 is needed to be recalculated between the nodes and , i.e. to generate a local path, the child source node is , the child destination node is , and the remaining TTL packet survival time is 5. The above system is taken as an example to illustrate the resource reallocation of the failed node.

[0064] , , the path is needed to be recalculated if the nodes , and are failed, i.e. to generate a local path. The previous node and the next node of the node are read respectively, and the adjusted node set is obtained as follows: The average service adaptability of the local paths ,

[0065] , , and is calculated, and if the following is obtained:

[0066] ​​Then the local path sign matrix is adjusted is:

[0067] For , the local paths between and , and , and are generated respectively, obtaining:

[0068] Then, after integrating the local paths, , since there are repeated nodes, the obtained after deletion.

[0069] Similarly, for , the obtained ; for , the obtained .

[0070] Finally, from the paths and , the path with the optimal service adaptability is selected as the path after resource reallocation .

[0071] Similarly, take the system shown in Figure 4 as an example to illustrate the resource reallocation of the failed link. , For path , if the link between node and , the link between node and fails, resource reallocation of the failed link is performed. Read the front-end node , the back-end node , the front-end pre-node , and the back-end post-node to obtain the adjusted node set:

[0072] Calculate the average service adaptability of the local paths , , , , , , , , , and , if the obtained:

[0073] then adjust the local path flag matrix is:

[0074] for , respectively generate and , and between the local paths, if

[0075] then integrate the local paths to obtain .

[0076] Similarly, for , obtain ; for , obtain .

[0077] Finally, select the path with the optimal service fitness from the paths and as the path after resource reallocation .

[0078] The application further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements each step of the optimization method when executing the computer program, and can achieve the same beneficial effects as the optimization method. To avoid repetition, no further elaboration is given here.

[0079] In the embodiments provided in the present application, it should be understood that the disclosed method and device can be implemented by other means. For example, the above-described device embodiments are only illustrative, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0080] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can be physically included separately, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of hardware plus software functional unit.

[0081] The above are preferred embodiments of the present application, it should be pointed out that, for those skilled in the technical field, without departing from the principles described in the present application, can also make a number of improvements and refinements, these improvements and refinements should also be considered as the scope of protection of the present application.

Claims

1. A method of network resource reallocation, the method comprising: The method comprises the following steps: S1, entering a path maintenance period, setting a current path serial number, and entering the path of the current serial number; S2, generating a scheduled path for the path of the current serial number; S3, checking the state of each node in the scheduled path, and if there is a failed node, performing resource reallocation for the failed node; S4, checking the state of each link in the scheduled path, and if there is a failed link or a congested link, regarding the failed link or the congested link as a failed link and performing resource reallocation for the failed link; S5, re-entering S1 until resource reallocation for a failed node and resource reallocation for a failed link are no longer needed.

2. The network resource re-allocation method of claim 1, wherein, The entering S2 further comprises: system initialization, parameters of the system initialization comprising current path , failure node quantity threshold , failure link quantity threshold , and sub-path adjustment quantity .

3. The network resource re-allocation method of claim 2, wherein, The S3 comprises the following steps: S301, detecting the number of path failure nodes ; S302, if , enter S303, otherwise enter S304; S303, judging the current service path as a failed path, regenerating the service path by using a network resource scheduling method based on a binary moth optimization method, and ending the resource reallocation; S304, Read all failed nodes in the path respectively The previous node in and the next node Obtain the set of adjusted nodes ; S305、According to the adjustment node set , the local path flag matrix of the failed node is obtained based on the service fitness weight multi-sampling method , wherein ; S306、initialize the local path sequence number ; S307, Configure Sub-Source Nodes and sub-target nodes To obtain the local path ,in For matrix Line 1, page q Column elements, and so on; S308, replace the path with each partial path replaced path sub-paths between the same nodes in the path, to obtain an adjusted path ; S309、if the adjustment path There are two and more same nodes Enter S310, otherwise enter S311; S310, delete the path between the frontmost node and the backmost node, only keep one node , enter S309; S311、if then go to S307, else go to S312; S312, complete the calculation of all adjustment paths, select the adjustment path with the optimal service fitness replacement path .

4. The network resource re-allocation method of claim 3, wherein, The step of generating the local path is: Designating sub-source node and sub-destination node, with the rest TTL message survival time as the maximum path length The network resource scheduling method based on the bisection moth optimization algorithm generates a local path, and the rest TTL message survival time is calculated by the following formula: wherein is a path removing the node and the node the length of the remaining path between the node and the node are the child source node and the child destination node respectively designated for the local path, and is the corresponding path to which the local path belongs.

5. The network resource re-allocation method of claim 4, wherein, The service fitness of the S312 is calculated according to the following formula: wherein , , and are a delay coefficient, a wideband coefficient, a packet loss coefficient and a jitter coefficient, respectively, , , and are a delay function, a bandwidth function, a packet loss function and a jitter function of the service path j , respectively.

6. The network resource re-allocation method of claim 5, wherein, The delay function of the S312 , the bandwidth function , the packet loss function and the jitter function are defined by the following equations, respectively: wherein , , and are coefficients greater than 0, , , and are respectively the latency, bandwidth, packet loss rate and jitter of the th node of the path , is the total number of nodes of the path .

7. The network resource re-allocation method of claim 6, wherein, The S305 comprises the following steps: S3051, calculating the average service fitness of the local path; in, For path Average business adaptability For path The number of jumps, For path Middle node With nodes The path between, and the nodes With nodes Cannot be used as a path at the same time The source node and the destination node; S3052、select the first local path with the largest service fitness wherein for marking the service fitness ranking number;​ S3053, obtain an adjusted local path sign matrix according to the following formula : 。 8. The network resource re-allocation method of claim 7, wherein, The S4 comprises the following steps: S401、Detect the number of failure paths ; S402, if go to S403, otherwise go to S404; S403、determine whether the current service path is a failure path j If the current service path is determined as a failure path, a network resource scheduling method based on a binary moth optimization method is used to regenerate the service path, and the resource redistribution is ended. S404, reading all invalid links in the path respectively middle front end node , back end node , front end front node and back end back node , obtaining an adjusted node set ; S405、According to the adjustment node set , based on the service fitness weight multi-sampling method to obtain the local path flag matrix ; S406, initialize the local path sequence number ; S407, set sub-source node and sub-destination node , get local path ; S408, use each local path replacement path sub-paths between the same nodes in the original path, to obtain an adjusted path ; S409、if the adjustment path there are two and more same nodes enter S410, otherwise enter S411; S410, delete the path between the frontmost node and the backmost node, only keep one node go to S409; S411、if then go to S407, else go to S412; S412, complete the calculation of all adjustment paths, select the adjustment path with the optimal service fitness replacement path .

9. The network resource re-allocation method of claim 8, wherein, The step S405 comprises the following steps: S4051, calculating the average service fitness of the local path; wherein the above local path is not the path ; S4052, select the first local path with the largest service fitness wherein the superscript ;​ S4053, obtain an adjusted local path sign matrix according to the following formula : 。 10. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the network resource reallocation method according to any one of claims 1-9 are realized.

Images