SDN (Software Defined Network) controller deployment method and device under hybrid band control

Abstract

The invention provides an SDN (Software Defined Network) controller deployment method and device under hybrid band control, and the method comprises the steps: taking the network delay, load difference degree and bootstrap time of an SDN system as optimization targets, taking the cost as a constraint condition, and determining a target function; based on a third-generation non-dominated sorting genetic algorithm, a feasible solution construction algorithm, a crossover and mutation algorithm, an isolated node removal algorithm and a random intervention and reference point updating algorithm, solving the target function, and determining an optimal solution of the Pareto frontier; and taking the optimal solution of the Pareto frontier as a controller deployment scheme of the SDN system, and deploying a controller of the SDN system. Therefore, the SDN controller deployment problem under the multi-target condition can be effectively optimized, and the requirements of low delay, load balancing and the like are met.

Description

Technical Field

[0001] The present invention relates to the field of computer network technology, and in particular to a method and device for deploying an SDN controller under hybrid frequency band control. Background Art

[0002] As the scale of modern networks continues to expand, SDN (Software-Defined Networking) has emerged as a new type of network architecture that separates network control and data forwarding and uses software to implement network control. Latency and control model are two key elements of flow transmission in SDN. First, SDN technology needs to support high-performance data forwarding, which means that network devices must have high processing power and low latency. Second, SDN technology needs to support large-scale network environments, which means that network controllers must have strong scalability and high availability.

[0003] There are two control modes for traffic transmission between the controller and the switch, namely in-band control and out-of-band control. The in-band control mode uses the service network to transmit control messages; the out-of-band control mode uses a separate control network to transmit control messages and further forwards them to the service network through some direct links between the two networks. However, due to the limitations of each of these two control modes, the hybrid band control model based on these two models has a better application prospect.

[0004] In the hybrid band control mode, some switches need to be selected to be directly connected to the control network. In the out-of-band control mode, these switches send their control signals to the control network through the control service connection. Other switches are not directly connected to the control network, and their control signals are transmitted to the switches with control signal connections through the data plane and further sent to the control network. In this mode, the deployment of the controller needs to be planned. However, current research only defines a single target for optimization, resulting in the SDN system not being able to meet actual application requirements. Summary of the invention

[0005] The present invention provides a method and device for deploying an SDN controller under hybrid frequency band control, so as to solve the defect that the SDN system under the hybrid frequency band control mode in the prior art is difficult to meet the actual application requirements.

[0006] The present invention provides a method for deploying an SDN controller under hybrid frequency band control, comprising the following steps: The network delay, load difference and bootstrap time of the SDN system are taken as optimization targets, and the cost is taken as a constraint to determine the objective function. Based on the third generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier; The optimal solution of the Pareto front is used as the controller deployment solution of the SDN system, and the controller of the SDN system is deployed.

[0007] Optionally, based on a third-generation non-dominated sorting genetic algorithm, a feasible solution construction algorithm, a crossover mutation algorithm, an isolated node removal algorithm, and a random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier, specifically including: Generate a first population through a feasible solution construction algorithm; each individual in the first population is represented by a set of decision variables; Based on the crossover mutation algorithm and the first population, a second population is obtained; Using a third generation non-dominated sorting genetic algorithm, non-dominated sorting is performed on the second population, and a reference point is selected based on the result of the non-dominated sorting, and an objective function is standardized to create a reference set, and a third population is obtained based on the reference set; By removing isolated nodes and randomly intervening and updating reference points, the isolated nodes of the third population are removed and random influences are added, and the obtained population is used as the updated first population; When the termination condition is met, the optimal solution of the Pareto front is obtained based on the updated first population; when the termination condition is not met, the crossover mutation algorithm and the first population are repeated to obtain the second population, the second population is non-dominated sorted by the third-generation non-dominated sorting genetic algorithm, and a reference point is selected based on the result of the non-dominated sorting, and the objective function is standardized to create a reference set, and a third population is obtained based on the reference set, and the isolated nodes of the third population are removed and random influences are added through the isolated node removal algorithm and the random intervention and reference point update algorithm, and the obtained population is used as the operation of the updated first population until the termination condition is met, and the optimal solution of the Pareto front is obtained based on the updated first population.

[0008] Optionally, the network delay of the SDN system is calculated according to the following formula: In the formula, Indicates the network delay of the SDN system, It represents the transmission delay of data from the source node to the destination node in the SDN system. It represents the computational delay required by the controller of the SDN system to process data after receiving it; The load difference of the SDN system is calculated according to the following formula: In the formula, Representation Controller The load, Indicates the load value of the controller with the highest load, Indicates the controller load value with the lowest load; The bootstrap time of the SDN system is calculated according to the following formula: In the formula, represents the bootstrap time, For minimum delay, is the linear correlation coefficient, Controller in SDN system To exchange The number of node hops passed by the shortest path, Represents the switch set of the SDN system, Represents the controller set of the SDN system.

[0009] Optionally, a first population is generated by a feasible solution construction algorithm, specifically including: The topology graph of the SDN system is divided based on connectivity to obtain a set of connected subgraphs; For any connected subgraph in the connected subgraph set, the number of controllers of the any connected subgraph is calculated according to a preset load and cost; Based on the number of controllers of the arbitrary connected subgraph, a clustering algorithm is used to determine the deployment position of the controller of the arbitrary connected subgraph; A first population is generated based on the controller deployment position of each connected subgraph in the connected subgraph.

[0010] Optionally, obtaining a second population based on a crossover mutation algorithm and the first population specifically includes: Select two parent individuals from the first population, and generate two offspring individuals based on the crossover probability; Based on a preset mutation reduction factor, performing polynomial mutation on the two offspring individuals to update the gene values ​​of the two offspring individuals; Based on a preset boundary value range, a boundary check is performed on the gene values ​​of the updated offspring individuals, and the offspring individuals whose gene values ​​are within the preset boundary value range are taken as individuals of the second population.

[0011] Optionally, removing the isolated nodes of the third population by using an isolated node removal algorithm specifically includes: Assign color information to each controller, check the connectivity between each controller and other controllers, and identify controllers that are not connected with other controllers as isolated nodes; Traversing all neighboring nodes of the isolated node, assigning the same color as any neighboring node to the isolated node, and updating individuals in the third population; By means of random intervention and updating reference point algorithm, random influence is added to the third population intervention, specifically including: adding randomly generated individuals to the third population; After merging the randomly generated individuals with the third population, the dominated solutions are removed, the non-dominated solutions are retained, and the reference set is updated based on the Das-Dennis method.

[0012] The present invention also provides an SDN controller deployment device under hybrid frequency band control, comprising the following modules: A determination module is used to determine the objective function by taking the network delay, load difference and bootstrap time of the SDN system as optimization targets and the cost as a constraint condition; A solution module is used to solve the objective function based on a third-generation non-dominated sorting genetic algorithm, a feasible solution construction algorithm, a crossover mutation algorithm, an isolated node removal algorithm, and a random intervention and reference point update algorithm to determine the optimal solution of the Pareto frontier; The deployment module is used to use the optimal solution of the Pareto front as the controller deployment solution of the SDN system to deploy the controller of the SDN system.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, an SDN controller deployment method under hybrid frequency band control as described in any one of the above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the SDN controller deployment method under hybrid frequency band control as described in any one of the above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for deploying an SDN controller under hybrid frequency band control as described in any one of the above is implemented.

[0016] The SDN controller deployment method and device under hybrid frequency band control provided by the present invention takes the network delay, load difference and bootstrap time of the SDN system as optimization targets and the cost as a constraint condition to determine the objective function, and then solves the objective function based on the third-generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm to determine the optimal solution of the Pareto frontier, thereby effectively optimizing the SDN controller deployment problem under multi-objective conditions and meeting the requirements of low delay, load balancing, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of a hybrid frequency band control scenario provided for related art.

[0019] Figure 2 A schematic diagram of a flow chart of a method for deploying an SDN controller under hybrid frequency band control provided by the present invention.

[0020] Figure 3 This is a flow chart of the algorithm RINSGA-III provided by the present invention.

[0021] Figure 4 This is a schematic diagram of the Chinanet topology provided by the present invention.

[0022] Figure 5 This is a schematic diagram of the Tw topology delay comparison provided by the present invention.

[0023] Figure 6 This is a schematic diagram of the Tw topology bootstrap time comparison provided by the present invention.

[0024] Figure 7 A schematic diagram for comparing Tw topology load balancing provided by the present invention.

[0025] Figure 8 This is a schematic diagram of Chinanet topology delay comparison provided by the present invention.

[0026] Fig. 9 This is a schematic diagram of the Chinanet topology bootstrap time comparison provided by the present invention.

[0027] Fig.10 A schematic diagram for comparing Chinanet topology load balancing provided by the present invention.

[0028] Fig.11 This is a schematic diagram of the Cogentco topology delay comparison provided by the present invention.

[0029] Fig.12 This is a schematic diagram of the bootstrap time comparison of the Cogentco topology provided by the present invention.

[0030] Fig.13 A schematic diagram of the Cogentco topology load balancing comparison provided by the present invention.

[0031] Fig.14 A structural schematic diagram of an SDN controller deployment device under hybrid frequency band control provided by the present invention.

[0032] Fig.15 This is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Figure 1 A schematic diagram of a hybrid frequency band control scenario provided for related technologies, such as Figure 1 As shown, in the mixed-band control mode, a selected portion of switches are directly connected to the control network. In the out-of-band control mode, these switches send their control signals to the control network through the control service connection (dashed line). Other switches are not directly connected to the control network, and their control signals are transmitted to the switches with control signal connections through the data plane and further sent to the control network.

[0035] In the embodiment of the present invention, the optimization goal is to minimize the total network delay, load difference and boot time. The mode also takes the cost budget into consideration.

[0036] The main ideas and basic methods of the embodiments of the present invention are as follows: placing a controller at an appropriate location, determining the management of the switch by the controller, selecting a switch in an out-of-band control mode, and selecting a node (switch or controller) to be connected to the above switch in a mixed-band control mode.

[0037] Figure 2 is a flow chart of the SDN controller deployment method under hybrid frequency band control provided by the present invention, such as Figure 2 As shown, the method comprises the following steps: Step 200: Taking the network delay, load difference and bootstrap time of the SDN system as optimization targets and cost as a constraint, determine the objective function.

[0038] Step 201: Based on the third generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto front.

[0039] Step 202: Use the optimal solution of the Pareto front as a controller deployment solution for the SDN system to deploy the controller of the SDN system.

[0040] Specifically, in the embodiment of the present invention, the network delay, load difference and bootstrap time of the SDN system need to be used as optimization targets, and the cost needs to be used as a constraint condition.

[0041] In some implementations, for the total network delay, a more perfect approach is to comprehensively consider the transmission delay and processing delay. The transmission delay is determined by the path. For the processing delay, due to the burstiness of message transmission, the arrival of data packets in the network does not obey the Poisson distribution, but the arrival process of the session flow is a Poisson flow, so the arrival method of the first message of all flows is also a Poisson flow, and the Packet-in message of each switch is the first data packet of all flows, so the arrival process of the Packet-in message in each switch follows the Poisson distribution. The Packet-in message flow at the input end of the controller is formed by the convergence of the Packet-in message flows sent by all routers, which can be regarded as the synthesis of multiple Poisson flows. Assuming that the Packet-in message flows sent by each router are independent of each other, according to the Poisson flow composition theorem, the Packet-in message flow of the controller is still a Poisson flow. Then the queuing theory model M / M / 1 can be applied to consider the processing time delay.

[0042] Optionally, the network delay of the SDN system is calculated according to the following formula: In the formula, Indicates the network delay of the SDN system, It represents the transmission delay of data from the source node to the destination node in the SDN system. It indicates the computational delay required by the controller of the SDN system to process data after receiving it.

[0043] In some implementations, for load diversity, the number of switches that each controller is responsible for is generally used as an indicator to measure node diversity. The controller with the most load nodes minus the controller with the least load nodes is obtained as shown in the following formula: in, For the controller in the network The number of switching nodes controlled.

[0044] In the embodiment of the present invention, for comprehensive consideration, the load can be broken down into the load of control traffic brought to the controller by different switches of the system. In the actual application of SDN, the load of the controller is mainly composed of processing Packet-In data packets sent from the switch, installing flow table entries, maintaining the network view, and communication between controllers. Since the forwarding path needs to be calculated when processing Packet-In data packets, this is a very time-consuming and resource-consuming process, and its load share is far greater than the other three parts. Therefore, the embodiment of the present invention uses the number of Packet-In data packets received by the controller from the switch per unit time as the load of the controller. Controller Load As shown in the following formula: in, For Switch The number of Packet-In packets sent per unit time. Indicates a switch and controller The affiliation between them.

[0045] Optionally, the load difference of the SDN system is calculated according to the following formula: In the formula, Representation Controller The load, Indicates the load value of the controller with the highest load, Indicates the controller load value with the lowest load.

[0046] In some implementations, the bootstrap time was not proposed as a metric in previous studies, but the comparison between in-band and out-of-band in the current study found that the bootstrap time is an important indicator for selecting in-band and out-of-band architectures.

[0047] When OpenFlow is controlled in in-band mode, the control traffic and data traffic of the deployed controller are transmitted on the same channel (in-band mode). Implementing this in-band mode is complicated because the switch must search and establish a path to the controller through other switches in the network (i.e. bootstrapping). Considering that the bootstrap time is closely related to the topology type, given the linear topology and mesh topology, the bootstrap time is linearly related to the number of node hops passed by the shortest path from the controller to the switch. When the network topology is a star topology, the bootstrap time can be ignored within the controller processing capacity limit. When the controller processing capacity limit is exceeded, the bootstrap time increases exponentially, which is related to the processing accumulation of packet-in. In general, it can be considered that the controller bootstrap time is linearly related to the number of hops in the shortest path between the controller and the switch.

[0048] Optionally, the bootstrap time of the SDN system is calculated according to the following formula: In the formula, represents the bootstrap time, For minimum delay, is the linear correlation coefficient, Controller in SDN system To exchange The number of node hops passed by the shortest path, Represents the switch set of the SDN system, Represents the controller set of the SDN system.

[0049] Therefore, the network delay, load diversity and bootstrap time of the SDN system can be used as optimization targets, and the cost can be used as a constraint to determine the objective function.

[0050] In order to solve the objective function, the embodiment of the present invention selects the third generation non-dominated sorting genetic algorithm (Reference-point Based Non-dominated Sorting Genetic Algorithm, NSGA-III) as the improved basic algorithm.

[0051] First, let's briefly introduce the algorithm flow of NSGA-Ⅲ. The traditional NSGA-Ⅲ algorithm includes the following steps: 1. Initialize the population: Randomly generate an initial population containing multiple individuals (solutions), each of which is represented by a set of decision variables.

[0052] 2. Non-dominated sorting: Perform non-dominated sorting on the individuals in the initial population and divide them into different levels (layers), where the first level contains non-dominated solutions on the Pareto front, the second level contains solutions dominated by the first level, and so on.

[0053] 3. Crossover and mutation: Use the crossover operation to generate one or more child solutions from two parent solutions. Use the mutation operation to make small random changes to the individual decision variables to increase the diversity of the population.

[0054] 4. Merge populations: Merge the newly generated offspring population with the parent population to form a new population (size is 2N).

[0055] 5. Environment selection: If the size of the merged population exceeds the required size (N), a selection strategy is used to select N individuals from the merged population as the next generation population. NSGA-III adopts a reference point-based selection mechanism to select individuals by calculating the distance between the individual and the reference point to ensure a uniform distribution of solutions on the Pareto frontier.

[0056] 6. Check termination conditions: Check whether the termination conditions are met (such as reaching the maximum number of iterations or meeting other convergence criteria). If the termination conditions are not met, return to step 2 and continue the next generation of evolution.

[0057] 7. Output results: If the termination condition is met, the set of high-quality solutions on the Pareto frontier found is output as the output of the algorithm.

[0058] The pseudo code of the NSGA-Ⅲ algorithm is as follows:

[0059] Based on the traditional NSGA-Ⅲ, at different stages of the algorithm solution, improvements are made through feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, so that the algorithm can have the ability to solve three optimization objectives simultaneously.

[0060] After determining the optimal solution of the Pareto front according to the improved NSGA-Ⅲ algorithm, the optimal solution of the Pareto front can be used as the controller deployment plan of the SDN system to deploy the controller of the SDN system.

[0061] The SDN controller deployment method under hybrid frequency band control provided by the present invention takes the network delay, load difference and bootstrap time of the SDN system as optimization targets and the cost as a constraint condition to determine the objective function, and then solves the objective function based on the third-generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm to determine the optimal solution of the Pareto frontier, thereby effectively optimizing the SDN controller deployment problem under multi-objective conditions and meeting the requirements of low delay, load balancing and the like.

[0062] Optionally, based on a third generation non-dominated sorting genetic algorithm, a feasible solution construction algorithm, a crossover mutation algorithm, an isolated node removal algorithm, and a random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier, specifically including: The first population is generated through a feasible solution construction algorithm; each individual in the first population is represented by a set of decision variables; Based on the crossover mutation algorithm and the first population, the second population is obtained; The second population is non-dominatedly sorted by the third generation non-dominated sorting genetic algorithm, and a reference point is selected based on the result of the non-dominated sorting, and the objective function is standardized to create a reference set, and the third population is obtained based on the reference set; Through the algorithm of removing isolated nodes and the algorithm of random intervention and updating reference points, the isolated nodes of the third population are removed and random influences are added, and the obtained population is used as the first population after update; When the termination condition is met, the optimal solution of the Pareto front is obtained based on the updated first population; when the termination condition is not met, the crossover mutation algorithm and the first population are repeated to obtain the second population, and the second population is non-dominated sorted by the third-generation non-dominated sorting genetic algorithm, and the reference point is selected based on the result of the non-dominated sorting, and the objective function is standardized to create a reference set, and the third population is obtained based on the reference set, and the isolated nodes of the third population are removed and random influences are intervened and added through the isolated node removal algorithm and the random intervention and reference point update algorithm, and the obtained population is used as the operation of the updated first population until the termination condition is met, and the optimal solution of the Pareto front is obtained based on the updated first population.

[0063] Specifically, as described above, the embodiment of the present invention improves the algorithm NSGA-III. Next, the basic principle of the improved algorithm will be introduced. The improved multi-objective evolutionary algorithm needs to have the ability to solve three optimization objectives at the same time. For this reason, the algorithm adds corresponding improvement mechanisms at different stages of the algorithm solution.

[0064] First, the first population (i.e., the initialization population, or the initialization feasible solution) is generated through a feasible solution construction algorithm, and each individual in the first population is represented by a set of decision variables.

[0065] Optionally, a first population is generated by a feasible solution construction algorithm, specifically including: The topology graph of the SDN system is divided based on connectivity to obtain a set of connected subgraphs; For any connected subgraph in the connected subgraph set, the number of controllers of any connected subgraph is calculated according to the preset load and cost; Based on the number of controllers in any connected subgraph, a clustering algorithm is used to determine the deployment location of the controllers in any connected subgraph; The first population is generated based on the controller deployment positions of each connected subgraph in the connected subgraph.

[0066] Specifically, in the process of initializing feasible solutions, the coverage and accuracy of the initialized feasible solutions are comprehensively considered. In previous fireworks algorithms, random feasible solutions were mostly used. Considering the particularity of topological scenarios, the importance of the locations where controllers are placed is obviously different. The weight of placing controllers in more central locations should be significantly higher than other locations. In addition, considering the connectivity of the graph, this patent proposes a solution initialization scheme. After calculating the number of controllers based on the load and actual conditions, a clustering algorithm is used to initialize the position of the topological graph, and this position is used as the position of the controller in the initial feasible solution.

[0067] The pseudo code of the feasible solution construction algorithm is as follows:

[0068] After the first population is generated, the second population can be obtained based on the crossover mutation algorithm and the first population.

[0069] Optionally, based on the crossover mutation algorithm and the first population, a second population is obtained, specifically including: Select two parent individuals from the first population and generate two offspring individuals based on the crossover probability; Based on the preset mutation reduction factor, perform polynomial mutation on the two offspring individuals and update the gene values ​​of the two offspring individuals; Based on the preset boundary value range, the gene values ​​of the updated offspring individuals are subjected to boundary checking, and the offspring individuals whose gene values ​​are within the preset boundary value range are taken as individuals of the second population.

[0070] Specifically, for the crossover mutation problem of operators, the original solution space contains a large number of 01 variables. These variables are directly negated in the process of generating offspring to improve the calculation speed. In addition, the segments of feasible solutions are divided, and different mutation strategies are used in different segments. The genetic operator is essentially a mutation operation on the graph abstracted into a one-dimensional solution set. Considering that the solution can be divided into three parts, the first part uses the original polynomial mutation as the position, and the second part uses the polynomial mutation method with moderate value reduction for distance. The third part uses the polynomial mutation method with reduced weight for the number of connections.

[0071] For binary crossover polynomial mutation, the algorithm pseudo code is as follows:

[0072] After the second population is obtained through the crossover mutation algorithm, the third generation non-dominated sorting genetic algorithm is used to perform non-dominated sorting on the second population, and a reference point is selected based on the result of the non-dominated sorting, and the objective function is standardized to create a reference set, and the third population is obtained based on the reference set. The specific process can be found in the NSGA-Ⅲ algorithm flow, which will not be repeated here.

[0073] After obtaining the reference set and the third population, the isolated nodes of the third population can be removed and random influences can be added through the isolated node removal algorithm and the random intervention and reference point update algorithm.

[0074] Optionally, the isolated nodes of the third population are removed by an isolated node removal algorithm, specifically including: Assign color information to each controller, check the connectivity between each controller and other controllers, and identify controllers that are not connected with other controllers as isolated nodes; Traverse all neighbor nodes of the isolated node, assign the same color to the isolated node as any neighbor node, and update the individuals in the third population.

[0075] Specifically, in the SDN scenario, the placement of the controller must take into account connectivity and isolated node issues. That is, when the child solution is generated, there will be a problem that a switch is in another control domain, which will cause a cross-domain problem, which is very unfavorable for the delay of the switch connection and the delivery of the flow table. Therefore, an embodiment of the present invention proposes an algorithm for removing isolated nodes.

[0076] In this algorithm for removing isolated nodes, it is necessary to assign color information to each controller and check the connectivity between each controller and other controllers, and determine the controllers that are not connected with other controllers as isolated nodes.

[0077] After obtaining the encouraged node, traverse all neighbor nodes of the isolated node, assign the same color to the isolated node as any neighbor node, and update the individuals in the third population.

[0078] The isolated node removal algorithm is applied in each iteration process to perform node correction on the mutated offspring solutions, removing isolated solutions and merging them into non-isolated solutions.

[0079] The pseudo code of the algorithm for removing isolated nodes is as follows:

[0080] Optionally, random influence is added to the third population intervention through random intervention and reference point update algorithm, specifically including: Add randomly generated individuals to the third population; After merging the randomly generated individuals with the third population, the dominated solutions are removed, the non-dominated solutions are retained, and the reference set is updated based on the Das-Dennis method.

[0081] Specifically, for each generation of offspring, the traditional algorithm selects individuals based on the reference point selection mechanism by calculating the distance between the individual and the reference point to ensure the uniform distribution of solutions on the Pareto front. In the SDN scenario, the solution set has a high dimension, and the mutation effect is difficult to guarantee compared with other scenarios, that is, it is difficult to obtain points on or close to the Pareto front, so it is not enough to only use a reference point selection mechanism to achieve a uniform distribution of Pareto. Therefore, the embodiment of the present invention introduces a part of the solutions whose number is negatively correlated with the number of iterations as random solutions into the next generation selection process to maintain a uniform distribution of solutions in scenarios with low mutation success rates. In addition, the reference point set is dynamically updated in real time to reflect the shape of the current Pareto front.

[0082] The number of random solutions is given by Decide, is a coefficient that decreases as the number of iterations increases, and the formula is as follows: The algorithm pseudo code is as follows:

[0083] By removing the isolated node algorithm and the random intervention and updating reference point algorithm, after removing the isolated nodes of the third population and intervening and adding random influences, the obtained population can be used as the updated first population, and the reference set is also updated at this time.

[0084] When the termination condition is met, the optimal solution of the Pareto front can be obtained based on the updated first population.

[0085] If the termination condition is not met, the crossover mutation algorithm and the first population are repeated to obtain the second population after the update. The second population is non-dominated sorted by the third-generation non-dominated sorting genetic algorithm, and the reference point is selected based on the result of the non-dominated sorting. The objective function is standardized and a reference set is created. The third population is obtained based on the reference set. The isolated nodes of the third population are removed and random influences are added through the isolated node removal algorithm and the random intervention and reference point update algorithm. The obtained population is used as the operation of the updated first population, and it is continuously iterated until the termination condition is met. When the termination condition is met, the optimal solution of the Pareto frontier is obtained based on the updated first population.

[0086] The termination condition may be that the number of iterations is reached, or the optimal solution reaches a preset range, etc. The specific setting may be determined according to the actual situation.

[0087] The SDN controller deployment method under hybrid frequency band control provided by the present invention is further explained below through embodiments in specific application scenarios.

[0088] This embodiment aims at the current research status that there is only a single-objective optimization for the SDN multi-controller placement problem in the mixed frequency band control scenario, and proposes a three-objective optimization algorithm for this scenario, which is a comprehensive optimization algorithm with comprehensive delay, bootstrap cost, and load balancing as the three objectives. Among them, the comprehensive delay comprehensively considers the transmission delay and propagation delay, representing the delay status of the entire system.

[0089] In addition, this embodiment selects NSGA-Ⅲ as the improved basic algorithm, proposes the algorithm RINSGA-Ⅲ, introduces the clustering strategy when initializing the feasible solution, proposes the algorithm for removing isolated nodes in the SDN scenario in the iteration process, and introduces the segmented mutation strategy in view of the fact that the original crossover mutation problem is not suitable for complex high-dimensional variables. In view of the problem that the offspring solution cannot be guaranteed to be fully distributed on the Pareto frontier during the algorithm process, it is proposed to add random solutions as intervention to improve the randomness of the algorithm and prevent the wrong evolution direction or premature convergence to the local situation. Figure 3 The flowchart of the algorithm RINSGA-III provided by the present invention is as follows: Figure 3 As shown, the pseudo code is as follows:

[0090] This embodiment uses Python language to carry out simulation experiments, and the comparison algorithms include: S-MOFWA algorithm based on S metric; original NSGA-Ⅲ algorithm; MaOSCA algorithm based on information feedback; improved FWA algorithm proposed for delay.

[0091] In order to evaluate the performance of the algorithm proposed in this embodiment, the conditions of the simulation experiment are as follows: Network topology: Tw topology: The real network topology structure KDL from the dataset “The Internet Topology Zoo”, with a total of 76 nodes and 116 links.

[0092] Cogentco topology: Cogent, a real network topology from the dataset “The Internet Topology Zoo”, with a total of 197 nodes and 245 links.

[0093] Chinanet topology: Chinanet, a real network topology from the dataset “The Internet Topology Zoo”, has 42 nodes and 66 links. Figure 4This is a schematic diagram of the Chinanet topology provided by the present invention.

[0094] Evaluation indicators: Three optimization objectives: comprehensive delay, load balancing, and bootstrap cost.

[0095] For comparison of the above four algorithms, Python simulation is used to run the simulation program ten times for each algorithm under each topology, and the average value is calculated as the simulation result.

[0096] Figure 5 A schematic diagram of Tw topology delay comparison provided by the present invention; Figure 6 A schematic diagram of the Tw topology bootstrap time comparison provided by the present invention; Figure 7 A schematic diagram of Tw topology load balancing comparison provided by the present invention; Figure 8 A schematic diagram of Chinanet topology delay comparison provided by the present invention; Fig. 9 A schematic diagram showing the comparison of Chinanet topology bootstrapping time provided by the present invention; Fig.10 A schematic diagram of Chinanet topology load balancing comparison provided by the present invention; Fig.11 A schematic diagram of the Cogentco topology delay comparison provided by the present invention; Fig.12 A schematic diagram of the bootstrap time comparison of the Cogentco topology provided by the present invention; Fig.13 A schematic diagram of the Cogentco topology load balancing comparison provided by the present invention.

[0097] like Figure 5-13 As shown in the figure, simulation experiments show that the improved algorithm has a significant improvement effect on all three indicators. Taking the smallest Chinnet topology as an example, Figure 8-10 As shown in the figure, under the indicator of latency, the improved algorithm has an improvement of about 8.61% for the S-MOFWA algorithm, about 9.98% for the NSGA-Ⅲ algorithm, and about 56% for the MaOSCA algorithm. Under the indicator of load balancing, the improved algorithm has an improvement of about 14.62% for the S-MOFWA algorithm, about 43.01% for the NSGA-Ⅲ algorithm, and about 45.08% for the MaOSCA algorithm. Under the indicator of bootstrap time, the improved algorithm has an improvement of about 19.31% for the S-MOFWA algorithm, about 6.92% for the NSGA-Ⅲ algorithm, and about 5.58% for the MaOSCA algorithm.

[0098] The following is a description of an SDN controller deployment device under hybrid frequency band control provided by the present invention. The SDN controller deployment device under hybrid frequency band control described below and the SDN controller deployment method under hybrid frequency band control described above can correspond to each other.

[0099] Fig.14 The schematic diagram of the structure of the SDN controller deployment device under the hybrid frequency band control provided by the present invention is as follows: Fig.14 As shown, the device includes the following modules: A determination module 1400 is used to determine an objective function by taking the network delay, load difference and bootstrap time of the SDN system as optimization targets and the cost as a constraint condition; A solution module 1410 is used to solve the objective function based on a third generation non-dominated sorting genetic algorithm, a feasible solution construction algorithm, a crossover mutation algorithm, an isolated node removal algorithm, and a random intervention and reference point update algorithm to determine the optimal solution of the Pareto frontier; The deployment module 1420 is used to deploy the controller of the SDN system by taking the optimal solution of the Pareto front as the controller deployment solution of the SDN system.

[0100] Optionally, the solution module 1410 is specifically configured to: The first population is generated through a feasible solution construction algorithm; each individual in the first population is represented by a set of decision variables; Based on the crossover mutation algorithm and the first population, the second population is obtained; The second population is non-dominatedly sorted by the third generation non-dominated sorting genetic algorithm, and a reference point is selected based on the result of the non-dominated sorting, and the objective function is standardized to create a reference set, and the third population is obtained based on the reference set; Through the algorithm of removing isolated nodes and the algorithm of random intervention and updating reference points, the isolated nodes of the third population are removed and random influences are added, and the obtained population is used as the first population after update; When the termination condition is met, the optimal solution of the Pareto front is obtained based on the updated first population; when the termination condition is not met, the crossover mutation algorithm and the first population are repeated to obtain the second population, and the second population is non-dominated sorted by the third-generation non-dominated sorting genetic algorithm, and the reference point is selected based on the result of the non-dominated sorting, and the objective function is standardized to create a reference set, and the third population is obtained based on the reference set, and the isolated nodes of the third population are removed and random influences are intervened and added through the isolated node removal algorithm and the random intervention and reference point update algorithm, and the obtained population is used as the operation of the updated first population until the termination condition is met, and the optimal solution of the Pareto front is obtained based on the updated first population.

[0101] Optionally, the network delay of the SDN system is calculated according to the following formula: In the formula, Indicates the network delay of the SDN system, It represents the transmission delay of data from the source node to the destination node in the SDN system. It represents the computational delay required by the controller of the SDN system to process data after receiving it; The load difference of the SDN system is calculated according to the following formula: In the formula, Representation Controller The load, Indicates the load value of the controller with the highest load, Indicates the controller load value with the lowest load; The bootstrap time of the SDN system is calculated according to the following formula: In the formula, represents the bootstrap time, For minimum delay, is the linear correlation coefficient, Controller in SDN system To exchange The number of node hops passed by the shortest path, Represents the switch set of the SDN system, Represents the controller set of the SDN system.

[0102] Optionally, a first population is generated by a feasible solution construction algorithm, specifically including: The topology graph of the SDN system is divided based on connectivity to obtain a set of connected subgraphs; For any connected subgraph in the connected subgraph set, the number of controllers of any connected subgraph is calculated according to the preset load and cost; Based on the number of controllers in any connected subgraph, a clustering algorithm is used to determine the deployment location of the controllers in any connected subgraph; The first population is generated based on the controller deployment positions of each connected subgraph in the connected subgraph.

[0103] Optionally, based on the crossover mutation algorithm and the first population, a second population is obtained, specifically including: Select two parent individuals from the first population and generate two offspring individuals based on the crossover probability; Based on the preset mutation reduction factor, perform polynomial mutation on the two offspring individuals and update the gene values ​​of the two offspring individuals; Based on the preset boundary value range, the gene values ​​of the updated offspring individuals are subjected to boundary checking, and the offspring individuals whose gene values ​​are within the preset boundary value range are taken as individuals of the second population.

[0104] Optionally, the isolated nodes of the third population are removed by an isolated node removal algorithm, specifically including: Assign color information to each controller, check the connectivity between each controller and other controllers, and identify controllers that are not connected with other controllers as isolated nodes; Traverse all neighbor nodes of the isolated node, assign the same color to the isolated node as any neighbor node, and update the individuals in the third population; Through random intervention and updating reference point algorithm, random influence is added to the third population intervention, including: Add randomly generated individuals to the third population; After merging the randomly generated individuals with the third population, the dominated solutions are removed, the non-dominated solutions are retained, and the reference set is updated based on the Das-Dennis device.

[0105] Fig.15 A schematic diagram of the structure of an electronic device provided by the present invention, such as Fig.15 As shown, the electronic device may include: a processor 1510, a communications interface 1520, a memory 1530 and a communication bus 1540, wherein the processor 1510, the communications interface 1520, and the memory 1530 communicate with each other through the communication bus 1540. The processor 1510 may call the logic instructions in the memory 1530 to execute the SDN controller deployment method under hybrid frequency band control, and the method includes: The network delay, load difference and bootstrap time of the SDN system are taken as optimization targets, and the cost is taken as a constraint to determine the objective function. Based on the third generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier; The optimal solution of the Pareto frontier is used as the controller deployment solution of the SDN system, and the controller of the SDN system is deployed.

[0106] In addition, the logic instructions in the above-mentioned memory 1530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0107] On the other hand, the present invention further provides a computer program product, the computer program product comprising a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the SDN controller deployment method under hybrid frequency band control provided by the above methods, the method comprising: The network delay, load difference and bootstrap time of the SDN system are taken as optimization targets, and the cost is taken as a constraint to determine the objective function. Based on the third generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier; The optimal solution of the Pareto frontier is used as the controller deployment solution of the SDN system, and the controller of the SDN system is deployed.

[0108] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for deploying an SDN controller under hybrid frequency band control provided by the above methods is implemented, and the method includes: The network delay, load difference and bootstrap time of the SDN system are taken as optimization targets, and the cost is taken as a constraint to determine the objective function. Based on the third generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier; The optimal solution of the Pareto frontier is used as the controller deployment solution of the SDN system, and the controller of the SDN system is deployed.

[0109] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for deploying a software defined network (SDN) controller under hybrid frequency band control, characterized in that: include: The network delay, load difference and bootstrap time of the SDN system are taken as optimization targets, and the cost is taken as a constraint to determine the objective function. Based on the third generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier; The optimal solution of the Pareto front is used as the controller deployment solution of the SDN system, and the controller of the SDN system is deployed.

2. The SDN controller deployment method under hybrid frequency band control according to claim 1, characterized in that: Based on the third generation non-dominated sorting genetic algorithm, feasible solution construction algorithm, crossover mutation algorithm, isolated node removal algorithm, and random intervention and reference point update algorithm, the objective function is solved to determine the optimal solution of the Pareto frontier, specifically including: Generate a first population through a feasible solution construction algorithm; each individual in the first population is represented by a set of decision variables; Based on the crossover mutation algorithm and the first population, a second population is obtained; Using a third generation non-dominated sorting genetic algorithm, non-dominated sorting is performed on the second population, and a reference point is selected based on the result of the non-dominated sorting, and an objective function is standardized to create a reference set, and a third population is obtained based on the reference set; By removing isolated nodes and randomly intervening and updating reference points, the isolated nodes of the third population are removed and random influences are added, and the obtained population is used as the updated first population; When the termination condition is met, the optimal solution of the Pareto front is obtained based on the updated first population; when the termination condition is not met, the crossover mutation algorithm and the first population are repeated to obtain the second population, the second population is non-dominated sorted by the third-generation non-dominated sorting genetic algorithm, and a reference point is selected based on the result of the non-dominated sorting, and the objective function is standardized to create a reference set, and a third population is obtained based on the reference set, and the isolated nodes of the third population are removed and random influences are added through the isolated node removal algorithm and the random intervention and reference point update algorithm, and the obtained population is used as the operation of the updated first population until the termination condition is met, and the optimal solution of the Pareto front is obtained based on the updated first population.

3. The SDN controller deployment method under hybrid frequency band control according to claim 1, characterized in that: The network delay of the SDN system is calculated according to the following formula: In the formula, Indicates the network delay of the SDN system, It represents the transmission delay of data from the source node to the destination node in the SDN system. It represents the computational delay required by the controller of the SDN system to process data after receiving it; The load difference of the SDN system is calculated according to the following formula: In the formula, Representation Controller The load, Indicates the load value of the controller with the highest load, Indicates the controller load value with the lowest load; The bootstrap time of the SDN system is calculated according to the following formula: In the formula, represents the bootstrap time, For minimum delay, is the linear correlation coefficient, Controller in SDN system To exchange The number of node hops passed by the shortest path, Represents the switch set of the SDN system, Represents the controller set of the SDN system.

4. The SDN controller deployment method under hybrid frequency band control according to claim 2, characterized in that: The first population is generated through a feasible solution construction algorithm, which includes: The topology graph of the SDN system is divided based on connectivity to obtain a set of connected subgraphs; For any connected subgraph in the connected subgraph set, the number of controllers of the any connected subgraph is calculated according to a preset load and cost; Based on the number of controllers of the arbitrary connected subgraph, a clustering algorithm is used to determine the deployment position of the controller of the arbitrary connected subgraph; A first population is generated based on the controller deployment position of each connected subgraph in the connected subgraph.

5. The SDN controller deployment method under hybrid frequency band control according to claim 2, characterized in that: Based on the crossover mutation algorithm and the first population, a second population is obtained, which specifically includes: Select two parent individuals from the first population, and generate two offspring individuals based on the crossover probability; Based on a preset mutation reduction factor, performing polynomial mutation on the two offspring individuals to update the gene values ​​of the two offspring individuals; Based on a preset boundary value range, a boundary check is performed on the gene values ​​of the updated offspring individuals, and the offspring individuals whose gene values ​​are within the preset boundary value range are taken as individuals of the second population.

6. The SDN controller deployment method under hybrid frequency band control according to claim 2, characterized in that: Removing isolated nodes of the third population by using an isolated node removal algorithm specifically includes: Assign color information to each controller, check the connectivity between each controller and other controllers, and identify controllers that are not connected with other controllers as isolated nodes; Traversing all neighboring nodes of the isolated node, assigning the same color as any neighboring node to the isolated node, and updating individuals in the third population; By means of random intervention and updating reference point algorithm, random influence is added to the third population intervention, specifically including: adding randomly generated individuals to the third population; After merging the randomly generated individuals with the third population, the dominated solutions are removed, the non-dominated solutions are retained, and the reference set is updated based on the Das-Dennis method.

7. A software defined network (SDN) controller deployment device under hybrid frequency band control, characterized in that: include: A determination module is used to determine the objective function by taking the network delay, load difference and bootstrap time of the SDN system as optimization targets and the cost as a constraint condition; A solution module is used to solve the objective function based on a third-generation non-dominated sorting genetic algorithm, a feasible solution construction algorithm, a crossover mutation algorithm, an isolated node removal algorithm, and a random intervention and reference point update algorithm to determine the optimal solution of the Pareto frontier; The deployment module is used to use the optimal solution of the Pareto front as the controller deployment solution of the SDN system to deploy the controller of the SDN system.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the SDN controller deployment method under hybrid frequency band control is implemented as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the SDN controller deployment method under hybrid frequency band control is implemented as claimed in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the SDN controller deployment method under hybrid frequency band control is implemented as claimed in any one of claims 1 to 6.

Images